#27813: Tor 0.3.4.8 is leaking memory
-------------------------------------------------+-------------------------
Reporter: anong | Owner: (none)
Type: defect | Status:
| needs_information
Priority: Very High | Milestone: Tor:
| 0.3.5.x-final
Component: Core Tor/Tor | Version: Tor:
| 0.3.4.8
Severity: Critical | Resolution:
Keywords: regression? memleak oom | Actual Points:
034-backport tor-relay 035-must |
Parent ID: | Points:
Reviewer: | Sponsor:
-------------------------------------------------+-------------------------
Comment (by starlight):
Replying to [comment:30 dgoulet]:
>. . .
>KIST should prevent that in theory but not entirely. Some could always go
as in flushing 1M of cells and only one goes through leaving the rest in
the kernel.
>
>The bigger issue is that we still have cells bypassing the KIST scheduler
that are directly written on the socket. We have several ticket open about
these issues.
Thank you for the update, the details are informative and interesting.
After posting I realized stuck sockets will hide an EOF condition and when
the other side closes EOFs are not detected resulting in a socket leak,
but one cured by the fix. A more carefully read turns up that performance
degradation did result.
Newer kernels account kernel SKBs under the owning process for correct
CGROUP resource control, and graphs above show the leaked memory as "app"
memory.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/27813#comment:31>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs
