| Commit message (Expand) | Author | Age |
|---|
| * | Rename aci to circ_id throughout.•••svn:r784
| Nick Mathewson | 2003-11-11 |
| * | change WARNING to WARN•••and fix a few typos
svn:r571
| Roger Dingledine | 2003-10-10 |
| * | Update LICENSE and copyright dates.•••svn:r560
| Nick Mathewson | 2003-10-08 |
| * | refactor around connection_edge_send_command()•••svn:r539
| Roger Dingledine | 2003-10-04 |
| * | wrap strdup; prefer time() to gettimeofday()•••svn:r538
| Roger Dingledine | 2003-10-04 |
| * | Add new cell fullness and bandwidth stats.•••svn:r533
| Nick Mathewson | 2003-10-02 |
| * | bugfixes and features: closer to making dirserv work•••fix a variety of seg faults
don't try to list OPs in running-routers
write cached-directory to disk when rebuilding the dir
on boot, dirservers load approved-routers file
on boot, dirservers load cached directory file
svn:r508
| Roger Dingledine | 2003-09-29 |
| * | clean up receiver buckets; prepare for payloads in relay_end; note a few bugs•••svn:r502
| Roger Dingledine | 2003-09-27 |
| * | first pass: obey log convention•••ERR is if something fatal just happened
WARNING is something bad happened, but we're still running. The bad thing
is either a bug in the code, an attack or buggy protocol/implementation
of the remote peer, etc. The operator should examine the bad thing and
try to correct it.
(No error or warning messages should be expected. I expect most people
to run on -l warning eventually.)
NOTICE is never ever used.
INFO means something happened (maybe bad, maybe ok), but there's nothing
you need to (or can) do about it.
DEBUG is for everything louder than INFO.
svn:r486
| Roger Dingledine | 2003-09-26 |
| * | clean up exported api's•••svn:r461
| Roger Dingledine | 2003-09-16 |
| * | implemented cpuworkers•••please poke at it and report bugs
still needs polishing, and only handles onions now (should handle
OR handshakes too)
svn:r402
| Roger Dingledine | 2003-08-20 |
| * | Change many files to new log_fn format•••svn:r333
| Nick Mathewson | 2003-06-17 |
| * | relay queues are obsolete (woo!)•••they used to be used for
* queueing relay cells at the edge of the network, when windows are empty
* queueing relay cells that arrive after an onion but before the onion
has been processed.
both of these uses are gone. so out they go.
svn:r315
| Roger Dingledine | 2003-06-13 |
| * | send truncates AP-ward in a circuit, not destroys•••svn:r313
| Roger Dingledine | 2003-06-13 |
| * | implement truncate and truncated (untested)•••clean up circuit_deliver_relay_cell convention
svn:r312
| Roger Dingledine | 2003-06-12 |
| * | remove dead code•••circuits no longer queue more cells when the windows are empty --
they simply don't package it from the buffer if they're not going to want it.
we can restore this code later if we need to resume queueing.
svn:r294
| Roger Dingledine | 2003-05-20 |
| * | add circuit-level sendme relay cells•••remove sendme cells
replace malloc with tor_malloc
patch (but not track down) bug in onion pending list
streamline connection_ap handshake
svn:r293
| Roger Dingledine | 2003-05-20 |
| * | sign directories with the signing key•••svn:r274
| Roger Dingledine | 2003-05-07 |
| * | put some symbolic constants to the onion skin lengths•••svn:r265
| Roger Dingledine | 2003-05-06 |
| * | incremental path building in; uses ephemeral DH; onions are gone•••still need to change circuit-level sendmes
svn:r264
| Roger Dingledine | 2003-05-05 |
| * | streams are now 8 bytes, and are recognized by intermediate hops•••the OP only crypts the appropriate number of times depending on which
layer (hop on the path) it's for/from.
svn:r262
| Roger Dingledine | 2003-05-02 |
| * | terminology shift: data->relay, topic->relay, topic->stream•••svn:r258
| Roger Dingledine | 2003-05-01 |
| * | bugfix: a circ can't be youngest if it's still connecting to the first hop•••svn:r255
| Roger Dingledine | 2003-04-20 |
| * | bugfix: refactor to always use circuit_remove•••this way we can always check if a new circ needs to be launched
svn:r254
| Roger Dingledine | 2003-04-20 |
| * | further cleanup, test.c still has some bugs•••svn:r241
| Roger Dingledine | 2003-04-16 |
| * | Factor out timeval-related functions.•••svn:r237
| Nick Mathewson | 2003-04-16 |
| * | Add magic to end of C files to make emacs happy; split test invocation into s...•••svn:r224
| Nick Mathewson | 2003-04-07 |
| * | be less noisy at -l info, now that flow control bug is solved•••svn:r172
| Roger Dingledine | 2003-03-10 |
| * | Bugfixes and enhancements in sendmes and dns farm•••svn:r161
| Roger Dingledine | 2003-02-18 |
| * | major overhaul: dns slave subsystem, topics•••on startup, it forks off a master dns handler, which forks off dns
slaves (like the apache model). slaves as spawned as load increases,
and then reused. excess slaves are not ever killed, currently.
implemented topics. each topic has a receive window in each direction
at each edge of the circuit, and sends sendme's at the data level, as
per before. each circuit also has receive windows in each direction at
each hop; an edge sends a circuit-level sendme as soon as enough data
cells have arrived (regardless of whether the data cells were flushed
to the exit conns). removed the 'connected' cell type, since it's now
a topic command within data cells.
at the edge of the circuit, there can be multiple connections associated
with a single circuit. you find them via the linked list conn->next_topic.
currently each new ap connection starts its own circuit, so we ought
to see comparable performance to what we had before. but that's only
because i haven't written the code to reattach to old circuits. please
try to break it as-is, and then i'll make it reuse the same circuit and
we'll try to break that.
svn:r152
| Roger Dingledine | 2003-01-26 |
| * | bugfix: couldn't send two creates, two datas, and the destroy all at once•••(amazing the odd behavior you get to test when you have a flaky modem
connection)
svn:r148
| Roger Dingledine | 2002-12-23 |
| * | create cells are now queued and processed only when idle•••we also queue data cells destined for a circuit that is
pending, and process them once the circuit opens
destroys reach into the queue and remove the pending onion,
along with its collected data cells
svn:r142
| Roger Dingledine | 2002-11-27 |
| * | per-second cell statistics to help with profiling•••svn:r140
| Roger Dingledine | 2002-11-24 |
| * | added OnionsPerSecond to prevent create flooding•••first cut, probably needs more playing with
svn:r137
| Roger Dingledine | 2002-11-23 |
| * | cell.c is now obsolete•••svn:r133
| Roger Dingledine | 2002-10-03 |
| * | added 'connected' cell type•••see http://archives.seul.org/or/dev/Sep-2002/msg00018.html
svn:r103
| Roger Dingledine | 2002-09-17 |
| * | port is now kept in host order except in sin_port•••svn:r82
| Roger Dingledine | 2002-08-24 |
| * | Folded cell.? into src/or•••svn:r64
| Roger Dingledine | 2002-07-19 |
| * | Implemented router twins•••I modified new_route so we don't pick twins back-to-back in the path.
I also had to patch my previous uses of connection_twin_get_by_addr_port()
because they assumed that "addr" and "port" would be the same for a twin
as for the original router.
svn:r56
| Roger Dingledine | 2002-07-18 |
| * | Implemented congestion control•••Servers are allowed to send 100 cells initially, and can't send more until
they receive a 'sendme' cell from that direction, indicating that they
can send 10 more cells. As it currently stands, the exit node quickly
runs out of window, and sends bursts of 10 whenever a sendme cell gets
to him. This is much much much faster (and more flexible) than the old
"give each circuit 1 kB/s and hope nothing overflows" approach.
Also divided out the connection_watch_events into stop_reading,
start_writing, etc. That way we can control them separately.
svn:r54
| Roger Dingledine | 2002-07-18 |
| * | Implemented link padding and receiver token buckets•••Each socket reads at most 'bandwidth' bytes per second sustained, but
can handle bursts of up to 10*bandwidth bytes.
Cells are now sent out at evenly-spaced intervals, with padding sent
out otherwise. Set Linkpadding=0 in the rc file to send cells as soon
as they're available (and to never send padding cells).
Added license/copyrights statements at the top of most files.
router->min and router->max have been merged into a single 'bandwidth'
value. We should make the routerinfo_t reflect this (want to do that,
Mat?)
As the bandwidth increases, and we want to stop sleeping more and more
frequently to send a single cell, cpu usage goes up. At 128kB/s we're
pretty much calling poll with a timeout of 1ms or even 0ms. The current
code takes a timeout of 0-9ms and makes it 10ms. prepare_for_poll()
handles everything that should have happened in the past, so as long as
our buffers don't get too full in that 10ms, we're ok.
Speaking of too full, if you run three servers at 100kB/s with -l debug,
it spends too much time printing debugging messages to be able to keep
up with the cells. The outbuf ultimately fills up and it kills that
connection. If you run with -l err, it works fine up through 500kB/s and
probably beyond. Down the road we'll want to teach it to recognize when
an outbuf is getting full, and back off.
svn:r50
| Roger Dingledine | 2002-07-16 |
| * | it wouldn't return when the next router isn't up, leading to later seg fault•••i'm going to take a shower, and then solve the deadlocking problem mat found
svn:r38
| Roger Dingledine | 2002-07-10 |
| * | put in the support for 'router twins'•••basically, a twin is a router which is different except it shares
the same keypair. so in cases where we want to find a "next router"
and all we really care is that it can decrypt the next onion layer,
then a twin is just as good.
we still need to decide how to mark twins in the routerinfo_t and in
the routers config file.
svn:r30
| Roger Dingledine | 2002-07-08 |
| * | Integrated onion proxy into or/•••The 'or' process can now be told (by the global_role variable) what
roles this server should play -- connect to all ORs, listen for ORs,
listen for OPs, listen for APs, or any combination.
* everything in /src/op/ is now obsolete.
* connection_ap.c now handles all interactions with application proxies
* "port" is now or_port, op_port, ap_port. But routers are still always
referenced (say, in conn_get_by_addr_port()) by addr / or_port. We
should make routers.c actually read these new ports (currently I've
kludged it so op_port = or_port+10, ap_port=or_port+20)
* circuits currently know if they're at the beginning of the path because
circ->cpath is set. They use this instead for crypts (both ways),
if it's set.
* I still obey the "send a 0 back to the AP when you're ready" protocol,
but I think we should phase it out. I can simply not read from the AP
socket until I'm ready.
I need to do a lot of cleanup work here, but the code appears to work, so
now's a good time for a checkin.
svn:r22
| Roger Dingledine | 2002-07-02 |
| * | made 'app' connection be 'exit' connection•••general cleanup, particularly in buffers.c
svn:r17
| Roger Dingledine | 2002-06-30 |
| * | Initial revision•••svn:r2
| Roger Dingledine | 2002-06-26 |