- Dec 28, 2021
-
-
zzz authored
Actually update the stat for stores in dbStoreSent(); we are generally storing to different ffs than lookups, so we need the stat for stores as well, since we use it as the timeout in StoreJob. Change from 1-day to 1-hour stat. Switch to avgOrLifetimeAvg() so the rate is always valid. Reduce max time used for timeout. This allows more peers to be tried before total timeout Previously, the per-peer timeout was almost always the max. Make sendStore() package private. Javadocs and cleanups.
-
zzz authored
-
- Dec 27, 2021
-
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
so that torrents stop when the router stops and restart when the router restarts. - Use BWLimits from the DirMonitor as a periodic test that the router is there - DirMonitor does not attempt to autostart torrents if BWLimits test fails - DirMonitor does autostart existing torrents when BWLimits test passes again - Register disconnect listener with socket manger and stop all torrents on disconnect - Use stopTorrent(true) on router errors to prevent changing the persisted torrent running status - Change autostart default to true for standalone Possibly more todo for corner cases or other start/stop/fail scenarios.
-
zzz authored
in client-side AppContext SimpleSession, to prevent router-side error message when closing socket, e.g. for BW limits check
-
zzz authored
it's not a i2pd bug at startup that fixes itself.
-
- Dec 23, 2021
- Dec 21, 2021
-
-
zzz authored
as requested by R4SAS
-
zzz authored
- Write directly from writer threads, except for during establishment and when write doesn't complete; throw those to the pumper as usual - New NTCPCon writelock, readlock, and statlock (formerly all on NTCPCon.this) to prevent deadlocks - NTCPCon chan and key now volatile, remove synch to prevent deadlocks - All interestOps changes now lock on the key via setInterest() and clearInterest() since changes may now happen in multiple threads - Set _paddingConfig at initialization to avoid NPE Greatly reduces pumper loops and CPU As proposed by jogger Reviewed by zlatinb Ref: http://zzz.i2p/topics/3192
-
- Dec 19, 2021
- Dec 18, 2021
-
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
-
zzz authored
Check bounds at initialization Remove unused setIPRestriction()
-
zzz authored
Removed in May 2011 when we added fast tier slices Also add support in exploratory tunnels Create MaskedIPSet in peer selectors, pass to ProfileOrganizer.selectXXX() for each call. Not required for one-hop tunnels. Disable for test networks (i2np.allowLocal) Reported by 'vulnerability_reports' http://zzz.i2p/topics/3215
-
- Dec 13, 2021
-
-
zzz authored
-
- Dec 11, 2021
-
-
idk authored
disable any chance of JNDI lookups in log4j.properties file by setting %m{nolookups}. I don't think we're actually vulnerable to CVE-2021-44228 if I'm understanding correctly, by default it doesn't seem like we actually use log4j for much of anything and we don't do much logging of arbitrarily crafted remote inputs, but also it seems like this JNDI lookups thing is way more trouble than it could possibly be worth to us. Maybe it's a good idea to make sure it's turned off by default.
-
- Dec 10, 2021
-
-
zzz authored
-
- Dec 08, 2021
-
-
zzz authored
- Add fromPort and toPort to Sink interface (breaking API change) - Change cache maps from Destination to I2PSocketAddress to include port - Accept host:port for destination in Streamr Client, use port as toPort in pinger - Change to muxed listener in I2PSource, only listen for specified protocols - Eliminate thread and queue in I2PSource, process messages inline in the listener - Add support for handling both repliable and raw datagrams in a single I2PSource instance - Remove verify option from I2PSource and I2PTunnelUDPServerBase, always verify repliable datagrams - Add getPort() method to UDPSource - Add a constructor to UDPSink to pass in an existing DatagramSocket - Change I2PTunnelUDPClientBase to receive both repliable and raw - Change SOCKSUDPTunnel reply handling strategy to key on I2CP toPort; remove ReplyTracker; the tunnel would not have worked before, because it expected raw replies only but MultiSink required a destination to look up where to forward the reply. - Mark SOCKSUDPTunnel as preliminary; note lack of support for raw replies; untested - Change Streamr Client Pinger to support fromPort - Change Streamr Server to remember fromPort in subscriptions and use it as toPort in data stream - Move fields to top of classes for sanity - Cleanups and log tweaks
-
- Dec 07, 2021
-
-
zzz authored
-