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.
- 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)
<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"/>
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.