Re: Packet loss every 30.999 seconds

In any case, it appears that my patch is a no-op, at least for the
problem I was trying to solve. This has me confused, however, because at
one point the problem was mitigated with it. The patch has gone through
several iterations, however, and it could be that it was made to the top
of the loop, before any of the checks, in a previous version. Hmmm.

(replying to myself)

I just found an earlier version of the patch, and sure enough, it was
to the top of the loop. Unfortunately, that version caused the system to
crash because vp was occasionally invalid after the wakeup.

Anyway, let's see if Mark's packet loss problem is indeed related to
this code. If he does the find just after boot and immediately sees the
problem, then I would say that is fairly conclusive. He could also release
the cached vnodes by temporarily setting kern.maxvnodes=10000 and then
setting it back to whatever it was previously (probably 60000-100000).
If the problem then goes away for awhile, that would be another good
indicator.

-DG

David G. Lawrence
President
Download Technologies, Inc. - http://www.downloadtech.com - (866) 399 8500
The FreeBSD Project - http://www.freebsd.org
Pave the road of life with opportunities.
_______________________________________________
freebsd-net@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • [x86_64 MCE] [RFC] mce.c race condition (or: when evil hacks are the only options)
    ... The race requires a large number of machine checks to be occurring in order ... In the normal case, the rest would get cleaned up by the subsequent loop, ... fact is waiting for all CPUs to be done, which could take up to a tick -- or ... I've come up with a patch that does this, ...
    (Linux-Kernel)
  • Re: Problem with inotify
    ... > Thanks for writing that patch, ... > inotify-test before unmounting results in a clean unmount. ... My analysis was that there is an infinite loop and this is what ... > loop when unmounting with inotify watches active. ...
    (Linux-Kernel)
  • Re: Packet loss every 30.999 seconds
    ... of the loop, before any of the checks, in a previous version. ... I just found an earlier version of the patch, and sure enough, it was ... If he does the find just after boot and immediately sees the ... the cached vnodes by temporarily setting kern.maxvnodes=10000 and then ...
    (freebsd-stable)
  • Re: [PATCH] make-cancel_rearming_delayed_work-reliable-fix
    ... but this needs another patch not related to ... all this made per cpu. ... than one loop of one cpu workqueue to cancel any "normal" work ... _PENDING flag is overloaded here and one flag still free. ...
    (Linux-Kernel)
  • Re: [PATCH 00/17] per device dirty throttling -v7
    ... Patches are against 2.6.22-rc4-mm2 ... writes in a setup involving a loop dev ... ext3 over loop over ext3 ... Without the patch, in all the cases I've seen deadlocks or long ...
    (Linux-Kernel)