- Change translation of some terms:
+ тред -> поток
+ журнал -> лог
+ аптайм -> время работы
- use lowercase for second word where needed
- Reorder words where needed
These things was discussed on irc #ru
- Limit number of parallel requests of a single piece when in the end game
- Shorten and weight the speed tracker so the display is more
reflective of current speed
and '4ef1d99f924ee431587d6f85241540e031fecbc1'
to branch 'i2p.i2p'
Explicit merge from test4.
1.6 JDK now required to build.
1.5 JRE still OK to run I2P.
* configlogging.jsp:
- Add easy way to add an override
- Make file size specifier more flexible
* Console:
- Sort RouterAddress options on netdb.jsp and peers.jsp
- Remove unused web-*.xml file from war
* Crypto:
- Convert all ArrayList caching to LBQs in YKGenerator,
HMACGenerator, and AESKeyCache.
- Change DSAEngine params from Hash to new SHA1Hash, since
these were really 20 byte hashes, not 32 byte Hashes.
- Add stats to track YKGenerator caching success
- Fix YKGenerator precalculation to be much more useful by
increasing the cache size and dramatically shortening the delay
- Option cleanups
- YKGenerator cleanups
- Mark HMAC256Generator unused
* EepGet: Reset length variable on redirect
* Files: Change permissions to 600/700 for all written files/directories.
Now requires Java 1.6 to build, but only 1.5+ to run.
(requires 1.6 to set permissiomns)
* GeoIP: Fix locking bug causing lookups to stop
* Hash: Throw IAE if data length is not 32 bytes,
now that DSAEngine abuse is gone
* HTTPResponseOutputStream:
- More caching
- Stats cleanup
- Max header length check
- Catch OOM
- Initializer cleanup
- Javadoc
* I2CP:
- Add new option i2cp.messageReliability=none, which prevents the
router from sending MessageStatusMessages back in reply to an
outbound SendMessageMessage. Since the streaming lib always ignored
the MSMs anyway, make it the default for streaming.
This will reduce the I2CP traffic significantly.
MSM handling now avoided, but it is still fairly broken, see
comments in I2PSessionImpl2.
- Cleanups to replace method calls with fields
- More cleanups, javadoc, rate reduction
* i2psnark:
- Compact response format
- Add link to finished torrent in message box
- Don't let one bad torrent prevent others from
starting or stopping
- Sort peers by completion %
- Add some missing mime types to web.xml
- shouldLog() cleanup
* i2ptunnel:
- Now that streaming flush() is fixed, use it in IRCClient, and
for initial data in I2PTunnel runner, to avoid the 250 ms
passive flush delay
- Add hostname DSA signature field, to be used for addkey forms.
Experimental, may be commented out later.
- More header blocking (thanks telecomix!)
- Remove unused web-*.xml file from war
* Installer: Add startup hint for non-x86
* Javadoc updates all over the place
* LogConsoleBuffer: Java 5
* Naming:
- Increase cache size and expiration time
- Add clearCache() method
- Don't use EepGet or Exec for b32
- Javadoc updates
* NetDB:
- Expire unreachable routers quickly, even if they don't have introducers,
so we don't have old data on routers that ran out of introducers.
- Fix rare NPEs at shutdown
* NTCP:
- Workaround for a rare 100% CPU bug in EventPumper
- Cleanups
* Streaming:
- Make flush() block less, by waiting only for "accept" into the
streaming queue rather than "completion" (i.e. ACK from the far end).
This prevents complete stalls when flushing, and should help performance
of apps that use flush(), like i2psnark (and SAM?).
close() still does a flush that waits for completion, as i2ptunnel
doesn't like a fast return from close().
- cleanups
* SusiDNS:
- Remove unused web-*.xml file from war
* TransportManager: Convert _transports from a List to a CHM
to prevent a rare concurrent exception
* Tunnels:
- Don't use peers < 0.7.9 for tunnels due to the old
message corruption bugs
- Javadoc
- Cleanups
* UDP:
- Beginnings of destroy message support
- Try to avoid running out of introducers by relaxing selection criteria
and increasing minimum number of potential introducers
- Avoid rare AIOOBE
- PacketBuilder refactor
- Make most classes package private
- Comments
- Logging cleanup
- Comment out a main()
- Make flush() block less, by waiting only for "accept" into the
streaming queue rather than "completion" (i.e. ACK from the far end).
This prevents complete window stalls when flushing, and should help performance
of apps that use flush(), like i2psnark (and SAM?).
close() still does a flush that waits for completion, as i2ptunnel
doesn't like a fast return from close().
- flush/close javadocs and comments
* i2ptunnel:
- Now that streaming flush() is fixed, use it in IRCClient, and
for initial data in I2PTunnelRunner, to avoid the 250 ms
passive flush delay
- Expire unreachable routers quickly, even if they don't have introducers,
so we don't have old data on routers that ran out of introducers.
- Fix rare NPEs at shutdown
- Add new option i2cp.messageReliability=none, which prevents the
router from sending MessageStatusMessages back in reply to an
outbound SendMessageMessage. Since the streaming lib always ignored
the MSMs anyway, make it the default for streaming.
This will reduce the I2CP traffic significantly.
MSM handling now avoided, but it is still fairly broken, see
comments in I2PSessionImpl2.
- Cleanups, javadoc, rate reduction
- Convert all ArrayList caching to LBQs in YKGenerator,
HMACGenerator, and AESKeyCache.
- Change DSAEngine params from Hash to new SHA1Hash, since
these were really 20 byte hashes, not 32 byte Hashes.
- Add stats to track YKGenerator caching success
- Fix YKGenerator precalculation to be much more useful by
increasing the cache size and dramatically shortening the delay
- Option cleanups
- YKGenerator cleanups
- Mark HMAC256Generator unused
* Base64: comment out some unused methods
* Remove huge whitespace in CryptoConstants
* ElGamalAESEngine:
- Reduce rates
- Check number of tags earlier
- Rework PeerID class so it depends only on peer desthash, since
peer ID is not available with compact format. Implement deferred
lookup of destination.
- Implement compact tracker requests and response handling
(may not be the final format)
- Fix Peer hashCode()
- Java 5
- Use new receivedAsReply flag in LeaseSet to mark
those received as response to a query
- Mark which methods in FloodfillPeerSelector may return
our own hash
- Redefine selectNearest() so it may return our own hash,
so it can be used for closeness measurement
- Redefine findNearestRouters() to return Hashes
instead of RouterInfos
- Fix LeaseSet response decisions for floodfills, based
on partial keyspace and closeness measurements
- List only count of published leasesets in netdb
Show definitions panel if any transport is enabled (was: only for UDP).
Use div.wideload for the whole page (was: only for transports and broken if only one of them enabled).
jbigi's build.sh had a number of failed assumptions as per where I2P and JAVA_HOME were which needed to be removed and a warning put in their place. A better solution would be to have some way to search for JAVA_HOME and I2P in common locations, but at least this solution works if you do it manually:
I2P=~i2p JAVA_HOME=/usr/lib/jvm/java-6-sun-1.6.0.20 sh build.sh dynamic
thank zzz for prodding me to do this
Loading a class would be one of them!
So we do a quick check first, If it bombs out, we delay and try again.
If it bombs after that, then we throw the ClassNotFoundException.
- Add a per-cache stat
- Limit each cache based on max memory
- Disable in UDP MessageReceiver
- Add clearAll() method to be called when under
severe memory pressure; call from Router
- add configs/win batchfiles to installer/resources/portable
* currently only pkg-portable-win32 on win32 available
need linuxers to write target preppkg-portable-nix/pkg-portable-linux
and enable pkg-portable-win32 on linux (i doubt anyone need it ?)
shell scripts should goto installer/resources/portable/configs/linux/
Enter b32 or b64 hash or dest into list box, and
check enable for whitelist. Uncheck enable and enter
i2cp.enableBlackList=true in advanced i2cp options for
blacklist. Todo: make black/whitelists radio buttons.
Fix the window size increment logic so it does it much more often.
The code increased the window size by MSS * MSS / N, like
in RFC 2581, but it did it only once every N,
so that was like MSS * MSS / N**2.
Now do it all the time, except for isolated packets like keepalives
that aren't using more than one message of the window.
Seems to speed up outbound significantly, without any
noticable increase in stream.sendsBeforeAck.
- Detect and drop dup SYNs rather than create
a duplicate connection - will hopefully fix
"Received a syn with the wrong IDs"
- Send reset for a SYN ACK with the wrong IDs
- Don't send a reset to a null dest
- Logging tweaks
- Cleanups
- Always show start button for webapps and plugins
* configclients.jsp, configupdate.jsp:
- Fix submission when entering CR in a text box
* Plugins:
- Stop all plugins at shutdown
- Log tweaks
* WebApps:
- Remove the WAC after stopping it
- Stop a WAC before starting it to prevent dups
- Send numwant=0 if we don't need peers
- Report returned complete and incomplete counts
if higher than peer count
- Allow missing peer list
- Log tweaks
- Negative delay means run immediately and inline
- Add methods to test class and run inline,
to propagate errors to the console
- Add javadoc for clients.config format
- Use new methods for plugins
- Big refactor of several classes for concurrent,
elimination of several locks
- Reduce max number of resent acks in a packet to
lower overhead
- Take incoming messages from the head of the queue,
not sure why taking them from the tail "reduces latency"
- Java 5 cleanup
- New, for filtering IRC client traffic when using SOCKS
* I2PTunnelIRCClient:
- Make filter classes static and public for use by SOCKS
- Eliminate redundant case conversion
- Pass ISON message through (jIRCii uses it for pings)
- Switch back to StringBuffer since it's used by 2 threads
- Set daemon on filter threads
* SOCKS5Server:
- Fix handling of multiple authentication methods
- Put B32 instead of B64 in Host: header, saves 450 bytes
- Eliminate some redundant lookups
- Fix http://i2p/b64/ and /eepproxy/site/ requests
- Disallow a port specified for an i2p address
- Cleanup and comments
- Replace the unused-since-2006 TimedWeightedPriorityMessageQueue
with DummyThrottle
- Don't instantiate and start TWPMQ Cleaner and OutboundRefiller
threads, part of priority queues unused since 0.6.1.11
- Don't instantiate and start UDPFlooder, it is for testing only
will be transparently unpacked with unpack200. Savings is about 60%.
Someday we will do this for suds, but we can do it for xpi2ps now.
* build: Add updater200 target
- Add js delete confirm
- Remove delete button for webapps
* i2psnark:
- Ignore a non-i2p tracker in a torrent rather than deleting
the torrent, thus "converting" a torrent to in-netowrk use
via the open trackers
- Add js delete confirm
- getFramedAveragePeerClockSkew() now returns a long (ms);
was a Long (s)
- Implement NTP-style clock slewing so the clock is adjusted
gradually
- Implement clock strata so we prefer better clocks
- Implement a timestamper in the transport so we will periodically
update the clock even if NTP is not working
This allows the router to converge the clock instead of simply
hoping the first connected peer is correct.
- Slow down NTP attempts after several consecutive failures
- Check plugin key against all installed plugins
- Reword some error messages
* VersionComparator: add '-' and '_' as valid separators
to better support plugin and java version checking
The license is BSD for launch4j and MIT for the wrapper code in head/
Changelog is in installer/lib/launch4j/web/changelog.html
Hopefully this will fix installs for 64-bit JRE on 64-bit windows.
The previous version was 2.0-RC3 2005-08-13.
The previous license was GPLv2 for launch4j and LGPLv2.1 for the wrapper code in head/
The bin/ld.exe and bin/windres.exe files were contributed by
i2p users in 2005 so the i2p installer could be built on windows.
They have not been updated for 3.0.1, so pkg builds on windows
will presumably still get 2.0-RC3.
* configclients.jsp: Use new WebAppStarter so webapps that are
started later also get the temp dir, password, and classpath
configuration just like if they were started at the beginning
* configupdate.jsp: Delay after checking for update so the
summary bar will have buttons.
- Randomize Build Message expiration to make it harder to guess hop position
- Save expired tunnel build configs for a while, so that we will still use the tunnel
and update peer stats if the reply comes in late
- Don't update our own profile for Tunnel Build Replies
- Add getRecordCount() to TunnelBuildMessage and TunnelBuildReplyMessage
so they can be extended.
- New I2NP Messages VariableTunnelBuildMessage and VariableTunnelBuildReplyMessage,
which contain the number of request slots in them.
- Convert all static assumptions of 8 slots to getRecordCount()
- Use the new VTBM if all hops in the tunnel and the OBEP or IBGW of the reply tunnel
support it, and the tunnel is 4 hops or shorter.
- Reply to a VTBM with a VTBRM of the same size
- Make BuildReplyHandler static
- Convert the currentlyBuilding List to a ConcurrentHashMap to speed reply lookups
and eliminate a global lock; don't put fallback tunnels in there
- Add new tunnel.corruptBuildReply stat
- Various cleanups and javadoc
Tested as compatible with current network, new messages untested.
- Store and report UDP clock skews even for large values, so
a badly skewed local clock will be reported to the console
- Don't shitlist for NTCP clock skew if we don't know what time it is
- If NTP hasn't worked yet, have NTCP or SSU update the clock one time
- Include failed clock skew in NTCP skew vector if there aren't many connections
- Don't include NTCP clock skews for non-established connections
- Fix framed clock skew frame size
- Report framed clock skew even if for only one peer, if NTP hasn't worked yet
- Don't log RRD errors after clock adjustment
- Reduce min skew for console warning to 30s (was 45s)
- More Java 5 cleanups
- Enable multiple parallel job runners much sooner to speed startup
- Rearrange the startup order to get the long jobs started sooner
- Don't allow the netDb readin job to clog the job queue
on a stream... AKA setting keepalive. This should fix the stuck NTCP
issue that has been bothing zzz for years.
* Set keepalive on BOB connections too, since this will assist closing
the connections in the event of a crash on a client.
Since the earliest date functions as a version number,
this will force the floodfill to flood each new version;
otherwise it won't if the earliest time hasn't changed.
- Do LS stores and verifies through client tunnels
to prevent correlation by the OBEP or FF
- Encrypt LS stores to prevent snooping by the OBEP
- Encrypt LS and RI verifies to prevent snooping by the OBEP
- Extend verify delay and timeout
- Reenable RI verifies
- Disallow simultaneous verifies for the same key
- Don't resend on verify timeout; try a different peer instead
- Adjust ff selection criteria
- More code to detect improper reuse of cached objects
after release
- Don't pass a msg with a failed IV on to
the FragmentHandler at the OBEP
- More cleanups and comments
as it may be reused if retried in another transport;
a nasty bug causing corrupt messages
- Lots of code to detect improper reuse of cached objects
after release
- Do RED dropping before the IBGW fragmenter, not after
- Change batch time to 250ms for IBGWs (was 100ms)
- Change batch time to 150ms for exploratory OBGWs (was 100ms)
- Start a new message in the fragmenter if almost full
- Fix a major, longstanding synchronization bug in the FragmentHandler
which led to corrupt messages at the endpoints
- More cleanups and comments
- Fix status to show a disconnected network error rather than
clock skew or UDP error when disconnected
- Use peer clock skew rather than clock offset for determining
whether to display clock skew error, i.e. display what matters
* Transport: Rework peer clock skew method to always return a value
by falling back to router clock offset;
Fix possible AIOOBE and divide by zero; remove logging;
reduce min number of peers
- Switch all I2PThreads to I2PAppThreads
- Run an InternalSocket as well for the HTTP Proxy
* EepGet: Use InternalSocket
* Console: Change "depth" to "length"
- Switch from ArrayList to ConcurrentHashSet in
KBucketImpl to reduce chance of deadlock;
remove periodic shuffling of the bucket,
needs to be addressed elsewhere
- When an "immediate" ack is requested, do it within
250 ms (was 2000)
- Request immediate acks when < 1/3 of window remains,
or when < 3 packets remain in window,
and every 8 packets (was when < 2 packets in window remain)
- Change requested delay to RTT/2 (was RTO/2)
- Log cleanup and javadoc
- Add support for error page translations
- Add support for external pages for all errors
- Fix lack of \r in error page headers
- HTML transitional fixes
- Cleanups
was not doing unsigned comparisons!
* Floodfill rework part 2 of N:
Store closest to the key, subject to last failed
lookup and store stats.
* FloodfillPeerSelector: Use standard XORComparator
instead of messing with BigInteger
* FloodfillVerifyStoreJob: Set correct timeout for
requeued store job
* KNDF: Rework getPeerTimout() to use 1 day averages,
and lower the min, max, and multiplication factor.
* Publish jobs: Lengthen timeout to 90s (was 30s for
routerinfos and 60s for leasesets)
* StoreJob: Limit max peer timeout to 15s for direct stores
- Fix bug where reply selector wasn't registered for
routerinfo stores, so we didn't get stats, and
we kept retrying. This also prevented verification
and profile updates for routerinfo stores.
This bug was introduced 4 years ago by the change to
store routerinfos directly.
- Add dbStoreSuccessful() to profile, and have FVSJ
call it or dbStoreFailed() as appropriate to give
credit or blame to the floodfill we stored to.
- Don't let FVSJ verify using the peer we stored to
- Reject torrents with too many pieces
- Reject torrents with a single file named *.torrent
- Increase max piece size to 2MB (was 1MB), but reduce
max number of connections to lessen ooms
- countries.txt: Convert to mixed case, include in update
- netdb.jsp: Hide all routers by default, sort and tag country names
- oldstats.jsp: Move to stats.jsp
- profiles.jsp: Show new DBH times instead of counts
* Profiles:
- Track last good and bad lookup times
and last good and bad store times,
to prep for floodfill changes
- Don't reset last-heard-about at router startup
* Checklist and Android readme fixups
- Remove the almost-unused send and receive size RateStats;
this also reduces the effective time for isActive()
- Only store DB history on-demand to save space;
fix up resulting NPEs
- Rewrite TrustedUpdate version comparator, use for netdb version table
so 0.7.10 will be sorted correctly
- Reduce netdb.jsp memory usage
- More tagging fixups
- configclients.jsp fixup for "Web console"
- Don't save config when checking for updates on configupdate.jsp
- Rework ConfigRestartBean and tag
- More tag fixups
- Add lang=xx for testing
- Add file for additional tagged strings
- summerbar related files since drz is working on it
- string requiring word reordering
- string with link in the middle
- string which looks unusual
- string found not in router console source
update the POs
- Don't hide link to configui.jsp for IE any more
- Add lang selection on configui.jsp
- Tag strings in configui.jsp
- Load console_big.css if lang == zh
- Add _x() tag for static iniitializers
- HTML transitional input tags
- Persistent lang setting with routerconsole.lang=xx
- Loading any page with ?lang=xx changes the persistent setting
- Add a custom Jetty handler to load foo_xx.jsp if it
exists for language xx. This is for jsp files with lots
of text in them. Otherwise use inline translate methods.
Not for included jsps.
- Add a script to create and update messages_xx.po translation
files, and create ResourceBundles from them
- Add class to translate strings from cached ResourceBundles
- Add translate wrappers to HelperBase, FormHandler, and *Renderer,
so calls can be made from both jsp and java files
- Add two example translations on configupdate.jsp - one in
the jsp itself and one in the helper.
- This is for strings in routerconsole only. Will be expanded
to other webapps and the router later.
- Enable per-client SessionKeyManagers for better anonymity
- tagsDelivered() now means tags are sent, not acked.
- OCMOSJ uses the new TagSetHandle object returned from tagsDelivered()
to call tagsAcked() or failTags() as appropriate.
- Assume tags delivered on an established session to
reduce streaming lib stalls caused by massive tag deliveries;
should increase throughput and window sizes on long-lived streams
- Unacked tagsets on a new session are stored on a separate list
- Don't kill an OB Session just because it's temporarily out of tags
- Increase min tag threshold to 30 (was 20) due to new speculative
tags delivered scheme, and to increase effective max window
- More Java 5 and dead code cleanups, and more comments and javadoc,
debug logging cleanups
- New I2PSocketEepGet fetches through existing tunnels
rather than through the proxy
- Use new eepget for i2psnark
- Add a fake user agent for non-proxied fetches
- Cleanups
- oops, store leaseset locally even when shutting down
(fix -16)
- Java 5 cleanups
* PRNG:
- Rename config option to prng.buffers (was router.prng.buffers)
- Change the default from 16 to 2 for I2PAppContext (saves 3.5MB)
* TunnelPool:
- Don't test tunnels when shutting down
- Less rates
- Java 5 cleanups
Prep for SessionKeyManager work in the router -
Fix up SKM renderStatusHTML(); add debug.jsp to see it;
Redefine getClientSessionKeyManager();
More cleanups
- Fix problems where a requested unsigned update would actually
kick off a signed update
- Fix problem when policy set to notify, and clicking
check for update, incorrectly causing unsigned update download
and bad messages
- Verify zip integrity of unsigned updates
- Move zip files to router dir, not base dir
- More tweaks and cleanup
- Fix BufferedStatsLog so it works at all
- Don't instantiate BufferedStatsLog unless stats.logFilters
property is defined (restart now required to enable logging)
This eliminates the StatLogWriter thread and a decent
amount of memory.
- Move two CLI classes out of the lib
- Commment out places where getStatLog() isn't checked for null
- Cleanups
* DataHelper, Router:
- Save config files in UTF-8 rather than the default encoding,
since we read them in UTF-8
* jetty.xml: Change encoding to UTF-8
* logs.jsp: Add system encoding
* NTCP: Clean up clock skew shitlist message
* Shitlist: Clean up expire message
* WorkingDir: Ensure modified files are processed with UTF-8 encoding
* Merge in dr|z3d and my own html fixes for router console java and jsp
files so that Opera (and now IE?) doesn't puke anymore on the missing
and misplaced HTML tags.
* Optimized all jsp files so that they are shorter to save space, which
is then used to fix the broken HTML. We should break even space-wise.
* Bump to -13.
- Make IRC Proxy non-shared, delayed-start, close-on-idle
for new users, for the anonymity benefits (see "Shared Clients,
Correlation and Collusion" http://zzz.i2p/topics/217 )
- Remove "experimental" flag on new client options
* configclients.jsp: Add full path to config file
* Add some path help in some config files
* Code Janitor time! Many fixes and documenting fixes that should be
done in the future. for the most part, this is a general code cleanup.
* On smaller/embedded systems, the "final" keyword cleanups will have
more of an impact than on larger systems.
* Document missing hashCode() methods.
* Unhide more variables to make code easier to read.
- Add new unsigned update option, triggered by
last-modified date, using the new EepHead.
Buttons still are not hidden after download complete.
- Make the .sud updater use the temp dir when proxied
- Several cleanups
- small pedantic fix
* streaming:
- Fix a deadly race condition.
- Some small pedantic fixes.
* core:
- Fix a deadly race condition.
* BOB:
- Fixed some races that occured from fixing races in streaming and core.
- Some badly needed code refactoring to depend less on the database.
under the assumption that they are already talking
to most of the routers, so there's no reason to reject. This may drive them
to their conn. limits, but it's hopefully a temporary solution to the
tunnel build congestion. As the net grows this will have to be revisited.
Under the one-packet-enough theory, and the fact that most
tunnels in a x+1 pool are of length x, variable lengths
don't really help that much. Also, a default of 1 led
to all sorts of problems with iMule/SAM, who was not
setting the variance properties.
This will affect exploratory tunnels for new users,
and those that have never saved a change on configtunnels.jsp,
and iMule users 1.4.5 and earlier.
- Move from a single connection limit threshold (80%) to
two (75% and 87%), and only start rejecting tunnels
at the higher threshold, to increase build success
- Move some limit methods from the transports to TransportImpl
- Add limit methods with a threshold argument
address to no address, so that inbound NTCP is disabled
after SSU detects a firewall. When UPnP was apparently successful
but the router is still firewalled (due to an additional
software firewall or a bad UPnP indication, for example)
the router will now remove the NTCP address.
* General cleanup on streaming and ministreaming.
This fixes some compile warnings, and prepares for a larger fix.
There is no code-flow changes, just lint. One warning remains as I am
unsure exactly how to solve the problem yet.
* Remove the last reference to my eepsite as a "news.xml" source,
and likewise stop my public key from being included
among valid release signing keys.
---------------------------------------------------------------------------------------
* netdb.jsp: Add country chart at bottom, clean up version chart
---------------------------------------------------------------------------------------
* Browser Launch: Wait until the routerconsole is up before launching the browser
---------------------------------------------------------------------------------------
* ReseedHandler:
- check for upper case HREF to be compatible with apache indexes
- Limit exploratory tunnels to connected peers when over
half the connection limit (was 80%)
- Have the high capacity tier fall back to a new connected tier
before moving on to the not failing tier
so that tunnel build success doesn't collapse
* PeerTestJob:
- Limit to connected peers
- Move reseeding from the routerconsole app to
the router, so that we can bootstrap an embedded router lacking a routerconsole
(iMule or android for example), without additional modifications.
This allows better integration between the reseeding function
and the netDb.
- Call reseed from PersistentDataStore, not from the
routerconsole init, and start seeding as soon as the netdb has read
the netDb/ directory, not when the console starts.
- Wake up the netdb reader as soon as reseeding is done,
rather than waiting up to 60s.
- Don't display the reseed button on the console until the
netdb initialization is done.
* NetDb:
- Fix an NPE on early shutdown
* RouterConsoleRunner:
- Catch a class not found error better
- Put the duplicate classes back in the war, because
the jar is not in the wrapper.config classpath in existing
installs. We could take them out of the jar, but then
they won't be available for standalone snark and future
updates via snark.
- Delete the dist/ dir in distclean
- Move FetchAndAdd to static inner class
- Remove duplicate classes from i2psnark.war (120KB);
fixes sporadic FetchAndAdd IllegalAccessError
- Fix standalone build to include i2psnark.jar since classes
aren't in the .war anymore
- Have standalone jetty use I2PAppContext temp directory
- Replace launch-i2psnark.jar with launch-i2psnark script,
since RunStandalone is in i2p.jar
- Clean up jetty-i2psnark.xml, turn off jetty logging
- Remove standalone build from the pkg target in the main build.xml
- Extract files from jar to temp dir, load from that dir, then
copy to the base dir if we have permissions (and failing silently
if we don't), so we have optimized libs and no complaints
when we have a read-only base dir.
- Don't cd to script location, no longer required
* RouterLaunch:
- If no wrapper, put wrapper.log in system temp dir
unless specified with -Dwrapper.logfile=/path/to/wrapper.log
or it already exists in CWD (for backward compatibility)
- Append rather than replace wrapper.log
- Pass wrapper log location to router as a property, so that logs.jsp can find it
* logs.jsp:
- Get wrapper log location from a property too
* runplain.sh:
- Add path substitution to runplain.sh on install
- Pass I2P base dir to the router as a property
* wrapper.config:
- Put wrapper.log in system temp dir for new installs
- Pass I2P base dir to the router as a property
* WorkingDir:
- Don't migrate an existing install by default
- Never migrate the data (too hard)
- Change the wrapper.config classpath to one line: lib/*.jar
This means we lose control of classpath load order, so move the windows installer
jars copy.jar, delete.jar, and exec.jar to a new installer/ directory so
these jars won't be in the classpath or potentially conflict, since
copy.jar and delete.jar include FileUtil.class, and we don't want to have
to remember to add them to the updater if we ever change FileUtil.class.
Delete the installer/ directory in postinstall.sh since it is windows-only.
- Move the console css from default.css in the .war to docs/themes/console/console.css,
and support console themes in the main console with routerconsole.theme=foo
(previous izpack was 3.7.2 from 2005-04-22)
izpack 4.3.0 from :
http://dist.codehaus.org/izpack/releases/4.3.0/IzPack-install-4.3.0.jar
SHA1 f06da6b26ac2c68fed64ab38980352989b8d8841
(no signatures or sha1sums found on website, and the jar is unsigned)
License: Apache 2.0
upack izpack:
java -jar IzPack-install-4.3.0.jar
or
java -jar IzPack-install-4.3.0.jar -console
get the standalone-compiler.jar from the installation lib/ directory:
SHA1 6d2b4a5657bfb864a333b1c4b1c0f8223aa57d80
(no signatures or sha1sums found on website, and the jar is unsigned)
This fixes the bug with the install windows centered in all the
workspaces, not the current workspace. And who knows what other
bugs in the last 4 years.
To fix Vista (and presumably Windows 7) permissiom problems,
add a run-privileged flag for those, and run the new fixperms.bat
which calls icacls to add the privileges to the install directory.
Add support for 6 more language packs found in the new release.
Change from ISO3 codes to native language names.
Disable creation of the i2p.tar.bz2 file in build.xml
(distributed as i2pheadless-0.7.x.tar.bz2), as izpack 4.3.0 now
supports headless installation with java -jar i2pinstall.exe -console.
Update INSTALL.txt and INSTALL-headless.txt accordingly.
* BOB now cleans up tunnels, although they can take up to 5 minutes to
disapear. This is due to the fact that the streaming lib doesn't
actually remove the connections properly and kill them off when the
manager is destroyed. I'm not certain if this is a bug, or a feature,
but it sure is annoying, and you have to wait for the connections to
time out. What should happen is the streaming lib should cause an IO
error to the pending read or write.
Eliminate all uses of the current working directory, and
set up multiple directories specified by absolute paths for various uses.
Add a WorkingDir class to create a user config directory and
migrate files to it for new installs.
The directory will be $HOME/.i2p on linux and %APPDIR%\I2P on Windows,
or as specified in the system property -Di2p.dir.config=/path/to/i2pdir
All files except for the base install and temp files will be
in the config directory by default.
Temp files will be in a i2p-xxxxx subdirectory of the system temp directory
specified by the system property java.io.tmpdir.
Convert all file opens in the code to be relative to a specific directory,
as specified in the context. Code and applications should never open
files relative to the current working directory (e.g. new File("foo")).
All files should be accessed in the appropriate context directory,
e.g. new File(_context.getAppDir(), "foo").
The router.config file location may be specified as a system property on the
java command line with -Drouter.configLocation=/path/to/router.config
All directories may be specified as properties in the router.config file.
The migration will copy all files from an existing installation,
except i2psnark/, with the system property -Di2p.dir.migrate=true.
Otherwise it will just set up a new directory with a minimal configuration.
The migration will also create a modified wrapper.config and (on linux only)
a modified i2prouter script, and place them in the config directory.
There are no changes to the installer or the default i2prouter, i2prouter.bat,
i2prouter, wrapper.config, runplain.sh, windows service installer/uninstaller,
etc. in this checkin.
* Directories. These are all set at instantiation and will not be changed by
* subsequent property changes.
* All properties, if set, should be absolute paths.
*
* Name Property Method Files
* ----- -------- ----- -----
* Base i2p.dir.base getBaseDir() lib/, webapps/, docs/, geoip/, licenses/, ...
* Temp i2p.dir.temp getTempDir() Temporary files
* Config i2p.dir.config getConfigDir() *.config, hosts.txt, addressbook/, ...
*
* (the following all default to the same as Config)
*
* Router i2p.dir.router getRouterDir() netDb/, peerProfiles/, router.*, keyBackup/, ...
* Log i2p.dir.log getLogDir() wrapper.log*, logs/
* PID i2p.dir.pid getPIDDir() wrapper *.pid files, router.ping
* App i2p.dir.app getAppDir() eepsite/, ...
*
* Note that we can't control where the wrapper actually puts its files.
All these will be set appropriately in a Router Context.
In an I2P App Context, all except Temp will be the current working directory.
Lightly tested so far, needs much more testing.
- Don't instantiate unused SessionKeyPersistenceHelper
- Use TransientSessionKeyManager instead of PersistentSessionKeyManager
- Add generics to TransientSessionKeyManager to help understand it
- Change initial session map size to 64 (was 1024)
- Prepare for per-destination SessionKeyManagers in ElGamalAESEngine
- Adjust interface to have persistent and non-persistent methods,
to prepare for partial storage in RAM
* PersistentDataStore:
- Cleanup, simplify, and concurrentify
- Tweak stats
- Remove write limit
- Flush to disk on shutdown
- Don't write out what we just read in
- Reduce displayed entries from 300 to 100
- Add ability to go forward or back
- Make textareas bigger
- Clean up file names
- Clarify messages about filter and search
- No longer use i2np.udp.forceIntroducers
- Tweak UDP port qualification
- Fix allowing low ports again
- Add option to completely disable NTCP, for those behind nasty firewalls
- Use SSU reachability rather than global reachability for determining NTCP reachability,
since we are now reporting NTCP reachability too
the UPnP device goes away - same as for HTTP POST
- Stuff the port mapping requester into a thread so it doesn't
delay everything for several seconds
- Handle UPnP devices that return IP = 0.0.0.0
- Better HTML output when no IP found
- Tweak logging
- Set Disposer thread name
- Keep the control point running after we find an IGD,
so that we get notifications of it leaving or
coming back or replaced.
- Set Status to OK for local public addresses or UPnP port open
- Allow UDP address changes after we transition to firewalled
- Have NTCP start reporting reachability status, this will
get OK on the console more often and mask UDP problems,
which might be good or bad...
- Fix UDP port configuration
- Reword and rearrange configuration options again
- Rearrange configuration help
- More right-alignment on config
- Prevent Concurrent modification exception in UPnP
- UPnP HTML output tweaks - remove "plugin" references
- Move UDP message failed log from WARN to INFO
- Short-circuit message history call in UDP
- Try to fix locking to prevent duplicate destinations when using
the new option new-dest-on-resume. Still not right for shared clients
but should be better for non-shared.
SSU reachability is OK. i2np.ntcp.autoip=always for the old behavior.
autoip default is now "true".
i2np.ntcp.hostname=xxx now trumps i2np.tcp.autoip.
- SSU always tells NTCP when status changes.
now defaults to auto; and configured now trumps auto.
Port configuration now does not affect whether inbound
NTCP is enabled - the host configuration alone can do that.
* more BOB fixes, complete with warnings when things go wrong, and
success messages when things turn around and go right. Terminates
early so that applications wait no more than 10 seconds or so.
* Reversed a few earlier patches that caused some odd behavior.
* Changed some core println()'s to debugging messages.
- Correct the meanings of the i2np.ntcp.autoip and i2np.ntcp.autoport
advanced config. If you have one of these set but not the other, you
will have to adjust your configuration on config.jsp.
Jetty forms the temp directory name from, among other things, the bound
address. The patch in 0.7.2 to bind to IPV6 addresses led to ':' in the
directory name, which is not allowed on windows. Change these to '_'.
* PeerProfile:
- Replace a hot lock with concurrent RW lock
- Rewrite ugly IP Restriction code
- Also use transport IP in restriction code
* Transport: Start the previously unused CleanupUnreachable
java.lang.IllegalStateException: Unable to serialize the message (net.i2p.data.i2np.DatabaseStoreMessage): Invalid key type
at net.i2p.data.i2np.I2NPMessageImpl.toByteArray(I2NPMessageImpl.java:238)
at net.i2p.router.transport.ntcp.NTCPConnection.bufferedPrepare(NTCPConnection.java:668)
at net.i2p.router.transport.ntcp.NTCPConnection.send(NTCPConnection.java:295)
at net.i2p.router.transport.ntcp.NTCPConnection.enqueueFloodfillMessage(NTCPConnection.java:412)
at net.i2p.router.transport.ntcp.NTCPConnection.enqueueInfoMessage(NTCPConnection.java:373)
at net.i2p.router.transport.ntcp.NTCPTransport.outboundMessageReady(NTCPTransport.java:179)
at net.i2p.router.transport.TransportImpl.send(TransportImpl.java:339)
...
* Code janator work, basic corrections involving @Override, and
appling final where it is important. Also fixed some equals methods
and commented places that need fixing.
* Catch NPE in NTCP.
This possibly augments fix 2009-04-11 welterde below.
* Various LINT on NTCP sources, and removal of space-wasting
spaces at end of lines in sources touched.
* i2ptunnel janitorial work and fixes on most locks.
Some locks still need work, and are marked with LINT in the comment.
Just grep for "LINT" to see where the remaining places are.
* More BOB threadgroup fixes, plus debug dump when things go wrong.
* Fixes to streaminglib, I2CP, which are related to the TG problem.
* JavaDocs fixups.
- Enable IPv6 stack in the JVM, hopefully won't break anything
- Patch Jetty to support binding to IPv6 addresses
- Allow multiple bind addresses for the router console
in the clients.config file; for new installs the
default is now "127.0.0.1,::1"
- Change most instances of "localhost" to "127.0.0.1"
throughout the code
* Router:
- Move some classes to private static inner
* SimpleTimer2, SimpleScheduler fixed so that the threads all run from
The main threadgroup, not in the current possible child threadgroup.
So long as any SimpleTimer2/SimpleScheduler is started *BEFORE* any
child threadgroups, the constructors are threadgroup safe. What would
be super cool is if they were to be all jailed within thier very own
threadgroup too, but, I2P isn't up to the task of this yet.
* Fixes to BOB to ensure the above is true.
To use the application, you need to (by default) add the following to clients.config :
# desktopgui
clientApp.6.args=
clientApp.6.delay=5
clientApp.6.main=desktopgui.Main
clientApp.6.name=desktopgui
clientApp.6.startOnLoad=true
- Maintain a router hash -> IP map in transport,
to support additional IP checks
(unused for now)
- Catch error on pre-2.6 kernels
- Some concurrent conversion
- Fix an HTML error on peers.jsp
- Remove unused calculators and RateStats:
CapacityCalculator, StrictSpeedCalculator, IsFailingCalculator;
sendFailureSize, processSuccessRate, processfailureRate, commErrorRate,
tunnelTestResponseTimeSlow
- Reduced number of Rates in these RateStats:
sendSuccessSize, receiveSize, rejectRate, failRate
- ~5KB/profile savings total
- Deflate speed calculation once an hour instead of once a day,
to improve fast tier selection
the code CWD is / but the only writable directory is
/data/data/net.i2p.router/files/ - still a ton of places to be
fixed, will be fixed up as things get working
- Load some config files from resources at startup
- Fix up logging
- Add reseed capability, by copying some code over from routerconsole
- Deal with conflicting bouncycastle libs
- Move ConEvent from SimpleTimer to SimpleScheduler
- Move RetransmissionTimer (ResendPacketEvent)
from SimpleTimer to new SimpleTimer2
- Move ActivityTimer and Flusher from SimpleTimer to RetransmissionTimer
- SimpleTimer2 allows specifying "fuzz" to reduce
timer queue churn further
As Debian's package building system is rather complicated and requires root
access unconditionally for some reason, doing it from ant isn't really
feasible. However to build any debian package anywhere is the same system, so
including helpful documentation on how to use that system as an ant build
target would be most useful in this case. Hopefully Debian users will
only have to deal with the already built .deb anyway.
- Fix tunnel reduction/restore, hook in the GUI
- Hook leaseset encryption into the GUI
- Implement saves for all the new stuff
- Add cancel button
- Add b32 display for non-http servers
- Prep for CONNECT
- Fix error msg when connection goes away
This sets i2p up as a functional Debian source package. dpkg-buildpackage
will build i2p using ant preppkg (tarball takes too long and not
helpful). It creates a binary .deb archive of the i2p installation,
which when installed goes into /var/lib/i2p as the non-root user i2p,
and adds an /etc/init.d script to start it up.
Some problems not yet solved:
1) under Debian the conf should go into /etc/i2p, but since it doesn't
things like the eepsite index file get overwritten if you reinstall.
should check for those somehow and not replace them, or ask the user.
2) under Debian they like it if you split the generated data from the
static code, so i2p should go into /usr/lib/i2p maybe, but its
netDB and any other cache files into /var/cache/i2p
that's important not just for organization, but also /var is often
on a filesystem optimized for churn. For now just put it in /var/lib
3) i2p is supposedly architecture independant, but it does choose a
native jbigi library on postinstall, so does that really count
as architecture independant?
- Change default length to 2+0
- Cleanup helper code
- Stub out the following new options (C=client, S=server):
+ Access list (S)
+ Certificate type (S)
+ Encrypted LeaseSet (S)
+ New dest on idle restart (C)
+ Tunnel closure on idle (C)
+ Tunnel reduction on idle (C,S)
Pass message timeout through new I2CP message
SendMessageExpiresMessage, so that the router
uses the same expiration as the streaming lib.
Should help reliability.
* I2CP:
Implement new I2CP message ReconfigureSessionMessage.
Will be used for tunnel reduction.
- Disable ExploreKeySelectorJob completely, just have
StartExplorersJob select a random key if queue is empty
- Add netDb.alwaysQuery=[B64Hash] for debugging
- Queue results of exploration for more exploration
- Floodfills periodically shuffle their KBuckets, and
FloodfillPeerSelector sorts more keys, so that
exploration works well
- Give each search a minimum of time even at the end
- Fix ExploreJob exclude peer list
- Always add floodfills to exclude peer list
- Don't queue keys for exploration or run ExploreJob
if floodfill
- Allow floodfills to return non-floodfills in
a DSRM msg so exploration works
- Expire routers with introducers after 90m.
This should improve reachability to firewalled routers
by keeping introducer info current.
- Expire routers with no addresses after 90m.
Implement Base32 Hash hostnames, via the naming service.
Names are of the form [52-characters].i2p, where
the 52 characters are the Base32 representation of our
256-byte hash. The client requests a lookup of the hash
via a brief I2CP session using new I2CP request/reply
messages. The router looks up the leaseset for the hash
to convert the hash to a dest. Convert the I2PTunnel
'preview' links to use Base32 hostnames as a
demonstration.
- Cleanup max connections code
- Add i2np.udp.maxConnections
- Set max connections based on share bandwidth
- Add haveCapacity() that can be used for connection
throttling in the router
- Reject IBGW/OBEP requests when near connection limit
- Reduce idle timeout when near connection limit
* Tunnel request handler:
- Require tunnel.dropLoad* stats
- Speed up request loop
is closed from under it.
Enhancement: BOB can now clear a destination in under 1 second with the above fix.
BOB also will do a thread dump when something really aweful happens,
so that developers/users can help in debugging.
- Fixes and cleanups when NTCP and/or UDP transports disabled
- More TCP removal cleanup
- Clean up bandwidth limiting, centralize defaults
- Force burst to be >= limit
- Increase default bw to 48/24, burst 64/32
* Delay shitlist cleaner at startup
* Strip wrapper properties from client config
* Define multiple cert type
* Prohibit negative maxSends in streaming
* HTML fixup on configtunnels.jsp
* Increase wrapper exit timeout from default 15s to 30s
- Better handling of 504 gateway timeout
(keep going up to limit of retry count rather
than just one more partial fetch)
- Add -t cmd line option for timeout
- Better handling of 403, 409, 503 errors
- Don't keep going after unknown return code
- Don't delay before exiting after a failure
- Don't create SnarkManager instance until first call,
so it doesn't create the i2psnark dir, read the config,
etc., for single Snark instances.
- Don't read i2psnark.config twice; fix setting
i2psnark.dir
- More Snark constructor changes for calling from router
- Make max connections per torrent configurable
- Remove static instances of I2PSnarkUtil, ConnectionAcceptor,
and PeerCoordinatorSet
- Convert static classes in Snark to listeners
- Fix Snark to work in single torrent mode again
- Should now work with multiple single Snarks
- Refactor to allow running a single Snark without a SnarkManager again,
by moving some things from SnarkManager to I2PSnarkUtil,
having Snark call completeListener callbacks,
and having Storage call storageListener callbacks.
This is in preparation for using Snark for router updates.
Step 2 is to allow multiple I2PSnarkUtil instances.
- Big rewrite of Storage to open file descriptors on demand, and
close them when unused, so we can support large numbers of torrents.
- Add more info to Connection.toString() for debugging
- Fix lifetimeMessages{Sent,Received} stats
- Reduce RTT damping to 0.875 (was 0.9)
- Add a stream.con.initialRTT.{in,out} stats
- Use the depend task with caching for more accurate dependencies
- Make sure the routerconsole gets the latest router version
- Fix addressbook repeated builds
when the initial data is larger than one MTU,
e.g. HTTP GETs with large URLs, CGI params or cookies,
or large HTTP POSTS:
- Don't reject additional packets received without a
send stream ID (i.e. sent before the SYN ACK was received)
- Put unknown non-SYN packets on the SYN queue also
so they won't be rejected
- Reduce flusher delay to 250ms (was 500)
- Flush unless window is full (was window is non-empty)
- Don't distclean in the updaterRouter target
- Don't make prepUpdate and prepupdateSmall depend
on distclean
- Don't make susimail build always clean
- Make pkg depend on distclean to be sure
- Clean out more routerconsole and susidns files in 'ant clean'
- i2ptunnel, routerconsole, susidns:
Only build WEB-INF when necessary
- systray: Only build jar when necessary
- Don't build i2psnark standalone for the updater target
- Add a signed Certificate type
- Add a main() to PrivateKeyFile to generate
Destinations with various Certificate types
- Add a VerifiedDestination class to check Certificates
of various types
- Add a HashCash library from http://www.nettgryppa.com/code/
(no distribution restrictions)
- Allow non-null Certificates in addressbook
- Update dbLookup profile stats in FloodOnlySearchJob
and FloodfillVerifyStoreJob
- Fix response time store in profile in SearchJob
* profiles.jsp: Don't override locale number format,
clean up the response time output for floodfills
- Don't do peer tests when hidden
- Don't offer to introduce when hidden
- Don't continually rebuild routerInfo when hidden
- Don't continually rebuild routerInfo when
i2np.udp.internalPort is set but i2np.udp.port is not
- Remove some unused functions
Added socketSoTimeout
CHANGED RetransmissionTimer is now public
FIXED SimpleTimer has a way to be stopped, and reap it's children
CLEANUP A few javadoc additions, where I could figgure out bits
CLEANUP all code that needed to catch the timeout exception for socketSoTimeout
Added socketSoTimeout
CHANGED RetransmissionTimer is now public
FIXED SimpleTimer has a way to be stopped, and reap it's children
FIXED Lots of javadoc additions, where I could
CLEANUP all code that needed to catch the timeout exception for socketSoTimeout
and fixed all the broken mainstream applications depending on it.
Fixed a grave bug in SimpleTimer.
Fixed Steraming Timer to be public.
Fixed a pile of JavaDoc comments, and reformatted the files I touched.
* Tunnels:
- Add missing message accounting for inbound gateways,
we were underestimating participating traffic because of it,
and the tunnels were classified "inactive"
- Add participating tunnel role on tunnels.jsp
- Correctly check inbound and outbound total bw limits separately
- Fix up and actually use the tunnel.participatingMessageCount stat,
favor it if lower than the total bw stat, so that
client traffic isn't included for throttle decisions
- Reduce min message count from 60 to 40
* Tunnel Dispatcher:
- Add tunnel.participatingBandwidth stat
- Remove all 3h and 24h stats
* Throttle:
- Use 60s rather than 10m tunnel.participatingMessageCount stat
- Fix a summary bar message
* Tunnel Dispatcher: Update tunnel.participatingMessageCount
every 20s, rather than at tunnel expiration, to maintain
a more current stat
* NetDb: Add netDb.exploreKeySet stat
* netdb.jsp: Add parameter ?r=xxxxxx to view a single routerinfo,
and ?r=. to view our own; change links on other pages too
- Add config i2psnark.linkPrefix to enable access to completed
torrents from a different machine - examples:
i2psnark.linkPrefix=file://///localserver/path/to/files/
i2psnark.linkPrefix=http://localwebserver/path/
Stop i2psnark, add to i2psnark.config, restart
- Remove Galen and NickyB trackers
peer for direct RouterInfo stores, to mimimize floodfill
connections
* Peer Profiles: Classify connected peers as "active",
which will help improve the fast pool
* Transport Manager: Add isEstablished(Hash)
- New, disabled by default, except for blocking of
forever-shitlisted peers. See source for instructions
and file format.
* Transport - Reject peers from inbound connections:
- Check IP against blocklist
- Check router hash against forever-shitlist, then block IP
- Try to pick better introducers by checking shitlist,
wasUnreachable list, failing list, and idle times
- To keep introducer connections up and valid,
periodically send a "ping" (a data packet with no data and no acks)
to everybody that has been an introducer in the last two hours
- Add a stat udp.receiveRelayRequestBadTag, make udp.receiveRelayRequest only for good ones
- Remove some 60s and 5m stats, leave only the 10m ones
- Narrow the range for the retransmit time after an allocation fail
- Adjust some logging
- Don't write router key backup when leaseSet keys are updated
- Synchronize to prevent concurrent writes (thanks Galen!)
- Backup keys every 7 days instead of every 5 minutes
- Repair corrupted files with wrong length rather than die
- Register shutdown hook to properly shutdown torrents when
the router shuts down, hopefully will reduce corruption
- Add Galen tracker
- Add a note about how to chane directory
to control which .wars in webapps/ get run. Apps are enabled
by default; disable by (e.g.) webapps.syndie.startOnLoad=false
Config file is written if it does not exist.
Implement methods for use by upcoming configclients.jsp.
- Remove several more stats
- Don't publish bw stats in first hour of uptime
- Publish floodfill stats even if other stats are disabled
- Changes not effective until 0.6.2.1 to provide cover.
* Reachability: Restrict peers with no SSU address at all from inbound tunnels
* News:
- Add display of last updated and last checked time
on index.jsp and configupdate.jsp
- Add a function to get update version (unused for now)
* config.jsp: Add another warning
* Shitlist:
- Add shitlistForever() and isShitlistedForever(), unused for now
- Sort the HTML output by router hash
* config.jsp: Add another warning
* netdb.jsp:
- Sort the lease HTML output by dest hash, local first
- Sort the router HTML output by router hash
- Sort the leases by expiration date in TunnelPool.locked_buildNewLeaseSet()
to make later LeaseSet comparisons reliable. This cleans up the code too.
- Fix broken old vs. new LeaseSet comparison
in ClientConnectionRunner.requestLeaseSet(),
so that we only sign and publish a new LeaseSet when it's really new.
Should reduce outbound overhead both in LeaseSet publishing and LeaseSet bundling,
and floodfill router load, since locked_buildNewLeaseSet() generates
the same LeaseSet as before quite frequently, often just seconds apart.
- Add exception to enforce max # of leases = 6, should be plenty
- Rewrite TunnelPool.locked_buildNewLeaseSet() so it doesn't add lots of
leases and then immediately remove them again, triggering
the new leaseSet size exception
- Remove the now unused LeaseSet.removeLease(lease) and
LeaseSet.removeLease(index)
- Store first and last expiration for efficiency
- Use CapacityBonus rather than ReachablilityBonus in the Capacity calculation
- Persist CapacityBonus rather than ReachabilityBonus
- Include SpeedBonus in the Speed calculation
- Prevent negative values in Speed and Capacity when using bonuses
- Clean up SpeedCalculator.java
* Stats: Add a fake uptime if not publishing stats, to get participating tunnels
* build.xml:
- Add an updateSmall target which includes only the essentials
- Clean up the build file some
- Remove empty eepsite/ and subdirs from i2pupdate.zip
* configtunnels.jsp: Add warning
* i2psnark: Catch a bencode exception (bad peer from tracker) earlier
* i2psnark-standalone: Fix exception http://forum.i2p/viewtopic.php?p=12217
- Call the previously unused profile.tunnelTestFailed()
(redefined to include a probability argument)
and severely downgrade a peer's capacity upon failures,
depending on tunnel length and direction.
This will help push unreachable and malicious peers
out of the High Capacity tier.
- Put recent fail rate on profiles.jsp
* ProfileOrganizer: Logging cleanup
* eepsite_index.html: Update add-host and jump links
* HTTP Proxy: Remove trevorreznik jump server from list
* TunnelPeerSelectors:
- Re-enable strict ordering of peers,
based on XOR distance from a random hash
- Restrict peers with uptime < 90m from tunnels (was 2h),
which is really 60m due to rounding in netDb publishing.
* i2psnark:
- Limit max pipelined requests from a single peer to 128KB
(was unlimited; i2p-bt default is 5 * 64KB)
- Increase max uploaders per torrent to 6 (was 4)
- Reduce max connections per torrent to 16 (was 24) to increase
unchoke time and reduce memory consumption
- Strictly enforce max connections per torrent
- Choke more gradually when over BW limit
* help.jsp: Add a link to the FAQ
* peers.jsp: Fix UDP direction indicators
* hosts.txt: Add update.postman.i2p
- Randomize the PeerCheckerTask start times to make global limiting
work better
- Calculate bw limits using 40s rather than 4m averages to make
bw limiting work better
- Change default bw limit from uplimit/3 to uplimit/2 due to
overhead reduction from the leaseset bundling change
- Restrict peers requiring introducers from inbound tunnels,
since it's slow and unreliable... and many of them advertise
NTCP, which seems unlikely to work
- Provide warning on summary bar if firewalled with inbound NTCP enabled
* Stats: Remove the bw.[send,recv]Bps[1,15]s stats unless
log level net.i2p.router.transport.FIFOBandwidthLimiter >= WARN
at startup (you didn't get any data unless you set the log level anyway)
* oldstats.jsp: Don't put 2 decimal places on integer event counts
* Remove the Internals link from the menu bar
* i2psnark: Extend startup delay from 1 to 3 minutes
- Fix a couple of tunnel cache cleaning bugs
- Cache based on source+dest pairs rather than just dest
- Send the reply leaseSet only when necessary,
rather than all the time (big savings in overhead)
- Enable persistent lease selection again
- Logging tweaks
- Fix a bug from -19 causing the persistent lease selection
removed in -17 to be back again
- Use netDb-listed-unreachable instead of detected-unreachable
for exclusion of unreachable peers from selected leases,
as there are potential anonymity problems with using
detected-unreachable
- Tweak logging some more
* NetDb stats: Remove a couple more including the inefficient stat_identities
- Track unreachable peers persistently
(i.e. separately from shitlist, and not cleared when they contact us)
- Exclude detected unreachable peers from inbound tunnels
- Exclude detected unreachable peers from selected leases
- Exclude detected unreachable floodfill peers from lookups
- Show unreachable status on profiles.jsp
- Extend shitlist time from 4-8m to 40-60m
- Add some shitlist logging
- Don't shitlist twice when unreachable on all transports
- Exclude netDb-listed unreachable peers from inbound tunnels;
this won't help much since there are very few of these now
- Remove 10s delay on inbound UDP connections used for the
0.6.1.10 transition
- Track and display UDP connection direction on peers.jsp
- Show shitlist status in-line on profiles.jsp
- Back out strict peer ordering until we fix SSU
- Back out persistent lease selection until we fix SSU
- Fix detection of UDP REJECT_UNSOLICITED by recording status on expiration
- Increase known Charlie time to 10m; 3m wasn't enough
- Don't continue retransmitting peer test if we know Charlie
- Don't run multiple peer tests at once
- Tighten test frequency range to 6.5-19.5m, was 0-26m
* NTCP: Don't drop a connection unless both directions are idle;
Fix idle time for outbound connections
* Outbound message: Make sure cached lease is in current leaseSet
* Stats: Put all NetworkDatabase stats in same group
* TunnelPool: Stop building tunnels and leaseSets after client shutdown
* i2psnark: Add locking to prevent two I2CP connections
* HostsTxtNamingService: Add reverse lookup support
* Outbound message: Minor cleanup
* i2psnark TrackerCLient: Minor cleanup
* checklist.txt: Minor edit
* hosts.txt: Add perv.i2p, false.i2p, mtn.i2p2.i2p
* i2ptunnel.config: Change CVS client to mtn
* netdb.jsp: Show leaseSet destinations using reverse lookup
* profiles.jsp: First cut at showing floodfill data
lease to reduce out-of-order delivery.
* ExploratoryPeerSelector: Back out the floodfill peer exclusion
for now, as it prevents speed rating of those peers
- Add support for secondary open trackers
- Refactor and simplify the TrackerClient code
- Add welterde's tracker to the default list
- Don't have eepget retry announces
- Slow down tracker contacts if they've failed for a while
- Add some debug support showing connections (?p=2)
- Fix counting so it really takes 4 consecutive failures
rather than 4 total to remove a tunnel
- Credit or blame goes to the exploratory tunnel as well
as the tunnel being tested
- Adjust tunnel test timeout based on tunnel length
* ExploratoryPeerSelector: Tweak logging
* ProfileOrganizer: Adjust integration calculation again
* build.xml: Add to help
* checklist.txt: Tweak
* readme.html: Fix forum links
* netDb: Remove tunnel.testFailedTime
- Exclude floodfill peers
- Tweak the HighCap vs. NonFailing decision
* i2psnark: Increase retries for .torrent fetch
* IRC Proxy: Prevent mIRC from sending an alternate DCC request
containing an IP
* readme.html: Reorder some items
* Stats: Add some more required stats
* Streaming lib: Fix slow start to be exponential growth,
fix congestion avoidance to be linear growth.
Should speed up local connections a lot, and remote
connections a little.
- Use more recent stats to calculate integrationory.txt
- Show that fast peers are also high-capacity on profiles.jsp
* readme.html: Update Syndie link
* TunnelPool: Update comments
* netDb: Report 1-2h uptime as 90m to further frustrate tracking,
get rid of the 60s tunnel stats
(effective as of .33 to provide cover)
- Fix a bug that caused a single FloodfillOnlySearchJob
instance to be run multiple times, with unpredictable
results
- Select ff peers randomly to improve reliability
- Add some bulletproofing
- Don't require a peer to be high-capacity to be
well-integrated (not used for anything right now,
but want to get it right for possible floodfill verification)
- Don't fall back to median for high-capacity threshold
if the mean is higher than the median, this prevents
frequent large high-capacity counts
- Fix high-capacity selector that picked one too many
* Console: put well-integrated count back in the summary
(fixes command line status)
* UpdateHandler:
- Fix byte count display
- Display final status on router console
- Don't allow multiple update jobs to queue up
- Increase max retries
- Code cleanup
- Don't show 'check for update' button when update in progress
- Enhance error messages
* Give the Jetty build file ability to ask permission
before downloading the Jetty archive from the web,
and to verify its SHA1 + MD5 hashes. Adjust the main build file
in accordance with this change.
* Improve the release checklist.
- Prevent peers with matching IPs from joining same tunnel.
Match 0-4 bytes of IP (0=off, 1=most restrictive, 4=least).
Default is 2 (disallow routers in same /16).
Set with router.defaultPool.IPRestriction=x
- Comment out unused RebuildPeriod pool setting
- Add random key to pool in preparation for XOR peer ordering
(was Meta = PetName + HostsTxt)
* Naming: Add two new experimental NamingServices, EepGet and Exec,
not enabled by default -
see source comments in core/java/src/net/i2p/client/naming
for configuration instructions
* i2psnark: Remove orion and gaytorrents from default tracker list
* Remove orion from jump list and from eepsite_index.html
* Jbigi: Change jbigi version to 4.2.2 in build scripts - tested by amiga
* Capitalize OutboundMessageDistributor job name
* TunnelPool: Add a warning if all tunnels are backlogged
previous tunnel when no longer backlogged
* Catch an nio exception in an NTCP logging statement if loglevel is WARN
* IRC Proxy: terminate all messages with \r\n (thanks TrivialPursuit!)
participating tunnel count to seesaw -
should increase network capacity
* Leave participating tunnels in 10s batches for efficiency
* Update participating tunnel ratestat when leaving a tunnel too,
to generate a smoother graph
* Fix tunnel.participatingMessageCount stat to include all
participating tunnels, not just outbound endpoints
* Simplify Expire Tunnel job name
rather than 1 every 10 sec;
Don't store leasesets to disk or read them in
* Combine rates for pools with the same length setting
in the new tunnel build algorithm
* Clarify a log message in the UpdateHandler
* addressbook: Limit size of subscribed hosts.txt,
don't save old etag or last-modified data
* EepGet: Add some logging,
enforce size limits even when size not in returned header,
don't return old etag or last-modified data,
don't call transferFailed listener more than once
* profiles.jsp formatting cleanup
* NTCP: Reduce max idle time from 60m to 20m
* NTCP: Fix idle time on connections with zero messages,
correctly drop these connections
* Add stats.i2p to the jump list
* Impose 20MB limit on POSTs and catch OOMs in POST
* eepsite_index.html: add stats.i2p services
* addressbook: log source of new keys; disallow dests > 516 bytes
* addressbook: convert hostnames to lower case to prevent duplicates
* susidns: generalize references to orion
* Escape both CR, LF and CR LF line breaks in Router.saveConfig()
and unescape them in DataHelper.loadProps() to support
saving and loading config properties with line breaks
* Change the update URLs textbox into a textarea like keys have,
so different URLs go on different lines
* Modify TrustedUpdate to provide a method which supplies a key list
delimited with CR LF line breaks
* Modify DEFAULT_UPDATE_URL to supply a default URL list
delimited with CR LF line breaks
* Modify selectUpdateURL() to handle URL lists
delimited by any kind of line breaks
* Start saving trusted update keys
* Improve formatting on configupdate.jsp
* Add support for multiple update URLs
* Change default for update to use i2p proxy,
add several URLs as defaults
* Enable trusted key form on configupdate.jsp
* Separate the checks "does Jetty .zip file need downloading"
and "does Jetty .zip file need extracting" in the Jetty buildfile.
First download (unless already done), then extract (unless done).
* i2psnark: increase streaming lib write timeout to 240 sec and change
timeout action from "ping" to "disconect", as the fix in .30 to
honor options on outbound connections led to hung outbound connections
(bitfield never transmitted, connection never dropped)
* IRC Proxy: Fix several possible anonymity holes:
- Block CTCP in NOTICE messages
- Block CTCP anywhere in PRIVMSG and NOTICE, not just at first character
- Check for lower case commands
(Thanks sponge!)
* Implement pushback of NTCP transport backlog to the outbound tunnel selection code
* Clean up the NTCP and UDP tables on peers.jsp to be consistent,
fix some of the sorting
* eepsite_index.html: Add links to trevorreznik address book
* streaming lib: Fix SocketManagerFactory to honor options on outbound connections
* streaming lib: Fix setDefaultOptions() when called with a ConnectionOptions parameter
* i2psnark: Don't make outbound connections to already-connected peers
* i2psnark: Debug logging cleanup
control of timeouts and transparent redirection). the users of eepget
in this source tree don't necessarily use the timeout controls, though
they can be updated to do so
* Lower the threshold between the K and L bandwidth class,
so that K is now < 12 KB/s, instead of <= 16 KB/s.
Hopefully this lets people with 128 kbit/s (16 KB/s) upload lines
participate in routing, if they keep the default share percentage.
* Take the post-download routerInfo size check back out of ReseedHandler,
since it wasn't helpful, and a lower limit caused false warnings.
* Give EepGet ability to enforce a min/max HTTP response size.
* Enforce a maximum response size of 8 MB when ReseedHandler
downloads into a ByteArrayOutputStream.
* Refactor ReseedHandler/ReseedRunner from static to ordinary classes,
change invocation from RouterConsoleRunner accordingly.
* Add an EepGet status listener to ReseedHandler to log causes of reseed failure,
provide status reports to indicate the progress of reseeding.
* Enable icon for default eepsite, and the index page
of the router console (more later).
* Clean up graphs.jsp - set K=1024 where appropriate,
output image sizes in html, catch ooms, other minor tweaks
* Fix current event count truncation which fixes graphs with low
60-sec event counts displaying high values
(bw.* and router.* graphs for example were 1.5x too high)
Affects all "events per period" (non-lifetime) counts.
* Add auto-detect IP/Port to NTCP. When enabled on config.jsp,
SSU will notify/restart NTCP when the external address changes.
Now you can enable inbound TCP without a static IP or dyndns service.
* Replace broken option i2np.udp.alwaysPreferred with
i2np.udp.preferred and adjust UDP bids; possible settings are
"false" (default), "true", and "always".
Default setting results in same behavior as before
(NTCP is preferred unless it isn't established and UDP is established).
Use to compare NTCP and UDP transports.
* First pass on EepGet and ReseedHandler improvements,
please avoid use on routers which matter!
* Give EepGet ability of downloading into an OutputStream,
such as the ByteArrayOutputStream of ReseedHandler.
* Detect failure to reseed better, report it persistently
and more verbosely, provide a link to logs
and suggest manual reseed.
* i2psnark tracker handling tweaks:
- Add link to tracker details page (Postman only for now, requires bytemonsoon patch)
- Add Base URL to tracker list configuration
- Web page links built from tracker list Base URLs
- Only build and sort tracker list once
- Add anonymityWeb tracker to default list
- Add tooltip info for TrackerErrs
- Stop torrent if not registered with tracker
- Mark temp files as delete on exit
* i2psnark: Cleanup some handling of saved partial pieces
* i2psnark: Put bit counting in Bitfield.java for efficiency
* i2psnark: Save torrent completion state in i2psnark.config
* Streaming lib changes to improve upstream performance during congestion:
* Change min window size from 12 to 1
* Change max timeout from 10 to 45 sec
* Change initial timeout from 10 to 15 sec
* Change intial window size for i2psnark from 12 to 1
* Change slow start growth rate for i2psnark from 1/2 to 1
* Implement priority sending for NTCP
* Disable trimForOverload() in tunnel BuildExecutor which
was preventing tunnel builds when outbound traffic was high
(i.e. most of the time when running i2psnark)
* Tell our peers about who we know in the floodfill netDb every
6 hours or so, mitigating the situation where peers lose track
of floodfill routers.
* Disable the Syndie updater (people should use the new Syndie,
not this one)
* Disable the eepsite tunnel by default
* Modify ReseedHandler to query the "i2p.reseedURL" property from I2PAppContext
instead of System, so setting a reseed URL in advanced configuration has effect.
* Clean out obsolete reseed code from ConfigNetHandler.
* i2psnark: More choking rotation tweaks
* Improve performance by not reading in the whole
piece from disk for each request. A huge memory savings
on 1MB torrents with many peers.
* i2psnark: Add 'Stop All' link on web page
* Add some links to trackers and forum on web page
* Don't start tunnel if 'Autostart' unchecked
* Fix torrent restart bug by reopening file descriptors
* i2psnark: Improvements for torrents with > 4 leechers:
choke based on upload rate when seeding, and
be smarter and fairer about rotating choked peers.
* Handle two common i2psnark OOM situations rather
than shutting down the whole thing.
* Fix reporting to tracker of remaining bytes for
torrents > 4GB (but ByteMonsoon still has a bug)
* i2psnark: Fix and enable generation of multifile torrents,
print error if no tracker selected at create-torrent,
fix stopping a torrent that hasn't started successfully,
add eBook and GayTorrents trackers to form,
web page formatting tweaks
* Ensure we get NTP samples from more diverse sources
(0.pool.ntp.org, 1.pool.ntp.org, etc)
* Discard median-based peer skew calculator as framed average works,
and adjusting its percentage can make it behave median-like
* Require more data points (from at least 20 peers)
before considering a peer skew measurement reliable
2006-10-10 jrandom
* Removed the status display from the console, as its more confusing
than informative (though the content is still displayed in the HTML)
* Add a framed average peer clock skew calculator
* Add config property "router.clockOffsetSanityCheck" to determine
if NTP-suggested clock offsets get sanity checked (default "true")
* Reject NTP-suggested clock offsets if they'd increase peer clock skew
by more than 5 seconds, or make it more than 20 seconds total
* Decrease log level in getMedianPeerClockSkew()
* Subclass from Clock a RouterClock which can access router transports,
with the goal of developing it to second-guess NTP results
* Make transports report clock skew in seconds
* Adjust renderStatusHTML() methods accordingly
* Show average for NTCP clock skews too
* Give transports a getClockSkews() method to report clock skews
* Give transport manager a getClockSkews() method to aggregate results
* Give comm system facade a getMedianPeerClockSkew() method which RouterClock calls
(to observe results, add "net.i2p.router.transport.CommSystemFacadeImpl=WARN" to
logging)
* Extra explicitness in NTCP classes to denote unit of time.
* Fix some places in NTCPConnection where milliseconds and seconds were confused
* i2psnark: Paranoid copy before writing pieces,
recheck files on completion, redownload bad pieces
* i2psnark: Don't contact tracker as often when seeding
* i2psnark: Implement retransmission of requests. This
eliminates one cause of complete stalls with a peer.
This problem is common on torrents with a small number of
active peers where there are no choke/unchokes to kickstart things.
* i2psnark: Mark a peer's requests as unrequested on disconnect,
preventing premature end game
* i2psnark: Randomize selection of next piece during end game
* i2psnark: Don't restore a partial piece to a peer that is already working on it
* i2psnark: strip ".torrent" on web page
* i2psnark: Limit piece size in generated torrent to 1MB max
* i2psnark: Fix bug where new peers would always be set to "interested"
regardless of actual interest
* i2psnark: Reduce max piece size from 10MB to 1MB; larger may have severe
memory and efficiency problems
* Tweak the PRNG logging so it only displays error messages if there are
problems
* Disable dynamic router keys for the time being, as they don't offer
meaningful security, may hurt the router, and makes it harder to
determine the network health. The code to restart on SSU IP change is
still enabled however.
* Disable tunnel load testing, leaning back on the tiered selection for
the time being.
* Spattering of bugfixes
* i2psnark: Implement basic partial-piece saves across connections
* i2psnark: Implement keep-alive sending. This will keep non-i2psnark clients
from dropping us for inactivity but also renders the 2-minute transmit-inactivity
code in i2psnark ineffective. Will have to research why there is transmit but
not receive inactivity code. With the current connection limit of 24 peers
we aren't in any danger of keeping out new peers by keeping inactive ones.
* i2psnark: Increase CHECK_PERIOD from 20 to 40 since nothing happens in 20 seconds
* i2psnark: Fix dropped chunk handling
* i2psnark: Web rate report cleanup
* i2psnark: Report cleared trackerErr immediately
* i2psnark: Add trackerErr reporting after previous success; retry more quickly
* i2psnark: Set up new connections more quickly
* i2psnark: Don't delay tracker fetch when setting up lots of connections
* i2psnark: Reduce MAX_UPLOADERS from 12 to 4
* Limit form size in SusiDNS to avoid exceeding a POST size limit on postback
* Print messages about addressbook size to give better overview
* Enable delete function in published addressbook
* Increase the tunnel building timeout
* Avoid a rare race (thanks bar!)
* Fix the bandwidth capacity publishing code to factor in share percentage
and outbound throttling (oops)
* Actually fix the threading deadlock issue in the netDb (removing
the synchronized access to individual kbuckets while validating
individual entries) (thanks cervantes, postman, frosk, et al!)
* Cut down NTCP connection establishments once we know the peer is skewed
(rather than wait for full establishment before verifying)
* Removed a lock on the stats framework when accessing rates, which
shouldn't be a problem, assuming rates are created (pretty much) all at
once and merely updated during the lifetime of the jvm.
* When dropping a netDb router reference, only accept newer
references as part of the update check
* If we have been up for a while, don't accept really old
router references (published 2 or more days ago)
* Drop router references once they are no longer valid, even if
they were allowed in due to the lax restrictions on startup
* When dropping a netDb router reference, only accept newer
references as part of the update check
* If we have been up for a while, don't accept really old
router references (published 2 or more days ago)
* Drop router references once they are no longer valid, even if
they were allowed in due to the lax restrictions on startup
* Every time we create a new router identity, add an entry to the
new "identlog.txt" text file in the I2P install directory. For
debugging purposes, publish the count of how many identities the
router has cycled through, though not the identities itself.
* Cleaned up the way the multitransport shitlisting worked, and
added per-transport shitlists
* When dropping a router reference locally, first fire a netDb
lookup for the entry
* Take the peer selection filters into account when organizing the
profiles (thanks Complication!)
* Avoid some obvious configuration errors for the NTCP transport
(invalid ports, "null" ip, etc)
* Deal with some small NTCP bugs found in the wild (unresolveable
hosts, strange network discons, etc)
* Send our netDb info to peers we have direct NTCP connections to
after each 6-12 hours of connection uptime
* Clean up the NTCP reading and writing queue logic to avoid some
potential delays
* Allow people to specify the IP that the SSU transport binds on
locally, via the advanced config "i2np.udp.bindInterface=1.2.3.4"
2006-07-18 jrandom
* Add a failsafe to the NTCP transport to make sure we keep
pumping writes when we should.
* Properly reallow 16-32KBps routers in the default config
(thanks Complication!)
* Improve the multitransport shitlisting (thanks Complication!)
* Allow routers with a capacity of 16-32KBps to be used in tunnels under
the default configuration (thanks for the stats Complication!)
* Properly allow older router references to load on startup
(thanks bar, Complication, et al!)
* Add a new "i2p.alwaysAllowReseed" advanced config property, though
hopefully today's changes should make this unnecessary (thanks void!)
* Improved NTCP buffering
* Close NTCP connections if we are too backlogged when writing to them
* New NIO-based tcp transport (NTCP), enabled by default for outbound
connections only. Those who configure their NAT/firewall to allow
inbound connections and specify the external host and port
(dyndns/etc is ok) on /config.jsp can receive inbound connections.
SSU is still enabled for use by default for all users as a fallback.
* Substantial bugfix to the tunnel gateway processing to transfer
messages sequentially instead of interleaved
* Renamed GNU/crypto classes to avoid name clashes with kaffe and other
GNU/Classpath based JVMs
* Adjust the Fortuna PRNG's pooling system to reduce contention on
refill with a background thread to refill the output buffer
* Add per-transport support for the shitlist
* Add a new async pumped tunnel gateway to reduce tunnel dispatcher
contention
* Ensure that the I2PTunnel web interface won't update tunnel settings
for shared clients when a non-shared client is modified
(thanks for spotting, BarkerJr!)
* Cut down the proactive rejections due to queue size - if we are
at the point of having decrypted the request off the queue, might
as well let it through, rather than waste that decryption
* Only send netDb searches to the floodfill peers for the time being
* Add some proof of concept filters for tunnel participation. By default,
it will skip peers with an advertised bandwith of less than 32KBps or
an advertised uptime of less than 2 hours. If this is sufficient, a
safer implementation of these filters will be implemented.
* Fix some oversights in my previous changes:
adjust some loglevels, make a few statements less wasteful,
make one comparison less confusing and more likely to log unexpected values
* Separate growth factors for tunnel count and tunnel test time
* Reduce growth factors, so probabalistic throttle would activate
* Square probAccept values to decelerate stronger when far from average
* Create a bandwidth stat with approximately 15-second half life
* Make allowTunnel() check the 1-second bandwidth for overload
before doing allowance calculations using 15-second bandwidth
* Tweak the overload detector in BuildExecutor to be more sensitive
for rising edges, add ability to initiate tunnel drops
* Add a function to seek and drop the highest-rate participating tunnel,
keeping a fixed+random grace period between such drops.
It doesn't seem very effective, so disabled by default
("router.dropTunnelsOnOverload=true" to enable)
* Allow a single build attempt to proceed despite 1-minute overload
only if the 1-second rate shows enough spare bandwidth
(e.g. overload has already eased)
* Correct a misnamed property in SummaryHelper.java
to avoid confusion
* Make the maximum allowance of our own concurrent
tunnel builds slightly adaptive: one concurrent build per 6 KB/s
within the fixed range 2..10
* While overloaded, try to avoid completely choking our own build attempts,
instead prefer limiting them to 1
* Adjust the tunnel build timeouts to cut down on expirations, and
increased the SSU connection establishment retransmission rate to
something less glacial.
* For the first 5 minutes of uptime, be less aggressive with tunnel
exploration, opting for more reliable peers to start with.
* Adjust how we pick high capacity peers to allow the inclusion of fast
peers (the previous filter assumed an old usage pattern)
* New set of stats to help track per-packet-type bandwidth usage better
* Cut out the proactive tail drop from the SSU transport, for now
* Reduce the frequency of tunnel build attempts while we're saturated
* Don't drop tunnel requests as easily - prefer to explicitly reject them
* Adjust the proactive tunnel request dropping so we will reject what we
can instead of dropping so much (but still dropping if we get too far
overloaded)
* Throttling improvements on SSU - throttle all transmissions to a peer
when we are retransmitting, not just retransmissions. Also, if
we're already retransmitting to a peer, probabalistically tail drop new
messages targetting that peer, based on the estimated wait time before
transmission.
* Fixed the rounding error in the inbound tunnel drop probability.
* Include a combined send/receive graph (good idea cervantes!)
* Proactively drop inbound tunnel requests probabalistically as the
estimated queue time approaches our limit, rather than letting them all
through up to that limit.
* Process inbound tunnel requests more efficiently
* Proactively drop inbound tunnel requests if the queue before we'd
process it in is too long (dynamically adjusted by cpu load)
* Adjust the tunnel rejection throttle to reject requeusts when we have to
proactively drop too many requests.
* Display the number of pending inbound tunnel join requests on the router
console (as the "handle backlog")
* Include a few more stats in the default set of graphs
* Cut down on the time that we allow a tunnel creation request to sit by
without response, and reject tunnel creation requests that are lagged
locally. Also switch to a bounded FIFO instead of a LIFO
* Threading tweaks for the message handling (thanks bar!)
* Don't add addresses to syndie with blank names (thanks Complication!)
* Further ban clearance
* Fix during the ssu handshake to avoid an unnecessary failure on
packet retransmission (thanks ripple!)
* Fix during the SSU handshake to use the negotiated session key asap,
rather than using the intro key for more than we should (thanks ripple!)
* Fixes to the message reply registry (thanks Complication!)
* More comprehensive syndie banning (for repeated pushes)
* Publish the router's ballpark bandwidth limit (w/in a power of 2), for
testing purposes
* Put a floor back on the capacity threshold, so too many failing peers
won't cause us to pick very bad peers (unless we have very few good
ones)
* Bugfix to cut down on peers using introducers unneessarily (thanks
Complication!)
* Reduced the default streaming lib message size to fit into a single
tunnel message, rather than require 5 tunnel messages to be transferred
without loss before recomposition. This reduces throughput, but should
increase reliability, at least for the time being.
* Misc small bugfixes in the router (thanks all!)
* More tweaking for Syndie's CSS (thanks Doubtful Salmon!)
* Take out the router watchdog's teeth (don't restart on leaseset failure)
* Filter the IRC ping/pong messages, as some clients send unsafe
information in them (thanks aardvax and dust!)
* Substantially reduced the lock contention in the message registry (a
major hotspot that can choke most threads). Also reworked the locking
so we don't need per-message timer events
* No need to have additional per-peer message clearing, as they are
either unregistered individually or expired.
* Include some of the more transient tunnel throttling
* Added a simple purge and ban of syndie authors, shown as the
"Purge and ban" button on the addressbook for authors that are already
on the ignore list. All of their entries and metadata are deleted from
the archive, and the are transparently filtered from any remote
syndication (so no user on the syndie instance will pull any new posts
from them)
* More strict tunnel join throtting when congested
* Try to desync tunnel building near startup (thanks Complication!)
* If we are highly congested, fall back on only querying the floodfill
netDb peers, and only storing to those peers too
* Cleaned up the floodfill-only queries
* Avoid a very strange (unconfirmed) bug that people using the systray's
browser picker dialog could cause by disabling the GUI-based browser
picker.
* Cut down on subsequent streaming lib reset packets transmitted
* Use a larger MTU more often
* Allow netDb searches to query shitlisted peers, as the queries are
indirect.
* Add an option to disable non-floodfill netDb searches (non-floodfill
searches are used by default, but can be disabled by adding
netDb.floodfillOnly=true to the advanced config)
* Made the netDb search load limitations a little less stringent
* Add support for specifying the number of periods to be plotted on the
graphs - e.g. to plot only the last hour of a stat that is averaged at
the 60 second period, add &periodCount=60
* Add support for graphing the event count as well as the average stat
value (done by adding &showEvents=true to the URL). Also supports
hiding the legend (&hideLegend=true), the grid (&hideGrid=true), and
the title (&hideTitle=true).
* Removed an unnecessary arbitrary filter on the profile organizer so we
can pick high capacity and fast peers more appropriately
* Integrate basic hooks for jrobin (http://jrobin.org) into the router
console. Selected stats can be harvested automatically and fed into
in-memory RRD databases, and those databases can be served up either as
PNG images or as RRDtool compatible XML dumps (see oldstats.jsp for
details). A base set of stats are harvested by default, but an
alternate list can be specified by setting the 'stat.summaries' list on
the advanced config. For instance:
stat.summaries=bw.recvRate.60000,bw.sendRate.60000
* HTML tweaking for the general config page (thanks void!)
* Odd NPE fix (thanks Complication!)
* Integrate basic hooks for jrobin (http://jrobin.org) into the router
console. Selected stats can be harvested automatically and fed into
in-memory RRD databases, and those databases can be served up either as
PNG images or as RRDtool compatible XML dumps (see oldstats.jsp for
details). A base set of stats are harvested by default, but an
alternate list can be specified by setting the 'stat.summaries' list on
the advanced config. For instance:
stat.summaries=bw.recvRate.60000,bw.sendRate.60000
* HTML tweaking for the general config page (thanks void!)
* Odd NPE fix (thanks Complication!)
* Further stat cleanup
* Keep track of how many peers we are actively trying to communicate with,
beyond those who are just trying to communicate with us.
* Further router tunnel participation throttle revisions to avoid spurious
rejections
* Rate stat display cleanup (thanks ripple!)
* Don't even try to send messages that have been queued too long
* Further stat cleanup
* Keep track of how many peers we are actively trying to communicate with,
beyond those who are just trying to communicate with us.
* Further router tunnel participation throttle revisions to avoid spurious
rejections
* Rate stat display cleanup (thanks ripple!)
* Don't even try to send messages that have been queued too long
* HTML fixes in Syndie to work better with opera (thanks shaklen!)
* Give netDb lookups to floodfill peers more time, as they are much more
likely to succeed (thereby cutting down on the unnecessary netDb
searches outside the floodfill set)
* Fix to the SSU IP detection code so we won't use introducers when we
don't need them (thanks Complication!)
* Add a brief shitlist to i2psnark so it doesn't keep on trying to reach
peers given to it
* Don't let netDb searches wander across too many peers
* Don't use the 1s bandwidth usage in the tunnel participation throttle,
as its too volatile to have much meaning.
* Don't bork if a Syndie post is missing an entry.sml
* More aggressive tunnel throttling as we approach our bandwidth limit,
and throttle based off periods wider than 1 second.
* Included Doubtful Salmon's syndie stylings (thanks!)
2006-02-27 jrandom
* Adjust the jbigi.jar to use the athlon-optimized jbigi on windows/amd64
machines, rather than the generic jbigi (until we have an athlon64
optimized version)
2006-02-27 jrandom
* Adjust the jbigi.jar to use the athlon-optimized jbigi on windows/amd64
machines, rather than the generic jbigi (until we have an athlon64
optimized version)
* Switch from the bouncycastle to the gnu-crypto implementation for
SHA256, as benchmarks show a 10-30% speedup.
* Removed some unnecessary object caches
* Don't close i2psnark streams prematurely
* Made the Syndie permalinks in the thread view point to the blog view
* Disabled TCP again (since the live net seems to be doing well w/out it)
* Fix the message time on inbound SSU establishment (thanks zzz!)
* Don't be so aggressive with parallel tunnel creation when a tunnel pool
just starts up
* Rounding calculation cleanup in the stats, and avoid an uncontested
mutex (thanks ripple!)
* SSU handshake cleanup to help force incompatible peers to stop nagging
us by both not giving them an updated reference to us and by dropping
future handshake packets from them.
* Fix to properly profile tunnel joins (thanks Ragnarok, frosk, et al!)
* More aggressive poor-man's PMTU, allowing larger MTUs on less reliable
links
* Further class validator refactorings
* Fix to properly profile tunnel joins (thanks Ragnarok, frosk, et al!)
* More aggressive poor-man's PMTU, allowing larger MTUs on less reliable
links
* Further class validator refactorings
* Throttle the outbound SSU establishment queue, so it doesn't fill up the
heap when backlogged (and so that the messages queued up on it don't sit
there forever)
* Further SSU memory cleanup
* Clean up the address regeneration code so it knows when to rebuild the
local info more precisely.
* Throttle the outbound SSU establishment queue, so it doesn't fill up the
heap when backlogged (and so that the messages queued up on it don't sit
there forever)
* Further SSU memory cleanup
* Properly enable TCP this time (oops)
* Deal with multiple form handlers on the same page in the console without
being too annoying (thanks blubb and bd_!)
* Major SSU and router tuning to reduce contention, memory usage, and GC
churn. There are still issues to be worked out, but this should be a
substantial improvement.
* Modified the optional netDb harvester task to support choosing whether
to use (non-anonymous) direct connections or (anonymous) exploratory
tunnels to do the harvesting. Harvesting itself is enabled via the
advanced config "netDb.shouldHarvest=true" (default is false) and the
connection type can be chosen via "netDb.harvestDirectly=false" (default
is false).
* Major SSU and router tuning to reduce contention, memory usage, and GC
churn. There are still issues to be worked out, but this should be a
substantial improvement.
* Modified the optional netDb harvester task to support choosing whether
to use (non-anonymous) direct connections or (anonymous) exploratory
tunnels to do the harvesting. Harvesting itself is enabled via the
advanced config "netDb.shouldHarvest=true" (default is false) and the
connection type can be chosen via "netDb.harvestDirectly=false" (default
is false).
* Moved the current net's reseed URL to a different location than where
the old net looks (dev.i2p.net/i2pdb2/ vs .../i2pdb/)
* More aggressively expire inbound messages (on receive, not just on send)
* Add in a hook for breaking backwards compatibility in the SSU wire
protocol directly by including a version as part of the handshake. The
version is currently set to 0, however, so the wire protocol from this
build is compatible with all earlier SSU implementations.
* Increased the number of complete message readers, cutting down
substantially on the delay processing inbound messages.
* Delete the message history file on startup
* Reworked the restart/shutdown display on the console (thanks bd_!)
* Moved the current net's reseed URL to a different location than where
the old net looks (dev.i2p.net/i2pdb2/ vs .../i2pdb/)
* More aggressively expire inbound messages (on receive, not just on send)
* Add in a hook for breaking backwards compatibility in the SSU wire
protocol directly by including a version as part of the handshake. The
version is currently set to 0, however, so the wire protocol from this
build is compatible with all earlier SSU implementations.
* Increased the number of complete message readers, cutting down
substantially on the delay processing inbound messages.
* Delete the message history file on startup
* Reworked the restart/shutdown display on the console (thanks bd_!)
* Migrate the outbound packets from a central component to the individual
per-peer components, substantially cutting down on lock contention when
dealing with higher degrees.
* Load balance the outbound SSU transfers evenly across peers, rather than
across messages (so peers with few messages won't be starved by peers
with many).
* Reduce the frequency of router info rebuilds (thanks bar!)
* Migrate the outbound packets from a central component to the individual
per-peer components, substantially cutting down on lock contention when
dealing with higher degrees.
* Load balance the outbound SSU transfers evenly across peers, rather than
across messages (so peers with few messages won't be starved by peers
with many).
* Reduce the frequency of router info rebuilds (thanks bar!)
* Add a new AIMD throttle in SSU to control the number of concurrent
messages being sent to a given peer, in addition to the throttle on the
number of concurrent bytes to that peer.
* Adjust the existing SSU outbound queue to throttle based on the queue's
lag, not an arbitrary number of packets.
* Add a new AIMD throttle in SSU to control the number of concurrent
messages being sent to a given peer, in addition to the throttle on the
number of concurrent bytes to that peer.
* Adjust the existing SSU outbound queue to throttle based on the queue's
lag, not an arbitrary number of packets.
* Disable the message history log file by default (duh - feel free to
delete messageHistory.txt after upgrading. thanks deathfatty!)
* Limit the size of the inbound tunnel build request queue so we don't
get an insane backlog of requests that we're bound to reject, and adjust
the queue processing so we keep on churning through them when we've got
a backlog.
* Small fixes for the multiuser syndie operation (thanks Complication!)
* Renamed modified PRNG classes that were imported from gnu-crypto so we
don't conflict with JVMs using that as a JCE provider (thanks blx!)
* Disable the message history log file by default (duh - feel free to
delete messageHistory.txt after upgrading. thanks deathfatty!)
* Limit the size of the inbound tunnel build request queue so we don't
get an insane backlog of requests that we're bound to reject, and adjust
the queue processing so we keep on churning through them when we've got
a backlog.
* Small fixes for the multiuser syndie operation (thanks Complication!)
* Renamed modified PRNG classes that were imported from gnu-crypto so we
don't conflict with JVMs using that as a JCE provider (thanks blx!)
* Bugfix to the I2PTunnel web config to properly accept i2cp port settings
* Initial sucker refactoring to simplify reuse of the html parsing
* Beginnings of hooks to push imported rss/atom out to remote syndie
archives automatically (though not enabled currently)
* Further SSU peer test cleanup
* Bugfix to the I2PTunnel web config to properly accept i2cp port settings
* Initial sucker refactoring to simplify reuse of the html parsing
* Beginnings of hooks to push imported rss/atom out to remote syndie
archives automatically (though not enabled currently)
* Further SSU peer test cleanup
* Add in per-blog RSS feeds to Syndie
* Upgraded sucker's ROME dependency to 0.8, bundling sucked enclosures
with the posts, marking additional attachments as Media RSS enclosures
(http://search.yahoo.com/mrss/), since RSS only supports one enclosure
per item.
* Don't allow the default syndie user to be set to something invalid if
its in single user mode.
* Add in per-blog RSS feeds to Syndie
* Upgraded sucker's ROME dependency to 0.8, bundling sucked enclosures
with the posts, marking additional attachments as Media RSS enclosures
(http://search.yahoo.com/mrss/), since RSS only supports one enclosure
per item.
* Don't allow the default syndie user to be set to something invalid if
its in single user mode.
* Run the peer profile coalescing/reorganization outside the job queue
(on one of the timers), to cut down on some job queue congestion. Also,
trim old profiles while running, not just when starting up.
* Slightly more sane intra-floodfill-node netDb activity (only flood new
entries)
* Workaround in the I2PTunnelHTTPServer for some bad requests (though the
source of the bug is not yet addressed)
* Better I2PSnark reconnection handling
* Further cleanup in the new tunnel build process
* Make sure we expire old participants properly
* Remove much of the transient overload throttling (it wasn't using a good
metric)
* Run the peer profile coalescing/reorganization outside the job queue
(on one of the timers), to cut down on some job queue congestion. Also,
trim old profiles while running, not just when starting up.
* Slightly more sane intra-floodfill-node netDb activity (only flood new
entries)
* Workaround in the I2PTunnelHTTPServer for some bad requests (though the
source of the bug is not yet addressed)
* Better I2PSnark reconnection handling
* Further cleanup in the new tunnel build process
* Make sure we expire old participants properly
* Remove much of the transient overload throttling (it wasn't using a good
metric)
* Run the peer profile coalescing/reorganization outside the job queue
(on one of the timers), to cut down on some job queue congestion. Also,
trim old profiles while running, not just when starting up.
* Slightly more sane intra-floodfill-node netDb activity (only flood new
entries)
* Workaround in the I2PTunnelHTTPServer for some bad requests (though the
source of the bug is not yet addressed)
* Better I2PSnark reconnection handling
* Further cleanup in the new tunnel build process
* Make sure we expire old participants properly
* Remove much of the transient overload throttling (it wasn't using a good
metric)
* New tunnel build process - does not use the new crypto or new peer
selection strategies. However, it does drop the fallback tunnel
procedure, except for tunnels who are configured to allow them, or for
the exploratory pool during bootstrapping or after a catastrophic
failure. This new process prefers to fail rather than use too-short
tunnels, so while it can do some pretty aggressive tunnel rebuilding,
it may expose more tunnel failures to the user.
* Always prefer normal tunnels to fallback tunnels.
* Potential fix for a bug while changing i2cp settings on I2PSnark (thanks
bar!)
* Do all of the netDb entry writing in a separate thread, avoiding
duplicates and batching them up.
* Add title attributes to all external links in Syndie, so we can rollover
and quickly see if it's worth clicking on.
* Fixed a minor compiler warning.
* First pass of the new tunnel creation crypto, specified in the new
router/doc/tunnel-alt-creation.html (referenced in the current
router/doc/tunnel-alt.html). It isn't actually used anywhere yet, other
than in the test code, but the code verifies the technical viability, so
further scrutiny would be warranted.
* Only create the loadtest.log if requested to do so (thanks zzz!)
* Make sure we cleanly take into consideration the appropriate data
points when filtering out duplicate messages in the message validator,
and report the right bloom filter false positives rate (not used for
anything except debugging)
* Syndie CSS tweaks to removed some redundant declarations, improve font
scaling and layout robustness. Improved cross browser compatibility
(in other words "kicked IE"). Tightened the look of the blog template
a little.
* Include the attachments/blogs/etc for comments on the blog view
* Syndie HTML fixes (thanks cervantes!)
* Make sure we fully reset the objects going into our cache before we
reuse them (thanks zzz!)
* Added the per-post list of attachments/blogs/etc to the blog view in
Syndie (though this does not yet include comments or some further
refinements)
* Have the I2P shortcut launch i2p.exe instead of i2psvc.exe on windows,
removing the dox box (though also removes the restart functionality...)
* Give the i2p.exe the correct java.library.path to support the systray
dll (thanks Bobcat, Sugadude, anon!)
* Removed a longstanding bug that had caused unnecessary router identity
churn due to clock skew
* Temporarily sanity check within the streaming lib for long pending
writes
* Added support for a blog-wide logo to Syndie, and automated the pushing
of updated extended blog info data along side the metadata.
* Removed a longstanding bug that had caused unnecessary router identity
churn due to clock skew
* Temporarily sanity check within the streaming lib for long pending
writes
* Added support for a blog-wide logo to Syndie, and automated the pushing
of updated extended blog info data along side the metadata.
* Removed a longstanding bug that had caused unnecessary router identity
churn due to clock skew
* Temporarily sanity check within the streaming lib for long pending
writes
* Added support for a blog-wide logo to Syndie, and automated the pushing
of updated extended blog info data along side the metadata.
* Rather than profile individual tunnels for throughput over their
lifetime, do so at 1 minute intervals (allowing less frequently active
tunnels to be more fairly measured).
* Run the live tunnel load test across two tunnels at a time, by default.
The load test runs for a random period from 90s to the tunnel lifetime,
self paced. This should help gathering data for profiling peers that
are in exploratory tunnels.
2006-01-03 jrandom
* Calculate the overall peer throughput across the 3 fastest one minute
tunnel throughput values, rather than the single fastest throughput.
* Degrade the profiled throughput data over time (cutting the profiled
peaks in half once a day, on average)
* Enable yet another new speed calculation for profiling peers, using the
peak throughput from individual tunnels that a peer is participating in,
rather than across all tunnels they are participating in. This helps
gather a fairer peer throughput measurement, since it won't allow a slow
high capacity peer seem to have a higher throughput (pushing a little
data across many tunnels at once, as opposed to lots of data across a
single tunnel). This degrades over time like the other.
* Add basic OS/2 support to the jbigi code (though we do not bundle a
precompiled OS/2 library)
* Disable multifile torrent creation in I2PSnark's web UI for the moment
(though it can still seed and participate in multifile swarms)
* Enable a new speed calculation for profiling peers, using their peak
1 minute average tunnel throughput as their speed.
* Include a simple torrent creator in the I2PSnark web UI
* Further streaming lib closing improvements
* Refactored the load test components to run off live tunnels (though,
still not safe for normal/anonymous load testing)
* Replaced the bundled linux jcpuid (written in C++) with scintilla's
jcpuid (written in C), removing the libg++.so.5 dependency that has bit
some distros (e.g. mandriva)
* Add a new Status: line on the router console - "ERR-ClockSkew", in case
the clock is too skewed to do anything useful (check the year and month,
not just the hour and minute).
* Fixed the read/write timeouts in the streaming lib (so that it actually
honors them now)
* Minor I2PSnark cleanups (no read timeout, more careful shutdown and
torrent closing)
* Handle an oddball tunnel creation failure (thanks Xunk)
2005-12-22 jrandom
* Bundle the standalone I2PSnark launcher in the installer and update
process (launch as "java -jar launch-i2psnark.jar", viewing the
interface on http://localhost:8002/)
* Don't autostart swarming torrents by default so that you can run a
standalone I2PSnark from the I2P install dir and not have the embedded
I2PSnark autolaunch the torrents that the standalone instance is running
* Fixed a rare streaming lib bug that could let a blocking call wait
forever.
* Fix for old Syndie blog bookmarks (thanks Complication!)
* Fix for I2PSnark to accept incoming connections again (oops)
* Randomize the order that peers from the tracker are contacted
* I2PSnark logging, disconnect old inactive peers rather than new ones,
memory usage reduction, better OOM handling, and a shared connection
acceptor.
* Cleaned up the Syndie blog page and the resulting filters (viewing a
blog from the blog page shows threads started by the selected author,
not those that they merely participate in)
* Added a standalone runner for the I2PSnark web ui (build with the
command "ant i2psnark", unzip i2psnark-standalone.zip somewhere, run
with "java -jar launch-i2psnark.jar", and go to http://localhost:8002/).
* Further I2PSnark error handling
2005-12-17 jrandom
* Let multiuser accounts authorize themselves to access the remote
functionality again (thanks Ch0Hag!)
* Adjust the JVM heap size to 128MB for new installs (existing users can
accomplish this by editing wrapper.config, adding the line
"wrapper.java.maxmemory=128", and then doing a full shutdown and startup
of the router). This is relevent for heavy usage of I2PSnark in the
router console.
* Added a standalone runner for the I2PSnark web ui (build with the
command "ant i2psnark", unzip i2psnark-standalone.zip somewhere, run
with "java -jar launch-i2psnark.jar", and go to http://localhost:8002/).
* Further I2PSnark error handling
* Added some I2PSnark sanity checks, an OOMListener when running
standalone, and a guard against keeping memory tied up indefinitely.
* Sanity check on the watchdog (thanks zzz!)
* Handle invalid HTTP requests in I2PTunnel a little better
* Moved I2PSnark from using Threads to I2PThreads, so we handle OOMs
properly (thanks Complication!)
* More guards in I2PSnark for zany behavior (I2PSession recon w/ skew,
b0rking in the DirMonitor, etc)
* Added multitorrent support to I2PSnark, accessible currently by running
"i2psnark.jar --config i2psnark.config" (which may or may not exist).
It then joins the swarm for any torrents in ./i2psnark/*.torrent, saving
their data in that directory as well. Removing the .torrent file stops
participation, and it is currently set to seed indefinitely. Completion
is logged to the logger and standard output, with further UI interaction
left to the (work in progress) web UI.
* Fix to drop peer references when we shitlist people again (thanks zzz!)
* Further I2PSnark fixes to deal with arbitrary torrent info attributes
(thanks Complication!)
* Don't test tunnels expiring within 90 seconds
* Defer Test Tunnel jobs if job lag too large
* Use JobQueue.getMaxLag() rather than the jobQueue.jobLag stat to measure
job lag for tunnel build backoff, allowing for more agile handling
(since the stat is only updated once a minute)
* Use tunnel length override if all tunnels are expiring within one
minute.
* Fixed I2PSnark's handling of some torrent files to deal with those
created by Azureus and I2PRufus (it didn't know how to deal with
additional meta info, such as path.utf-8 or name.utf-8).
* Create different strategies for exploratory tunnels (which are difficult
to create) and client tunnels (which are much easier)
* Gradually increase number of parallel build attempts as tunnel expiry
nears.
* Temporarily shorten attempted build tunnel length if builds using
configured tunnel length are unsuccessful
* React more aggressively to tunnel failure than routine tunnel
replacement
* Make tunnel creation times randomized - there is existing code to
randomize the tunnels but it isn't effective due to the tunnel creation
strategy. Currently, most tunnels get built all at once, at about 2 1/2
to 3 minutes before expiration. The patch fixes this by fixing the
randomization, and by changing the overlap time (with old tunnels) to a
range of 2 to 4 minutes.
* Reduce number of excess tunnels. Lots of excess tunnels get created due
to overlapping calls. Just about anything generated a call which could
build many tunnels all at once, even if tunnel building was already in
process.
* Miscellaneous router console enhancements
* Added an RDF and XML thread export to Syndie, reachable at
.../threadnav/rdf or .../threadnav/xml, accepting the parameters
count=$numThreads and offset=$threadIndex. If the $numThreads is -1, it
displays all threads.
* Bugfix in Syndie for a problem in the threaded indexer (thanks CofE!)
* Always include ourselves in the favorite authors (since we don't
bookmark ourselves)
* Added support for a 'most recent posts' view that CofE requested, which
includes the ability to filter by age (e.g. posts by your favorite
authors in the last 5 days).
* Adjusted Syndie to use the threaded view that cervantes suggested, which
displays a a single thread path at a time - from root to leaf - rather
than a depth first traversal.
* Package up a standalone Syndie install into a "syndie-standalone.zip",
buildable with "ant syndie". It extracts into ./syndie/, launches with
"java -jar launchsyndie.jar" (or javaw, on windows, to avoid a dos box),
running a single user Syndie instance (by default). It also creates a
default subscription to syndiemedia without any anonymity (using no
proxy). Upgrades can be done by just replacing the syndie.war with the
one from I2P.
* Cleaned up the build process to deal with Jetty 5.1.6 and rename the
new commons-logging-api.jar to commons-logging.jar, which it replaces.
Jetty 5.1.6 is pushed with all updates. Also, no need to push a
separate jdom or rome, as they're inside syndie.war.
* Don't let the TCP transport alone shitlist a peer, since other
transports may be working. Also display whether TCP connections are
inbound or outbound on the peers page.
* Fixed some substantial bugs in the SSU introducers where we wouldn't
talk to anyone who didn't expose an IP (even if they had introducers),
among other goofy things.
* When dealing with SSU introducers, send them all a packet at 3s/6s/9s,
rather than sending one a packet at 3s, then another a packet at 6s,
and a third a packet at 9s.
* Fixed Syndie attachments (oops)
* Further Syndie UI cleanup
* Bundled our patched MultiPartRequest code from jetty (APL2 licensed),
since it hasn't been applied to the jetty CVS yet [1]. Its packaged
into syndie.jar and renamed to net.i2p.syndie.web.MultiPartRequest, but
will be removed as soon as its integrated into Jetty. This patch allows
posting content in various character sets.
[1] http://article.gmane.org/gmane.comp.java.jetty.general/6031
* Upgraded new installs to the latest stable jetty (5.1.6), though this
isn't pushed as part of the update yet, as there aren't any critical
bugs.
* Added back in the OSX jbigi, which was accidentally removed a few revs
back (thanks for the bug report stoerte!) New installs will get the
full jbigi, or you can pull the jbigi.jar from CVS by going to
http://dev.i2p.net/cgi-bin/cvsweb.cgi/i2p/installer/lib/jbigi/jbigi.jar
and clicking on the first "download" link, saving that jbigi.jar to
lib/jbigi.jar in your I2P installation directory. After restarting your
router, it should load up fine.
* Inlined the Syndie CSS to reduce the number of HTTP requests (and
because firefox [and others?] delay rendering until they fetch the css).
* Make sure we fire the shutdown tasks when regenerating a new identity
(thanks picsou!)
* Cleaned up some of the things I b0rked in the 'dynamic keys' mode
* Don't drop SSU sessions if they're still transmitting data successfully,
even if there are transmission failures
* Adjusted the time summarization to display hours after 119m, not 90m
* Further EepGet cleanup (grr)
* Update the sorting in Syndie to consider children 'newer' than parents,
even if they have the same message ID (duh)
* Cleaned up some nav links in Syndie (good idea gloin, spaetz!)
* Added a bunch of tooltips to Syndie's fields (thanks polecat!)
* Force support for nonvalidating XML in Jetty (so we can handle GCJ/etc
better)
* Be more explicit about what messages we will handle through a client
tunnel, and how we will handle them. This cuts off a set of attacks
that an active adversary could mount, though they're probably nonobvious
and would require at least some sophistication.
* Added support for 'dynamic keys' mode, where the router creates a new
router identity whenever it detects a substantial change in its public
address (read: SSU IP or port). This only offers minimal additional
protection against trivial attackers, but should provide functional
improvement for people who have periodic IP changes, since their new
router address would not be shitlisted while their old one would be.
* Added further infrastructure for restricted route operation, but its use
is not recommended.
* Don't publish stats for periods we haven't reached yet (thanks zzz!)
* Cleaned up the syndie threaded display to show the last updated date for
a subthread, and to highlight threads updated in the last two days.
* IE doesn't strip SPAN from <button> form fields, so add in a workaround
within I2PTunnel.
* Increase the maximum SSU retransmission timeout to accomodate slower or
more congested links (though SSU's RTO calculation will usually use a
much lower timeout)
* Moved the streaming lib timed events off the main timer queues and onto
a streaming lib specific set of timer queues. Streaming lib timed
events are more likely to have lock contention on the I2CP socket while
other timed events in the router are (largely) independent.
* Fixed a case sensitive lookup bug (thanks tino!)
* Syndie cleanup - new edit form on the preview page, and fixed some blog
links (thanks tino!)
* IE doesn't strip SPAN from <button> form fields, so add in a workaround
within I2PTunnel.
* Increase the maximum SSU retransmission timeout to accomodate slower or
more congested links (though SSU's RTO calculation will usually use a
much lower timeout)
* Moved the streaming lib timed events off the main timer queues and onto
a streaming lib specific set of timer queues. Streaming lib timed
events are more likely to have lock contention on the I2CP socket while
other timed events in the router are (largely) independent.
* Fixed a case sensitive lookup bug (thanks tino!)
* Syndie cleanup - new edit form on the preview page, and fixed some blog
links (thanks tino!)
* Implemented a trivial pure java PMTU backoff strategy, switching between
a 608 byte MTU and a 1350 byte MTU, depending upon retransmission rates.
* Fixed new user registration in Syndie (thanks Complication!)
* More aggressive I2PTunnel content encoding munging to work around some
rare HTTP behavior (ignoring q values on Accept-encoding, using gzip
even when only identity is specified, etc). I2PTunnelHTTPServer now
sends "Accept-encoding: \r\n" plus "X-Accept-encoding: x-i2p-gzip\r\n",
and I2PTunnelHTTPServer handles x-i2p-gzip in either the Accept-encoding
or X-Accept-encoding headers. Eepsite operators who do not know to
check for X-Accept-encoding will simply use the identity encoding.
* Add filtering threads by author to Syndie, populated with authors in the
user's addressbook
* When creating the default user, add
"http://syndiemedia.i2p/archive/archive.txt" to their addressbook,
configured to automatically pull updates. (what other archives should
be included?)
* Tiny servlet to help dole out the new routerconsole themes, and bundle
the installer/resources/themes/** into ./docs/themes/** on both install
and update.
* Initial pass of the routerconsole revamp, starting with I2PTunnel and
being progressively rolled out to other sections at later dates.
Featuring abstracted W3C strict XHTML1.0 markup, with CSS providing
layout and styling.
* Implemented console themes. Users can create their own themes by
creating css files in: {i2pdir}/docs/themes/console/{themename}/
and activating it using the routerconsole.theme={themename} advanced
config property. Look at the example incomplete "defCon1" theme.
Note: This is very much a work in progress. Folks might want to hold-off
creating their own skins until the markup has solidified.
* Added "routerconsole.javascript.disabled=true" to disable console
client-side scripting and "routerconsole.css.disabled=true" to remove
css styling (only rolled out in the i2ptunnel interface currently)
* Fixed long standing bug with i2ptunnel client and server edit screens
where tunnel count and depth properties would fail to save. Added
backup quantity and variance configuration options.
* Added basic accessibility support (key shortcuts, linear markup, alt and
title information and form labels).
* So far only tested on IE6, Firefox 1.0.6, Opera 8 and lynx.
* Initial pass of the routerconsole revamp, starting with I2PTunnel and
being progressively rolled out to other sections at later dates.
Featuring abstracted W3C strict XHTML1.0 markup, with CSS providing
layout and styling.
* Implemented console themes. Users can create their own themes by
creating css files in: {i2pdir}/docs/themes/console/{themename}/
and activating it using the routerconsole.theme={themename} advanced
config property. Look at the example incomplete "defCon1" theme.
Note: This is very much a work in progress. Folks might want to hold-off
creating their own skins until the markup has solidified.
* Added "routerconsole.javascript.disabled=true" to disable console
client-side scripting and "routerconsole.css.disabled=true" to remove
css styling (only rolled out in the i2ptunnel interface currently)
* Fixed long standing bug with i2ptunnel client and server edit screens
where tunnel count and depth properties would fail to save. Added
backup quantity and variance configuration options.
* Added basic accessibility support (key shortcuts, linear markup, alt and
title information and form labels).
* So far only tested on IE6, Firefox 1.0.6, Opera 8 and lynx.
* Default Syndie to single user mode, and automatically log into a default
user account (additional accounts can be logged into with the 'switch'
or login pages, and new accounts can be created with the register page).
* Disable the 'automated' column on the Syndie addressbook unless the user
is appropriately authorized (good idea Polecat!)
* Include SSU establishment failure in the peer profile as a commError,
as we do for TCP establishment failures.
* Don't throttle the initial transmission of a message because of ongoing
retransmissions to a peer, since the initial transmission of a message
is more valuable than a retransmission (since it has less latency).
* Cleaned up links to SusiDNS and I2PTunnel (thanks zzz!)
* Include the most recent ACKs with packets, rather than only sending an
ack exactly once. SSU differs from TCP in this regard, as TCP has ever
increasing sequence numbers, while each message ID in SSU is random, so
we don't get the benefit of later ACKs implicitly ACKing earlier
messages.
* Reduced the max retransmission timeout for SSU
* Don't try to send messages queued up for a long time waiting for
establishment.
* Include the most recent ACKs with packets, rather than only sending an
ack exactly once. SSU differs from TCP in this regard, as TCP has ever
increasing sequence numbers, while each message ID in SSU is random, so
we don't get the benefit of later ACKs implicitly ACKing earlier
messages.
* Reduced the max retransmission timeout for SSU
* Don't try to send messages queued up for a long time waiting for
establishment.
* Fix for some syndie reply scenarios (thanks identiguy and CofE!)
* Removed a potentially infinitely recursive call (oops)
(forgot to commit this file before. oops)
* Merge sucker into syndie with a rssimport.jsp page.
* Add getContentType() to EepGet.
* Make chunked transfer work (better) with EepGet.
* Do replaceAll("<","<") for logs.
* Improved the bandwidth throtting on tunnel participation, especially for
low bandwidth peers.
* Improved failure handling in SSU with proactive reestablishment of
failing idle peers, and rather than shitlisting a peer who failed too
much, drop the SSU session and allow a new attempt (which, if it fails,
will cause a shitlisting)
* Clarify the cause of the shitlist on the profiles page, and include
bandwidth limiter info at the bottom of the peers page.
* Improved the bandwidth throtting on tunnel participation, especially for
low bandwidth peers.
* Improved failure handling in SSU with proactive reestablishment of
failing idle peers, and rather than shitlisting a peer who failed too
much, drop the SSU session and allow a new attempt (which, if it fails,
will cause a shitlisting)
* Clarify the cause of the shitlist on the profiles page, and include
bandwidth limiter info at the bottom of the peers page.
* In Syndie, propogate the subject and tags in a reply, and show the parent
post on the edit page for easy quoting. (thanks identiguy and CofE!)
* Streamline some netDb query handling to run outside the jobqueue -
which means they'll run on the particular SSU thread that handles the
message. This should help out heavily loaded netDb peers.
* Defer netDb searches for newly referenced peers until we actually want
them
* Ignore netDb references to peers on our shitlist
* Set the timeout for end to end client messages to the max delay after
finding the leaseSet, so we don't have as many expired messages floating
around.
* Add a floor to the streaming lib window size
* When we need to send a streaming lib ACK, try to retransmit one of the
unacked packets instead (with updated ACK/NACK fields, of course). The
bandwidth cost of an unnecessary retransmission should be minor as
compared to both an ACK packet (rounded up to 1KB in the tunnels) and
the probability of a necessary retransmission.
* Adjust the streaming lib cwin algorithm to allow growth after a full
cwin messages if the rtt is trending downwards. If it is not, use the
existing algorithm.
* Increased the maximum rto size in the streaming lib.
* Load balancing bugfix on end to end messages to distribute across
tunnels more evenly.
* Defer netDb searches for newly referenced peers until we actually want
them
* Ignore netDb references to peers on our shitlist
* Set the timeout for end to end client messages to the max delay after
finding the leaseSet, so we don't have as many expired messages floating
around.
* Add a floor to the streaming lib window size
* When we need to send a streaming lib ACK, try to retransmit one of the
unacked packets instead (with updated ACK/NACK fields, of course). The
bandwidth cost of an unnecessary retransmission should be minor as
compared to both an ACK packet (rounded up to 1KB in the tunnels) and
the probability of a necessary retransmission.
* Adjust the streaming lib cwin algorithm to allow growth after a full
cwin messages if the rtt is trending downwards. If it is not, use the
existing algorithm.
* Increased the maximum rto size in the streaming lib.
* Load balancing bugfix on end to end messages to distribute across
tunnels more evenly.
* Add a blocking fetch to EepGetScheduler and RemoteArchiveBean and use them from the updater, to prevent race conditions with multiple archive fetches.
* Integrated GNU-Crypto's Fortuna PRNG, seeding it off /dev/urandom and
./prngseed.rnd (if they exist), and reseeding it with data out of
various crypto operations (unused bits in a DH exchange, intermediary
bits in a DSA signature generation, extra bits in an ElGamal decrypt).
The Fortuna implementation under gnu.crypto.prng has been modified to
use BouncyCastle's SHA256 and Cryptix's AES (since those are the ones
I2P uses), and the resulting gnu.crypto.prng.* are therefor available
under GPL+Classpath's linking exception (~= LGPL). I2P's SecureRandom
wrapper around it is, of course, public domain.
* Fix bug in ircclient that prevented it to use its own dest (i.e. was
always shared. (thx for info Ragnarok)
* Fix crash in Sucker with some bad html.
* Workaround a bug in GCJ's Calendar implementation
* Propery throw an exception in the streaming lib if we try to write to a
closed stream. This will hopefully help clear some I2Phex bugs (thanks
GregorK!)
* Ported the snark bittorrent client to I2P such that it is compatible
with i2p-bt and azneti2p. For usage information, grab an update and run
"java -jar lib/i2psnark.jar". It isn't currently multitorrent capable,
but adding in support would be fairly easy (see PeerAcceptor.java:49)
* Don't allow leaseSets expiring too far in the future (thanks postman)
libjbigi.so. Yeah, this means that i2psnark uses pure java modPow, but it doesn't do
any real heavy lifting anyway, except a DSA signature every 5-10 minutes. whoop de do.
* Ported the snark bittorrent client to I2P such that it is compatible
with i2p-bt and azneti2p. For usage information, grab an update and run
"java -jar lib/i2psnark.jar". It isn't currently multitorrent capable,
but adding in support would be fairly easy (see PeerAcceptor.java:49)
* Don't allow leaseSets expiring too far in the future (thanks postman)
The build in tracker has been removed for simplicity.
Example usage:
java -jar lib/i2psnark.jar myFile.torrent
or, a more verbose setting:
java -jar lib/i2psnark.jar --eepproxy 127.0.0.1 4444 \
--i2cp 127.0.0.1 7654 "inbound.length=2 outbound.length=2" \
--debug 6 myFile.torrent
* Bugfix for the auto-update code to handle different usage patterns
* Decreased the addressbook recheck frequency to once every 12 hours
instead of hourly.
* Handle dynamically changing the HMAC size (again, unless your nym is
toad or jrandom, ignore this ;)
* Cleaned up some synchronization/locking code
* Allow an env prop to configure whether we want to use the backwards
compatible (but not standards compliant) HMAC-MD5, or whether we want
to use the not-backwards compatible (but standards compliant) one. No
one should touch this setting, unless your name is toad or jrandom ;)
* Added some new dummy facades
* Be more aggressive on loading up the router.config before building the
router context
* Added new hooks for apps to deal with previously undefined I2NP message
types without having to modify any code.
* Demo code for using a castrated router for SSU comm (SSUDemo.java)
* Bundled dust's Sucker for pulling RSS/Atom content into SML, which can
then be injected into Syndie with the Syndie CLI.
* Bundled ROME and JDOM (BSD and Apache licensed, respectively) for
RSS/Atom parsing.
2005-10-13 jrandom
* SSU retransmission choke bugfix (== != !=)
* Include initial transmissions in the retransmission choke, so that
if we are already retransmitting a message, we won't send anything
to that peer other than that message (or ACKs, if necessary)
* SSU retransmission choke bugfix (== != !=)
* Include initial transmissions in the retransmission choke, so that
if we are already retransmitting a message, we won't send anything
to that peer other than that message (or ACKs, if necessary)
* Choke SSU retransmissions to a peer while there is already a
retransmission in flight to them. This currently lets other initial
transmissions through, since packet loss is often sporadic, but maybe
this should block initial transmissions as well?
* Display the retransmission bytes stat on peers.jsp (thanks bar!)
* Filter QUIT messages in the I2PTunnelIRCClient proxy
* Piggyback the SSU partial ACKs with data packets. This is backwards
compatible.
* Syndie RSS renderer bugfix, plus now include the full entry instead of
just the blurb before the cut.
* Piggyback the SSU explicit ACKs with data packets (partial ACKs aren't
yet piggybacked). This is backwards compatible.
* SML parser cleanup in Syndie
* Implemented a new I2PTunnelIRCClient which locally filters inbound and
outbound IRC commands for anonymity and security purposes, removing all
CTCP messages except ACTION, as well as stripping the hostname from the
USER message (while leaving the nick and 'full name'). The IRC proxy
doesn't use this by default, but you can enable it by creating a new
"IRC proxy" tunnel on the web interface, or by changing the tunnel type
to "ircclient" in i2ptunnel.config.
2005-10-10 jrandom
* I2PTunnel http client config cleanup and stats
* Minor SSU congestion tweaks and stats
* Reduced netDb exploration period
* Implemented a new I2PTunnelIRCClient which locally filters inbound and
outbound IRC commands for anonymity and security purposes, removing all
CTCP messages except ACTION, as well as stripping the hostname from the
USER message (while leaving the nick and 'full name'). The IRC proxy
doesn't use this by default, but you can enable it by creating a new
"IRC proxy" tunnel on the web interface, or by changing the tunnel type
to "ircclient" in i2ptunnel.config.
2005-10-10 jrandom
* I2PTunnel http client config cleanup and stats
* Minor SSU congestion tweaks and stats
* Reduced netDb exploration period
* Now that the streaming lib works reasonably, set the default inactivity
event to send a 0 byte keepalive payload, rather than disconnecting the
stream. This should cut the irc netsplits and help out with other long
lived streams. The default timeout is now less than the old timeout as
well, so the keepalive will be sent before earlier builds fire their
fatal timeouts.
2005-10-07 jrandom
* Include the 1 second bandwidth usage on the console rather than the
1 minute rate, as the 1 second value doesn't have the 1m/5m quantization
issues.
* Allow the I2PTunnelHTTPServer to send back the first few packets of an
HTTP response quicker, and initialize the streaming lib's cwin more
carefully.
* Added a small web UI to the new Syndie scheduled updater. If you log in
as a user authorized to use the remote archive funtionality, you can
request remote archives in your address book to be automatically pulled
down by checking the "scheduled?" checkbox.
* Allow the first few packets in the stream to fill in their IDs during
handshake (thanks cervantes, Complication, et al!) This should fix at
least some of the intermittent HTTP POST issues.
* Killed three more streaming lib bugs, one of which caused excess packets
to be transmitted (dupacking dupacks), one that was the root of many of
the old hung streams (shrinking highest received), and another that was
releasing data too soon.
* Only allow autodetection of our IP address if we haven't received an
inbound connection in the last two minutes.
* Increase the default max streaming resends to 8 from 5 (and down from
the earlier 10)
* Support noreseed.i2p in addition to .i2pnoreseed for disabling automatic
reseeding - useful on OSes that make it hard to create dot files.
Thanks Complication (and anon)!
* Fixed the installer version string (thanks Frontier!)
* Added cleaner rejection of invalid IP addresses, shitlist those who send
us invalid IP addresses, verify again that we are not sending invalid IP
addresses, and log an error if it happens. (Thanks Complication, ptm,
and adab!)
2005-09-29 jrandom
* Let syndie users modify their metadata.
* Reseed the router on startup if there aren't enough peer references
known locally. This can be disabled by creating the file .i2pnoreseed
in your home directory, and the existing detection and reseed handling
on the web interface is unchanged.
2005-09-29 jrandom
* Let syndie users modify their metadata.
* Reseed the router on startup if there aren't enough peer references
known locally. This can be disabled by creating the file .i2pnoreseed
in your home directory, and the existing detection and reseed handling
on the web interface is unchanged.
* I2PTunnel bugfix (thanks Complication!)
* Increase the SSU cwin slower during congestion avoidance (at k/cwin^2
instead of k/cwin)
* Limit the number of inbound SSU sessions being built at once (using
half of the i2np.udp.maxConcurrentEstablish config prop)
* Don't shitlist on a message send failure alone (unless there aren't any
common transports).
* More careful bandwidth bursting
* Reworded the SSU introductions config section (thanks duck!)
* Force identity content encoding for I2PTunnel httpserver requests
(thanks redzara!)
* Further x-i2p-gzip bugfixes for the end of streams
* Reduce the minimum bandwidth limits to 3KBps steady and burst (though
I2P's performance at 3KBps is another issue)
* Cleaned up some streaming lib structures
* Allow reseeding on the console if the netDb knows less than 30 peers,
rather than less than 10 (without internet connectivity, we keep the
last 15 router references)
* Reenable the x-i2p-gzip HTTP processing by default, flushing the stream
more aggressively.
* Show the status that used to be called "ERR-Reject" as "OK (NAT)"
* Reduced the default maximum number of streaming lib resends of a packet
(10 retransmits is a bit much with a reasonable RTO)
* Better i2paddresshelper handling in the I2PTunnel httpclient, plus a new
conflict resolution page if the i2paddresshelper parameter differs from
an existing name to destination mapping.
2005-09-25 jrandom
* Fix a long standing streaming lib bug (in the inactivity detection code)
* Improved handling of initial streaming lib packet retransmissions to
kill the "lost first packet" bug (where a page shows up with the first
few KB missing)
* Add support for initial window sizes greater than 1 - useful for
eepsites to transmit e.g. 4 packets full of data along with the initial
ACK, thereby cutting down on the rtt latency. The congestion window
size can and does still shrink down to 1 packet though.
* Adjusted the streaming lib retransmission calculation algorithm to be
more TCP-like.
* Bugfix in susidns for deleting entries
2005-09-21 jrandom
* Add support for HTTP POST to EepGet
* Use HTTP POST for syndie bulk fetches, since there's a lot of data to
put in that URL.
* Added support for pure 64bit linux with jbigi and the java service
wrapper (no need for jcpuid if we're on os.arch=amd64). Thanks mule
et al for help testing!
* UI cleanup in Syndie (thanks gloin and bar!)
2005-09-17 jrandom
* Clean up syndie a bit more and bundle a default introductory post with
both new installs and updates.
* Typo fixes on the console (thanks bar!)
* Updated the bandwidth limiter to use two tiers of bandwidth - our normal
steady state rate, plus a new limit on how fast we transfer when
bursting. This is different from the old "burst as fast as possible
until we're out of tokens" policy, and should help those with congested
networks. See /config.jsp to manage this rate.
* Bugfixes in Syndie to handle missing cache files (no data was lost, the
old posts just didn't show up).
* Log properly in EepPost
* Added the natively compiled jbigi and patched java service wrapper for
OS X. Thanks Bill Dorsey for letting me use your machine!
* Don't build i2p.exe or i2pinstall.exe when run on OS X machines, as we
don't bundle the binutils necessary (and there'd be a naming conflict
if we did).
* Added 'single user' functionality to syndie - if the single user
checkbox on the admin page is checked, all users are allowed to control
the instance and sync up with remote syndie nodes.
* Temporarily disable the x-i2p-gzip in i2ptunnel until it is more closely
debugged.
* Adjust I2PTunnelHTTPServer so it can be used for outproxy operators
(just specify the spoofed host as an empty string), allowing them to
honor x-i2p-gzip encoding.
* Let windows users build the exes too (thanks bar and redzara!)
* Allow I2PTunnel httpserver operators to disable gzip compression on
individual tunnels with the i2ptunnel.gzip=false client option
(good idea susi!)
* Added the i2p.exe and i2pinstall.exe for windows users, using launch4j.
* Added runplain.sh for *nix/osx users having problems using the java
service wrapper (called from the install dir as: sh runplain.sh)
* Bundle susidns and syndie, with links on the top nav
* Have I2PTunnelHTTPClient and I2PTunnelHTTPServer use the x-i2p-gzip
content-encoding (if offered), reducing the payload size before it
reaches the streaming lib. The existing compression is at the i2cp
level, so we've been packetizing 4KB of uncompressed data and then
compressing those messages, rather than compressing and then packetizing
4KB of compressed data. This should reduce the number of round trips
to fetch web pages substantially.
* Adjust the startup and timing of the addressbook so that susidns always
has config to work off, and expose a method for susidns to tell it to
reload its config and rerun.
* More careful error handling with introductions (thanks dust!)
* Fix the forceIntroducers checkbox on config.jsp (thanks Complication!)
* Hide the shitlist on the summary so it doesn't confuse new users.
* Removed guaranteed delivery mode entirely (so existing i2phex clients
using it can get the benefits of mode=best_effort). Guaranteed delivery
is offered at the streaming lib level.
* Improve the peer selection code for peer testing, as everyone now
supports tests.
* Give the watchdog its fangs - if it detects obscene job lag or if
clients have been unable to get a leaseSet for more than 5 minutes,
restart the router. This was disabled a year ago due to spurious
restarts, and can be disabled by "watchdog.haltOnHang=false", but the
cause of the spurious restarts should be gone.
* More aggressively publish updated routerInfo.
* Expose the flag to force SSU introductions on the router console
* Don't give people the option to disable SNTP time sync, at least not
through the router console, because there is no reason to disable it.
No, not even if your OS is "ntp synced", because chances are, its not.
* Test the router's reachability earlier and more aggressively
* Use the low level bandwidth limiter's rates for the router console, and
if the router has net.i2p.router.transport.FIFOBandwidthLimiter=INFO in
the logger config, keep track of the 1 second transfer rates as the stat
'bw.sendBps1s' and 'bw.recvBps1s', allowing closer monitoring of burst
behavior.
* Added preliminary support for NAT hole punching through SSU introducers
* Honor peer test results from peers that we have an SSU session with if
those sessions are idle for 3 minutes or more.
* HTML cleanup for the router console (thanks!)
2005-09-07 jrandom
* Lay the foundation for 'client routers' - the ability for peers to opt
out of participating in tunnels entirely due to firewall/NAT issues.
Individual routers have control over where those peers are used in
tunnels - in outbound or inbound, exploratory or client tunnels, or
none at all. The defaults with this build are to simply act as before -
placing everyone as potential participants in any tunnel.
* Another part of the foundation includes the option for netDb
participants to refuse to answer queries regarding peers who are marked
as unreachable, though this too is disabled by default (meaning the
routerInfo is retrievable from the netDb).
* Expose the HTTP headers to EepGet status listeners
* Handle DSA key failures properly (if the signature is not invertable, it
is obviously invalid)
also, syndie now properly detects whether the remote archive can send a filtered export.zip
by examining the HTTP headers for X-Syndie-Export-Capable: true. If the remote archive
does not set that header (and neither freesites, nor apache or anything other than the ArchiveServlet will),
it uses individual HTTP requests for individual blog posts and metadata fetches.
rss.jsp can in turn receive all the filters that index.jsp can - e.g. ?blog=blah or ?selector=group://foo,
and by default returns the latest 10 values (overridden with ?wanted=15). If you want it to pull
with a user's blog's preferences (filters, groups, etc), you can specify ?login=user&password=password
* Don't persist peer profiles until we are shutting down, as the
persistence process gobbles RAM and wall time.
* Bugfix to allow you to check/uncheck the sharedClient setting on the
I2PTunnel web interface.
* Be more careful when expiring a failed tunnel message fragment so we
don't drop the data while attempting to read it.
* Don't publish leaseSets to the netDb if they will never be looked for -
namely, if they are for destinations that only establish outbound
streams. I2PTunnel's 'client' and 'httpclient' proxies have been
modified to tell the router that it doesn't need to publish their
leaseSet (by setting the I2CP config option 'i2cp.dontPublishLeaseSet'
to 'true').
* Don't publish the top 10 peer rankings of each router in the netdb, as
it isn't being watched right now.
* Don't publish leaseSets to the netDb if they will never be looked for -
namely, if they are for destinations that only establish outbound
streams. I2PTunnel's 'client' and 'httpclient' proxies have been
modified to tell the router that it doesn't need to publish their
leaseSet (by setting the I2CP config option 'i2cp.dontPublishLeaseSet'
to 'true').
* Don't publish the top 10 peer rankings of each router in the netdb, as
it isn't being watched right now.
* Minor logging and optimization tweaks in the router and SDK
* Use ISO-8859-1 in the XML files (thanks redzara!)
* The consolePassword config property can now be used to bypass the router
console's nonce checking, allowing CLI restarts
* Minor logging and optimization tweaks in the router and SDK
* Use ISO-8859-1 in the XML files (thanks redzara!)
* The consolePassword config property can now be used to bypass the router
console's nonce checking, allowing CLI restarts
* Catch errors with corrupt tunnel messages more gracefully (no need to
kill the thread and cause an OOM...)
* Don't skip shitlisted peers for netDb store messages, as they aren't
necessarily shitlisted by other people (though they probably are).
* Adjust the netDb store per-peer timeout based on each particular peer's
profile (timeout = 4x their average netDb store response time)
* Don't republish leaseSets to *failed* peers - send them to peers who
replied but just didn't know the value.
* Set a 5 second timeout on the I2PTunnelHTTPServer reading the client's
HTTP headers, rather than blocking indefinitely. HTTP headers should be
sent entirely within the first streaming packet anyway, so this won't be
a problem.
* Don't use the I2PTunnel*Server handler thread pool by default, as it may
prevent any clients from accessing the server if the handlers get
blocked by the streaming lib or other issues.
* Don't overwrite a known status (OK/ERR-Reject/ERR-SymmetricNAT) with
Unknown.
* Removed the concept of "no bandwidth limit" - if none is specified, its
16KBps in/out.
* Include ack packets in the per-peer cwin throttle (they were part of the
bandwidth limit though).
* Tweak the SSU cwin operation to get more accurrate estimates under
congestions.
* SSU improvements to resend more efficiently.
* Added a basic scheduler to eepget to fetch multiple files sequentially.
* Removed the concept of "no bandwidth limit" - if none is specified, its
16KBps in/out.
* Include ack packets in the per-peer cwin throttle (they were part of the
bandwidth limit though).
* Tweak the SSU cwin operation to get more accurrate estimates under
congestions.
* SSU improvements to resend more efficiently.
* Added a basic scheduler to eepget to fetch multiple files sequentially.
* Revise the SSU peer testing protocol so that Bob verifies Charlie's
viability before agreeing to Alice's request. This doesn't work with
older SSU peer test builds, but is backwards compatible (older nodes
won't ask newer nodes to participate in tests, and newer nodes won't
ask older nodes to either).
* Revise the SSU peer testing protocol so that Bob verifies Charlie's
viability before agreeing to Alice's request. This doesn't work with
older SSU peer test builds, but is backwards compatible (older nodes
won't ask newer nodes to participate in tests, and newer nodes won't
ask older nodes to either).
* Keep detailed stats on the peer testing, publishing the results in the
netDb.
* Don't overwrite the status with 'unknown' unless we haven't had a valid
status in a while.
* Make sure to avoid shitlisted peers for peer testing.
* When we get an unknown result to a peer test, try again soon afterwards.
* When a peer tells us that our address is different from what we expect,
if we've done a recent peer test with a result of OK, fire off a peer
test to make sure our IP/port is still valid. If our test is old or the
result was not OK, accept their suggestion, but queue up a peer test for
later.
* Don't try to do a netDb store to a shitlisted peer, and adjust the way
we monitor netDb store progress (to clear up the high netDb.storePeers
stat)
* Deployed the peer testing implementation to be run every few minutes on
each router, as well as any time the user requests a test manually. The
tests do not reconfigure the ports at the moment, merely determine under
what conditions the local router is reachable. The status shown in the
top left will be "ERR-SymmetricNAT" if the user's IP and port show up
differently for different peers, "ERR-Reject" if the router cannot
receive unsolicited packets or the peer helping test could not find a
collaborator, "Unknown" if the test has not been run or the test
participants were unreachable, or "OK" if the router can receive
unsolicited connections and those connections use the same IP and port.
state = 2 means all three handlers are blocking on udpReceiver.receive())
this can legitimately happen if the bandwidth limiter or router throttle chokes the receive for >= 1s.
* Add a configurable throttle to the number of concurrent outbound SSU
connection negotiations (via i2np.udp.maxConcurrentEstablish=4). This
may help those with slow connections to get integrated at the start.
* Further fixlets to the streaming lib
* Display the average clock skew for both SSU and TCP connections
2005-08-07 jrandom
* Fixed the long standing streaming lib bug where we could lose the first
packet on retransmission.
* Avoid an NPE when a message expires on the SSU queue.
* Adjust the streaming lib's window growth factor with an additional
Vegas-esque congestion detection algorithm.
* Removed an unnecessary SSU session drop
* Reduced the MTU (until we get a working PMTU lib)
* Deferr tunnel acceptance until we know how to reach the next hop,
rejecting it if we can't find them in time.
* If our netDb store of our leaseSet fails, give it a few seconds before
republishing.
* Fix an addressbook NPE when a new hostname from the master addressbook
didn't exist in the router addressbook.
* Fix an addressbook bug which caused subscriptions not to be parsed at
all. (Oops!)
* Fix an addressbook NPE when a new hostname from the master addressbook
didn't exist in the router addressbook.
* Fix an addressbook bug which caused subscriptions not to be parsed at
all. (Oops!)
* Adjust the netDb search and store per peer timeouts to match the average
measured per peer success times, rather than huge fixed values.
* Optimized and reverified the netDb peer selection / retrieval process
within the kbuckets.
* Drop TCP connections that don't have any useful activity in 10 minutes.
* If i2np.udp.fixedPort=true, never change the externally published port,
even if we are autodetecting the IP address.
(also includes most of the new peer/NAT testing, but thats not used atm)
2005-07-27 jrandom
* Enabled SSU as the default top priority transport, adjusting the
config.jsp page accordingly.
* Add verification fields to the SSU and TCP connection negotiation (not
compatible with previous builds)
* Enable the backwards incompatible tunnel crypto change as documented in
tunnel-alt.html (have each hop encrypt the received IV before using it,
then encrypt it again before sending it on)
* Disable the I2CP encryption, leaving in place the end to end garlic
encryption (another backwards incompatible change)
* Adjust the protocol versions on the TCP and SSU transports so that they
won't talk to older routers.
* Fix up the config stats handling again
* Fix a rare off-by-one in the SSU fragmentation
* Reduce some unnecessary netDb resending by inluding the peers queried
successfully in the store redundancy count.
* Fix in the SDK for a bug which would manifest itself as misrouted
streaming packets when a destination has many concurrent streaming
connections (thanks duck!)
* No more "Graceful shutdown in -18140121441141s"
* Further preparation for removing I2CP crypto
* Added some validation to the DH key agreement (thanks $anon)
* Validate tunnel data message expirations (though not really a problem,
since tunnels expire)
* Minor PRNG threading cleanup
* Added workaround for an odd win32 bug in the stats configuration
console page which meant only the first checkbox selection was saved.
2005-07-15 Romster
* Added per group selection toggles in the stats configuration console
page.
* Fixed a recently injected bug in the multitransport bidding which had
allowed an essentially arbitrary choice of transports, rather than the
properly ordered choice.
(getLatency() != getLatencyMs(). duh)
* Fixed a long standing bug where we weren't properly comparing session
tags but instead largely depending upon comparing their hashCode,
causing intermittent decryption errors.
* Reduced the growth factor on the slow start and congestion avoidance for
the streaming lib.
* Adjusted some of the I2PTunnelServer threading to use a small pool of
handlers, rather than launching off new threads which then immediately
launch off an I2PTunnelRunner instance (which launches 3 more threads..)
* Don't persist session keys / session tags (not worth it, for now)
* Added some detection and handling code for duplicate session tags being
delivered (root cause still not addressed)
* Make the PRNG's buffer size configurable (via the config property
"i2p.prng.totalBufferSizeKB=4096")
* Disable SSU flooding by default (duh)
* Updates to the StreamSink apps for better throttling tests.
* Use a buffered PRNG, pulling the PRNG data off a larger precalculated
buffer, rather than the underlying PRNG's (likely small) one, which in
turn reduces the frequency of recalcing.
* More tuning to reduce temporary allocation churn
* Within the tunnel, use xor(IV, msg[0:16]) as the flag to detect dups,
rather than the IV by itself, preventing an attack that would let
colluding internal adversaries tag a message to determine that they are
in the same tunnel. Thanks dvorak for the catch!
* Drop long inactive profiles on startup and shutdown
* /configstats.jsp: web interface to pick what stats to log
* Deliver more session tags to account for wider window sizes
* Cache some intermediate values in our HMACSHA256 and BC's HMAC
* Track the client send rate (stream.sendBps and client.sendBpsRaw)
* UrlLauncher: adjust the browser selection order
* I2PAppContext: hooks for dummy HMACSHA256 and a weak PRNG
* StreamSinkClient: add support for sending an unlimited amount of data
* Migrate the tests out of the default build jars
2005-06-22 Comwiz
* Migrate the core tests to junit
* Phase 1 of the unit test bounty completed. (The router build script was modified not to build the router
tests because of a broken dependancy on the core tests. This should be fixed in
phase 3 of the unit test bounty.)
* increased the maximum number of fragments allowed in a message from 31 to 127,
reducing the maximum fragment size to 8KB and moving around some bits in the fragment
info. This is not backwards compatible.
* removed the old (hokey) congestion control description, replacing it with the TCP-esque
algorithm implemented
note: the code for the ACK bitfields and fragment info changes have not yet been
implemented, so the old version of this document describes whats going on in the live net.
the new bitfields / fragment info should be deployed in the next day or so (hopefully :)
* Cleaned up the peers page a bit more.
more udp stuff:
* add new config option: i2np.udp.alwaysPreferred=true to adjust the bidding
so that UDP is picked first, even if a TCP connection exists
* fixed the initial clock skew problem (duh)
* reduced the MTU to 576 (largest nearly-universally-safe, and allows a
tunnel message in 2 fragments)
* handle some races @ connection establishment (thanks duck!)
* if there are more ACKs than we can send in a packet, reschedule another
ACK immediately
* Added a small new page to the web console (/peers.jsp) which contains
the peer connection information. This will be cleaned up a lot more
before 0.6 is out, but its a start.
* Reduced some SimpleTimer churn
* add hooks for per-peer choking in the outbound message queue - if/when a
peer reaches their cwin, no further messages will enter the 'active' pool
until there are more bytes available. other messages waiting (either later
on in the same priority queue, or in the queues for other priorities) may
take that slot.
* when we have a message acked, release the acked size to the congestion
window (duh), rather than waiting for the second to expire and refill the
capacity.
* send packets in a volley explicitly, waiting until we can allocate the full
cwin size for that message
* more stats. including per-peer KBps (updated every second)
* improved blocking/timeout situations on the send queue
* added drop simulation hook
* provide logical RTO limits
* Reduce the peer profile stat coallesce overhead by inlining it with the
reorganize.
* Limit each transport to at most one address (any transport that requires
multiple entry points can include those alternatives in the address).
udp stuff:
* change the UDP transport's style from "udp" to "SSUv1"
* keep track of each peer's skew
* properly handle session reestablishment over an existing session, rather
than requiring both sides to expire first
* More fixes for the I2PTunnel "other" interface handling (thanks nelgin!)
* Add back the code to handle bids from multiple transports (though there
is still only one transport enabled by default)
* Adjust the router's queueing of outbound client messages when under
heavy load by running the preparatory job in the client's I2CP handler
thread, thereby blocking additional outbound messages when the router is
hosed.
* No need to validate or persist a netDb entry if we already have it
And for some udp stuff:
* only bid on what we know (duh)
* reduceed the queue size in the UDPSender itself, so that ACKs go
through more quickly, leaving the payload messages to queue up in
the outbound fragment scheduler
* rather than /= 2 on congestion, /= 2/3 (still AIMD, but less drastic)
* adjust the fragment selector so a wsiz throttle won't force extra
volleys
* mark congestion when it occurs, not after the message has been
ACKed
* when doing a round robin over the active messages, move on to the
next after a full volley, not after each packet (causing less "fair"
performance but better latency)
* reduced the lock contention in the inboundMessageFragments by
moving the ack and complete queues to the ACKSender and
MessageReceiver respectively (each of which have their own
threads)
* prefer new and existing UDP sessions to new TCP sessions, but
prefer existing TCP sessions to new UDP sessions
* Added button to router console for manual update checks.
* Fixed bug in configupdate.jsp that caused the proxy port to be updated
every time the form was submitted even if it hadn't changed.
* Added a pool of PRNGs using a different synchronization technique,
hopefully sufficient to work around IBM's PRNG bugs until we get our
own Fortuna.
* In the streaming lib, don't jack up the RTT on NACK, and have the window
size bound the not-yet-ready messages to the peer, not the unacked
message count (not sure yet whether this is worthwile).
* Many additions to the messageHistory log.
* Handle out of order tunnel fragment delivery (not an issue on the live
net with TCP, but critical with UDP).
* Added a pool of PRNGs using a different synchronization technique,
hopefully sufficient to work around IBM's PRNG bugs until we get our
own Fortuna.
* In the streaming lib, don't jack up the RTT on NACK, and have the window
size bound the not-yet-ready messages to the peer, not the unacked
message count (not sure yet whether this is worthwile).
* Many additions to the messageHistory log.
* Handle out of order tunnel fragment delivery (not an issue on the live
net with TCP, but critical with UDP).
and for udp stuff:
* implemented tcp-esque rto code in the udp transport
* make sure we don't ACK too many messages at once
* transmit fragments in a simple (nonrandom) order so that we can more easily
adjust timeouts/etc.
* let the active outbound pool grow dynamically if there are outbound slots to
spare
* use a simple decaying bloom filter at the UDP level to drop duplicate resent
packets.
And this is the big "Fix the Parser" patch. It turns the sam_parse function in src/parse.c into something that actually works. Generating the argument list from an incoming SAM thingy is a bit memory churn-y; perhaps when I have time I'll replace all those strdups with structures that simply track the (start,end) indices.
Oh and also I moved i2p-ping to the new system. Which required 0 change in code. All I did was fix the Makefile, and add shared library libtool support. Anyway, so enjoy folks. It's rare I'm this productive
- polecat
* In the SDK, we don't actually need to block when we're sending a message
as BestEffort (and these days, we're always sending BestEffort).
* Pass out client messages in fewer (larger) steps.
* Have the InNetMessagePool short circuit dispatch requests.
* Have the message validator take into account expiration to cut down on
false positives at high transfer rates.
* Allow configuration of the probabalistic window size growth rate in the
streaming lib's slow start and congestion avoidance phases, and default
them to a more conservative value (2), rather than the previous value
(1).
* Reduce the ack delay in the streaming lib to 500ms
* Honor choke requests in the streaming lib (only affects those getting
insanely high transfer rates)
* Let the user specify an interface besides 127.0.0.1 or 0.0.0.0 on the
I2PTunnel client page (thanks maestro^!)
(plus minor udp tweaks)
* Added the possibility for i2ptunnel client and httpclient instances to
have their own i2p session (and hence, destination and tunnels). By
default, tunnels are shared, but that can be changed on the web
interface or with the sharedClient config option in i2ptunnel.config.
2005-04-17 jrandom
* Marked the net.i2p.i2ptunnel.TunnelManager as deprecated. Anyone use
this? If not, I want to drop it (lots of tiny details with lots of
duplicated semantics).
* Added new user-editable eepproxy error page templates.
2005-04-17 jrandom
* Revamp the tunnel building throttles, fixing a situation where the
rebuild may not recover, and defaulting it to unthrottled (users with
slow CPUs may want to set "router.tunnel.shouldThrottle=true" in their
advanced router config)
* use the new raw i2np message format (the previous corruptions were due to above)
* add a new test component (UDPFlooder) which floods all peers at the rate desired
* packet munging fix for highly fragmented messages
* include basic slow start code
* fixed the UDP peer rate refilling
* cleaned up some nextSend scheduling
* Make sure we don't get cached updates (thanks smeghead!)
* Clear out the callback for the TestJob after it passes (only affects the
job timing accounting)
* include a new signedOnTime so that we can prepare the packet at a different moment from
when we encrypt & send it (also allowing us to reuse that signature on resends for the same
establishment)
* Security improvements to TrustedUpdate: signing and verification of the
version string along with the data payload for signed update files
(consequently the positions of the DSA signature and version string fields
have been swapped in the spec for the update file's header); router will
no longer perform a trusted update if the signed update's version is lower
than or equal to the currently running router's version.
* Added two new CLI commands to TrustedUpdate: showversion, verifyupdate.
* Extended TrustedUpdate public API for use by third party applications.
* Retry I2PTunnel startup if we are unable to build a socketManager for a
client or httpclient tunnel.
* Add some basic sanity checking on the I2CP settings (thanks duck!)
* After a successfull netDb search for a leaseSet, republish it to all of
the peers we have tried so far who did not give us the key (up to 10),
rather than the old K closest (which may include peers who had given us
the key)
* Don't wait 5 minutes to publish a leaseSet (duh!), and rather than
republish it every 5 minutes, republish it every 3. In addition, always
republish as soon as the leaseSet changes (duh^2).
* Minor fix for oddball startup race (thanks travis_bickle!)
* Minor AES update to allow in-place decryption.
SigningPublicKey, which take a single String argument and construct
the object from the Base64 data in that string (where this data is
the product of a .toBase64() call on a prior instance).
2005-03-29 jrandom
* Decreased the initial RTT estimate to 10s to allow more retries.
* Increased the default netDb store replication factor from 2 to 6 to take
into consideration tunnel failures.
* Address some statistical anonymity attacks against the netDb that could
be mounted by an active internal adversary by only answering lookups for
leaseSets we received through an unsolicited store.
* Don't throttle lookup responses (we throttle enough elsewhere)
* Fix the NewsFetcher so that it doesn't incorrectly resume midway through
the file (thanks nickster!)
* Updated the I2PTunnel HTML (thanks postman!)
* Added support to the I2PTunnel pages for the URL parameter "passphrase",
which, if matched against the router.config "i2ptunnel.passphrase" value,
skips the nonce check. If the config prop doesn't exist or is blank, no
passphrase is accepted.
* Implemented HMAC-SHA256.
* Enable the tunnel batching with a 500ms delay by default
* Dropped compatability with 0.5.0.3 and earlier releases
* Implemented the news fetch / update policy code, as configurated on
/configupdate.jsp. Defaults are to grab the news every 24h (or if it
doesn't exist yet, on startup). No action is taken however, though if
the news.xml specifies that a new release is available, an option to
update will be shown on the router console.
* New initialNews.xml delivered with new installs, and moved news.xml out
of the i2pwww module and into the i2p module so that we can bundle it
within each update.
* New /configupdate.jsp page for controlling the update / notification
process, as well as various minor related updates. Note that not all
options are exposed yet, and the update detection code isn't in place
in this commit - it currently says there is always an update available.
* New EepGet component for reliable downloading, with a CLI exposed in
java -cp lib/i2p.jar net.i2p.util.EepGet url
* Added a default signing key to the TrustedUpdate component to be used
for verifying updates. This signing key can be authenticated via
gpg --verify i2p/core/java/src/net/i2p/crypto/TrustedUpdate.java
* New public domain SHA1 implementation for the DSA code so that we can
handle signing streams of arbitrary size without excess memory usage
(thanks P.Verdy!)
* Added some helpers to the TrustedUpdate to work off streams and to offer
a minimal CLI:
TrustedUpdate keygen pubKeyFile privKeyFile
TrustedUpdate sign origFile signedFile privKeyFile
TrustedUpdate verify signedFile
* Fixed the tunnel fragmentation handler to deal with multiple fragments
in a single message properly (rather than release the buffer into the
cache after processing the first one) (duh!)
* Added the batching preprocessor which will bundle together multiple
small messages inside a single tunnel message by delaying their delivery
up to .5s, or whenever the pending data will fill a full message,
whichever comes first. This is disabled at the moment, since without the
above bugfix widely deployed, lots and lots of messages would fail.
* Within each tunnel pool, stick with a randomly selected peer for up to
.5s before randomizing and selecting again, instead of randomizing the
pool each time a tunnel is needed.
* Fixed the tunnel fragmentation handler to deal with multiple fragments
in a single message properly (rather than release the buffer into the
cache after processing the first one) (duh!)
* Added the batching preprocessor which will bundle together multiple
small messages inside a single tunnel message by delaying their delivery
up to .5s, or whenever the pending data will fill a full message,
whichever comes first. This is disabled at the moment, since without the
above bugfix widely deployed, lots and lots of messages would fail.
* Within each tunnel pool, stick with a randomly selected peer for up to
.5s before randomizing and selecting again, instead of randomizing the
pool each time a tunnel is needed.
2005-03-18 jrandom
* Minor tweak to the timestamper to help reduce small skews
* Adjust the stats published to include only the relevent ones
* Only show the currently used speed calculation on the profile page
* Allow the full max # resends to be sent, rather than piggybacking the
RESET packet along side the final resend (duh)
* Add irc.postman.i2p to the default list of IRC servers for new installs
* Drop support for routers running 0.5 or 0.5.0.1 while maintaining
backwards compatability for users running 0.5.0.2.
* Update the old speed calculator and associated profile data points to
use a non-tiered moving average of the tunnel test time, avoiding the
freshness issues of the old tiered speed stats.
* Explicitly synchronize all of the methods on the PRNG, rather than just
the feeder methods (sun and kaffe only need the feeder, but it seems ibm
needs all of them synchronized).
* Properly use the tunnel tests as part of the profile stats.
* Don't flood the jobqueue with sequential persist profile tasks, but
instead, inject a brief scheduling delay between them.
* Reduce the TCP connection establishment timeout to 20s (which is still
absurdly excessive)
* Reduced the max resend delay to 30s so we can get some resends in when
dealing with client apps that hang up early (e.g. wget)
* Added more alternative socketManager factories (good call aum!)
* Adjust the old speed calculator to include end to end RTT data in its
estimates, and use that as the primary speed calculator again.
* Use the mean of the high capacity speeds to determine the fast
threshold, rather than the median. Perhaps we should use the mean of
all active non-failing peers?
* Updated the profile page to sort by tier, then alphabetically.
* Added some alternative socketManager factories (good call aum!)
* New strict speed calculator that goes off the actual number of messages
verifiably sent through the peer by way of tunnels. Initially, this only
contains the successful message count on inbound tunnels, but may be
augmented later to include verified outbound messages, peers queried in
the netDb, etc. The speed calculation decays quickly, but should give
a better differential than the previous stat (both values are shown on
the /profiles.jsp page)
2005-03-11 jrandom
* Rather than the fixed resend timeout floor (10s), use 10s+RTT as the
minimum (increased on resends as before, of course).
* Always prod the clock update listeners, even if just to tell them that
the time hasn't changed much.
* Added support for explicit peer selection for individual tunnel pools,
which will be useful in debugging but not recommended for use by normal
end users.
* More aggressively search for the next hop's routerInfo on tunnel join.
* Give messages received via inbound tunnels that are bound to remote
locations sufficient time (taking into account clock skew).
* Give alternate direct send messages sufficient time (10s min, not 5s)
* Always give the end to end data message the explicit timeout (though the
old default was sufficient before)
* No need to give end to end messages an insane expiration (+2m), as we
are already handling skew on the receiving side.
* Don't complain too loudly about expired TunnelCreateMessages (at least,
not until after all those 0.5 and 0.5.0.1 users upgrade ;)
* Properly keep the sendBps stat
* When running the router with router.keepHistory=true, log more data to
messageHistory.txt
* Logging updates
* Minor formatting updates
2005-03-06 jrandom
* Allow the I2PTunnel web interface to select streaming lib options for
individual client tunnels, rather than sharing them across all of them,
as we do with the session options. This way people can (and should) set
the irc proxy to interactive and the eepproxy to bulk.
* Added a startRouter.sh script to new installs which simply calls
"sh i2prouter start". This should make it clear how people should start
I2P.
* Filter HTTP response headers in the eepproxy, forcing Connection: close
so that broken (/malicious) webservers can't allow persistent
connections. All HTTP compliant browsers should now always close the
socket.
* Enabled the GZIPInputStream's cache (they were'nt cached before)
* Make sure our first send is always a SYN (duh)
* Workaround for some buggy compilers
* Loop while starting up the I2PTunnel instances, in case the I2CP
listener isn't up yet (thanks detonate!)
* Implement custom reusable GZIP streams to both reduce memory churn
and prevent the exposure of data in the standard GZIP header (creation
time, OS, etc). This is RFC1952 compliant, and backwards compatible,
though has only been tested within the confines of I2P's compression use
(DataHelper.[de]compress).
* Preemptively support the next protocol version, so that after the 0.5.0.2
release, we'll be able to drop protocol=2 to get rid of 0.5 users.
* Fix one substantial OOM cause (session tag manager was only dropping
tags once the critical limit was met, rather than honoring their
expiration) (duh)
* Lots of small memory fixes
* Double the allowable concurrent outstanding tunnel build tasks (20)
* Really disable the streaming lib packet caching
* Synchronized a message handling point in the SDK (even though its use is
already essentially single threaded, its better to play it safe)
* Don't add new RepublishLeaseSetJobs on failure, just requeue up the
existing one (duh)
* Throttle the number of concurrent pending tunnel builds across all
pools, in addition to simply throttling the number of new requests per
minute for each pool individually. This should avoid the cascading
failure when tunnel builds take too long, as no new builds will be
created until the previous ones are handled.
* Factored out and extended the DataHelper's unit tests for dealing with
long and date formatting.
* Explicitly specify the HTTP auth realm as "i2prouter", though this
alone doesn't address the bug where jetty asks for authentication too
much. (thanks orion!)
* Updated the StreamSinkServer to ignore all read bytes, rather than write
them to the filesystem.
* Don't rerequest leaseSets if there are already pending requests
* Reverted the insufficiently tested caching in the DSA/SHA1 impl, and
temporary disabled the streaming lib packet caching.
* Reduced the resend RTT penalty to 10s
* Further streaming lib caching improvements
* Reduce the minimum RTT (used to calculate retry timeouts), but also
increase the RTT on resends.
* Lower the default message size to 4KB from 16KB to further reduce the
chance of failed fragmentation.
* Extend tunnel rebuild throttling to include fallback rebuilds
* If there are less than 20 routers known, don't drop the last 20 (to help
avoid dropping all peers under catastrophic failures)
* New stats for end to end messages - "client.leaseSetFoundLocally",
"client.leaseSetFoundRemoteTime", and "client.leaseSetFailedRemoteTime"
* Throttle the number of tunnel rebuilds per minute, preventing CPU
overload under catastrophic failures (thanks Tracker and cervantes!)
* Block the router startup process until we've initialized the clock
* Cache temporary memory allocation in the DSA's SHA1 impl, and the packet
data in the streaming lib.
* Fixed a streaming lib bug where the connection initiator would fail the
stream if the ACK to their SYN was lost.
* Reworked the tunnel (re)building process to remove the tokens and
provide cleaner controls on the tunnels built.
* Fixed situations where the timestamper wanted to test more servers than
were provided (thanks Tracker!)
* Get rid of the dead SAM sessions by using the streaming lib's callbacks
(thanks Tracker!)
* Temporary workaround for the I2CP disconnect bug (have the streaminglib
try to automatically reconnect on accept()/connect(..)).
* Loop check for expired lease republishing (just in case)
* Temporary workaround for the I2CP disconnect bug (have the streaminglib
try to automatically reconnect on accept()/connect(..)).
* Loop check for expired lease republishing (just in case)
* Adjusted (and fixed...) the timestamper change detection
* Deal with a rare reordering bug at the beginning of a stream (so we
don't drop it unnecessarily)
* Cleaned up some dropped message handling in the router
* Reduced job queue churn when dealing with a large number of tunnels by
sharing an expiration job
* Keep a separate list of the most recent CRIT messages (shown on the
logs.jsp). This way they don't get buried among any other messages.
* For clarity, display the tunnel variance config as "Randomization" on
the web console.
* If lease republishing fails (boo! hiss!) try it again
* Actually fix the negative jobLag in the right place (this time)
* Allow reseeding when there are less than 10 known peer references
* Lots of logging updates.
* Allow the streaming lib resend frequency to drop down to 20s as the
minimum, so that up to 2 retries can get sent on an http request.
* Add further limits to failsafe tunnels.
* Keep exploratory and client tunnel testing and building stats separate.
* Only use the 60s period for throttling tunnel requests due to transient
network overload.
* Rebuild tunnels earlier (1-3m before expiration, by default)
* Cache the next hop's routerInfo for participating tunnels so that the
tunnel participation doesn't depend on the netDb.
* Fixed a long standing bug in the streaming lib where we wouldn't always
unchoke messages when the window size grows.
* Make sure the window size never reaches 0 (duh)
* Allow the streaming lib resend frequency to drop down to 20s as the
minimum, so that up to 2 retries can get sent on an http request.
* Add further limits to failsafe tunnels.
* Keep exploratory and client tunnel testing and building stats separate.
* Only use the 60s period for throttling tunnel requests due to transient
network overload.
* Rebuild tunnels earlier (1-3m before expiration, by default)
* Cache the next hop's routerInfo for participating tunnels so that the
tunnel participation doesn't depend on the netDb.
* Fixed a long standing bug in the streaming lib where we wouldn't always
unchoke messages when the window size grows.
* Make sure the window size never reaches 0 (duh)
* Only build failsafe tunnels if we need them
* Properly implement the selectNotFailingPeers so that we get a random
selection of peers, rather than using the strictOrdering (thanks dm!)
* Don't include too many "don't tell me about" peer references in the
lookup message - only send the 10 peer references closest to the target.
* Only build new extra tunnels on failure if we don't have enough
* Fix a fencepost in the tunnel building so that e.g. a variance of
2 means +/- 2, not +/- 1 (thanks dm!)
* Avoid an NPE on client disconnect
* Never select a shitlisted peer to participate in a tunnel
* Have netDb store messages timeout after 10s, not the full 60s (duh)
* Keep session tags around for a little longer, just in case (grr)
* Cleaned up some closing event issues on the streaming lib
* Stop bundling the jetty 5.1.2 and updated wrapper.config in the update
so that 0.4.* users will need to do a clean install, but we don't need
to shove an additional 2MB in each update to those already on 0.5.
* Imported the susimail css (oops, thanks susi!)
<failif="tests.failed">Some tests failed; see details above.</fail>
</target>
<targetdepends="init,-do-not-recompile,compile-test-single,-pre-test-run-single,-do-test-run-single,-post-test-run-single"description="Run single unit test."name="test-single"/>
<failif="tests.failed">Some tests failed; see details above.</fail>
</target>
<targetdepends="init,-do-not-recompile,compile-test-single,-pre-test-run-single,-do-test-run-single,-post-test-run-single"description="Run single unit test."name="test-single"/>
<failif="tests.failed">Some tests failed; see details above.</fail>
</target>
<targetdepends="init,-do-not-recompile,compile-test-single,-pre-test-run-single,-do-test-run-single,-post-test-run-single"description="Run single unit test."name="test-single"/>
<p>I2P builds up a new net inside the usual internet, connecting nodes together
via encrypted connections.
It is a JAVA prgram with its most used part (the encryption of the data) written
in handoptimized assembler code.
It will use your bandwith, your RAM and your CPU. It will use them all up if you
do not limit it.
I2P will route unknown traffic through your node, even stuff you dislike.
As that data is encrypted, nobody knows whats data went to or drom your node.
</p>
<p>
First, ALWAYS use the latest stable release.
Development releases are called "mtn version" and are marked with a -, e.g.
0.6.5-1. Those are usually useable by all but could do harm to your I2P
experience.
You can get the latest MTN builds from my eepsite echelon.i2p, but always
remember: I built them, you need to trust me not to changed the code!
After you get the right "i2pupdate.zip" file, put that file into the I2P
directory and hit restart on the router console http://127.0.0.1:7657.
Do NOT deflate the zip file!
</p>
<p>
I2P is very dynamic - after startup it tries to get known to other I2P routers
and measures their speed - you need to wait some 10-120 minutes until your
I2P router knows enough other ones to obtain full power of I2P.
</p>
<h2>Filesharing</h2>
<p>
I2P is able to do anonymous filesharing.
But as there are NO gateways between real net and I2P, you can only share/
download torrents from within I2P. Look e.g. postman.i2p or planet.i2p.
You CANNOT use azureus, utorrent or any other usual client.
You cannot download anonymous torrents from mininova, piratebay or else.
You need to use I2P internal torrents and I2P aware programs like
I2Psnark (builtin, suitable for 1-20 torrents)
I2PRufus (external, python, high CPU load, suitable >20 torrents) http://echelon.i2p/i2prufus
I2P-BT (external, python)
I2PsnarkXL (mod of I2Psnark made by fwd)
</p>
<p>
There are also gnutella and edonkey clients:
i2phex for gnutella (http://echelon.i2p/i2phex)
imule for edonkey (http://echelon.i2p/imule)
</p>
<p>
Remember, as I2P uses other routers to route your traffic via 1-6 other PCs,
your transferrates in P2P are slower than in usual internet.
But you are anonymous, no one can easily (within 2 months-2 years) get your IP!
torrents inside of I2P reaches up to 50 kb/sec, usual are 10-20 kb/sec per torrent
i2phex reaches up to 20 kb/sec, usually 5-10 kb/sec
imule in times reaches 10 kb/sec, usually 5-10 kb/sec
</p>
<p>
In I2PHex and imule you can just tell "share file or directory", in torrent
you need to create a .torrent file and upload that (and ONLY that small .torrent)
file to the trackers like tracker.postman.i2p/
</p>
<p>
I2P is a smaller net (1000 users) which grows slowly. As of which amount of shared
data will slowly rise.
</p>
<p>
I2P is anonymous and it does not censor - there is no administrator.
There IS unwanted stuff like kiddyporn, nazism, or else.
If you dislike all this, do not use I2P.
There is NO way to prohibite this stuff to appear in a anonymous net like I2P.
(as that stuff is available shows the anonymity and transfer function of I2P
is working well enough)
You can delete the destinations in question from your local hosts.txt file (or
deface them) which will partly prevent you to reach those bad sies by accident.
</p>
<h2>Internet (the websites)</h2>
<p>
Only one outproxy (gateway I2P - webpages in usual Internet) is working.
It is NOT official from I2P, I2P does work without.
If that outproxy is slow, offline, gone,.. I2P still works on and cannot do
anything to change that failure.
That outproxy translates usual internet webpages into the I2P net and you can
reach them via your Router.
The best way for usual webpages is TOR, not that outproxy.
Remember: the owner of the outproxy got ALL traffic from all I2P users
visiting Internet pages and will risk that into the police!
</p>
<p>
This proxy is false.i2p. In newer I2P routers it is enabled, but not in
older ones. Go to http://127.0.0.1:7657/i2ptunnel/index.jsp tunnels page and
click on the eepProxy tunnel. Change the entry for the "Outproxies" to false.i2p
and save. On the tunnels page, stop the epproxy tunnel and start it again,
now your router will use the false.i2p outproxy.
</p>
<p>
No other (known) gateways are setup and running. No one we know will run
the gateway for torrent or any other P2P data (and risk his life).
</p>
<h2>Bandwidth</h2>
<p>http://127.0.0.1:7657/config.jsp</p>
<p>Setup your bandwith wisely. Know your linespeed!</p>
<p>
E.g. most common terms are:
1Mbit = roughly 100 kbyte/sec
10 MBit = roughly 1100 kbyte/sec
512 kbit = roughly 50 kbyte/sec
or in germany:
16000er = roughly 1500 kbyte/sec
6000er = roughly 600 kbyte/sec
1000er = roughly 100 kb/sec
</p>
<p>
Set your bandwith limits to 10% under your line speed and burst rate to
your line speed.
Set the bandwith share percentage to:
>80% if lowest bandwith setting is >50k
>50% if lowest bandwith setting is >30k
>20% if lowest bandwith setting is >16k
</p>
<p>There is no shared bandwith under 16k.</p>
<p>
Limit your participating tunnels (shared bandwith) on:
http://127.0.0.1:7657/configadvanced.jsp
with the line:
router.maxParticipatingTunnels=500
</p>
<p>
2000 is for roughly 600 kb/sec - very high value with high CPU load
1000 is for roughly 300 kb/sec
600 is a good value for 150-200kb/sec
300 is roughly 90 kb/sec
150 roughly 50 kb/sec
Remember: even failed tunnel requests will result in a part tunnel on the hops in between!
Those said, there are far more part tunnels unused than used in the live net under load, which
results in slower bandwith per tunnel in the end.
It is wise to first limit the bandwith and afterwards the part tunnels, e.g. set some more part
tunnels and let I2P reach the bandwith limit instead of the part tunnels limit!
</p>
<h2>What is shared bandwidth?</h2>
<p>
I2P transports your date from the client to the server through 1-6 hops
(other I2P routers). Each of this hops sees the data from you as "participating
tunnel" - which is the shared bandwith of them.
With this in mind, I2P needs some amount of this shared bandwith at some
amount of routers.
Share as much as you are able of - others will thank you!
</p>
<p>
With the "share percentage" set like above, you will obtain enough speed for
your own traffic and obtain some participating tunnels (if >16kb/sec) with some
noise traffic to hide your traffic in the stream.
</p>
<p>
With release 0.6.5 there is some method to prefer your own traffic ahead
of shared traffic which will result in better experience to you!
</p>
<h2>Addressbook</h2>
<p>
I2P uses a local addressbook to link short DNS names with the internal used 512bit
hashes (which are destination IDs).
Those links are saved insside the hosts.txt and userhosts.txt files in the i2p
directory.
Hosts which are not in those files cannot be reached via the short DNS names
and a error message with "jumper links" will appear. Those links will ask
some hosts services and forward to the correct site (if the site is known to them).
Those hosts services just made a form to add new "hosts" and those results public
available.
You can subscribe to those hosts service and let your hosts.txt file be updated
automatic. Go to http://127.0.0.1:7657/susidns/subscriptions.jsp SusiDNS
and enter the hosts services into the textbox (and save afterwards):
http://www.i2p2.i2p/hosts.txt
http://stats.i2p/cgi-bin/newhosts.txt
http://tino.i2p/hosts.txt
http://i2host.i2p/cgi-bin/i2hostag
You can add one of them, two or all.
SusiDNS will now ask those hosts for new entries to the hosts.txt and those
will be added to your hosts.txt. The userhosts.txt will ONLY be updated by
yourself (the user) and not be published into the net!
Remember, names once set could not be changed! If you loose your key (destination
ID) to your eepsite, service,..., there is no way to change the linking
between the DNS name and the (lost) destination ID automatic! Only manual by each
user itself - great topic to discuss of need to renew DNS hostnames.
As this subscription will not update old entries, you can "deface" unwanted
eepsites with a false key and if you hit the bad name in browser by accident,
you will not see the bad stuff!
</p>
<h2>Out of Memory errors </h2>
<p>
If your router hits the Out of Memory error - check your logs!
Usual point for OOM are to much torrents in i2psnark - i2psnark is a real
memory hogg and >10 torrents it requiers hell a lot of memory!
</p>
<p>
Maybe it is possible for you to increase the wrapper memory config.
This ONLY works if you start the I2P service restartable with console
(on Windows).
In I2P directory edit the wrapper.config file and change the values:
wrapper.java.maxmemory=256 (or even to 512, IF possible)
Afterwards shutdown I2P complete (the service) and restart it.
</p>
<h2>Blocklists</h2>
<p>
Sometimes attackers trying to flood the I2P net and try to do some harm.
And some folks setting localnet IPs as their internet reachable address.
To prevent those bad router to harm the local router, I2P implemented
a local blocklist system. It is NOT integrated automatic as it could
really harm your I2P experience if setup the wrong way.
The way to enable blocklists is:
Get the file http://zzz.i2p/files/blocklist.txt and copy this file into the
I2P directory.
On http://127.0.0.1:7657/configadvanced.jsp set the option
router.blocklist.enable=true - click on Apply and restart the router
with the restart link left on router console.
The blockfile.txt file follows a special order, you<6F>ll get it if you read it.
The first entry is the reason to be shown on http://127.0.0.1:7657/profiles.jsp
at the bottom in the shitlist section.
The second entry is the IP or the dest ID of a router.
Right now there are only private subnets in the blocklist AND one chinese router
which floods the floodfill DB while restarting every few minutes with a different
router ID and far to less bandwith for being a floodfill router.
</p>
<p>(By echelon -- echelon.i2p )</p>
</body>
</html>
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
