diff options
| author | Nick Mathewson <nickm@torproject.org> | 2013-06-16 09:55:44 -0400 |
|---|---|---|
| committer | Nick Mathewson <nickm@torproject.org> | 2013-06-18 10:15:16 -0400 |
| commit | 2e1fe1fcf93c2a77805048bea5c535ca4456d583 (patch) | |
| tree | 6a396bc09f558b9611282de2d54a7f7824696095 /doc | |
| parent | 2a95f3171681ee53c97ccba9d80f4454b462aaa7 (diff) | |
| download | tor-2e1fe1fcf93c2a77805048bea5c535ca4456d583.tar.gz tor-2e1fe1fcf93c2a77805048bea5c535ca4456d583.zip | |
Implement a real OOM-killer for too-long circuit queues.
This implements "algorithm 1" from my discussion of bug #9072: on OOM,
find the circuits with the longest queues, and kill them. It's also a
fix for #9063 -- without the side-effects of bug #9072.
The memory bounds aren't perfect here, and you need to be sure to
allow some slack for the rest of Tor's usage.
This isn't a perfect fix; the rest of the solutions I describe on
codeable.
Diffstat (limited to 'doc')
| -rw-r--r-- | doc/tor.1.txt | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/doc/tor.1.txt b/doc/tor.1.txt index 773fccf536..1f3afef2ac 100644 --- a/doc/tor.1.txt +++ b/doc/tor.1.txt @@ -1475,6 +1475,15 @@ is non-zero): localhost, RFC1918 addresses, and so on. This can create security issues; you should probably leave it off. (Default: 0) +**MaxMemInCellQueues** __N__ **bytes**|**KB**|**MB**|**GB**:: + This option configures a threshold above which Tor will assume that it + needs to stop queueing cells because it's about to run out of memory. + If it hits this threshold, it will begin killing circuits until it + has recovered at least 10% of this memory. Do not set this option too + low, or your relay may be unreliable under load. This option only + effects circuit queues, so the actual process size will be larger than + this. (Default: 8GB) + DIRECTORY SERVER OPTIONS ------------------------ |