Big problems with 7.1 locking up :-(

heliocentric at gmail.com heliocentric at gmail.com
Sat Jan 10 18:07:49 PST 2009


I noticed a similar problem testing 7.1-RC1, It seemed to be a deep
deadlock, as it was triggered by lighttpd doing kern_sendfile, and
never returning. The side effects (being unable to create processes,
etc) is similar.

My kernconf is below, try building the kernel, and send an email
containing the backtrace from any process that has blocked (in my
case, lighttpd attempting to sendfile a large amount of data to php
fastcgi triggered it, but that's a guess on my part). Note that this
includes witness, and invariants, so performance will be hit. Also,
enable watchdogd, and add -e 'ls -al /etc' to it's flags. It should
drop you to a debugger with a backtrace within a few seconds of the
lock being triggered, and it should output a backtrace and any
invariant/witness lock warnings. Obviously if you don't have a serial
or local console, don't do this.

include         GENERIC
ident           DEBUG
options         KDB
options         DDB
options         SW_WATCHDOG
options DEBUG_VFS_LOCKS
options INVARIANTS
options WITNESS

On 1/10/09, Pete French <petefrench at ticketswitch.com> wrote:
>> FWIW, the other guy I know who is having this problem had already
>> switched to using ULE under 7.0-release, and did not have any
>> problems with it.  So *his* problem was probably not related to
>> SCHED_ULE, unless something has recently changed there.
>
> Well, one of my machines just locked up again, even with SCHED_4BSD
> on it, so I am now thinking it is unrelated.
>
> The machine has completely locked - no response to pings, no
> response to keypresses, nor to the power button. There is nothing
> printed on the console - it is just sitting there with a login prompt :-(
>
> This is really not good - these are extremely common servers after all, and
> I am just running bog standard 7.1 with apache and mysql. This is happening
> across several different servers, all of which are slight variants on
> the DL360, so I dont think it is something perculiar to me.
>
> -pete.
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"
>


More information about the freebsd-stable mailing list