Dave Jones <davej(a)redhat.com> írta:
On Thu, Aug 03, 2006 at 12:05:22AM +0200, Zoltan
Boszormenyi
wrote:
> ---------------------------------------------
> [ INFO: possible recursive locking detected ]
> ---------------------------------------------
> vol_id/973 is trying to acquire lock:
> (&md->io_lock){----}, at: [<ffffffff880c8653>]
dm_request+0x25/0x130
> [dm_mod]
>
> but task is already holding lock:
> (&md->io_lock){----}, at: [<ffffffff880c8653>]
dm_request+0x25/0x130
> [dm_mod]
That's a new one. I've posted it upstream, hopefully we'll
get a
fix for that quickly.
Thanks, I saw it.
> Lukewarm IQ detected in hotplug locking
> BUG: warning at kernel/cpu.c:43/lock_cpu_hotplug() (Not
tainted)
This one is already in bugzilla, and is a real pita to fix.
> The third is that httpd-2.2.2-7 segfaults.
> Downgrading to FC5 last update (httpd-2.2.2-1.2)
> doesn't help.
Don't know about that one. File a bug in bugzilla.
If it dumps core, grab the -debuginfo rpm too, and
feed it to gdb and get a backtrace, and add that to the bug.
Strace didn't give me any interesting info,
it segfaults after a bunch of munmap.
It doesn't dump core but I run it under gdb.
It suggests that the bug may be in glibc?
BZ #201145 , gdb output is quoted under additional info.
Thanks,
Zoltán Böszörményi
_____________________________________________________________________
A legjobb akciós utazások oldala, ahol most rengeteg nyeremény vár Rád:
http://ad.adverticum.net/b/cl,1,6022,99401,161631/click.prm