Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 30 Aug 1999 10:58:48 -0700 (PDT)
From:      Matthew Dillon <dillon@apollo.backplane.com>
To:        Suresh Rajagopalan <sraja@cinenet.net>
Cc:        freebsd-smp@FreeBSD.ORG
Subject:   Re: SMP freezes on 3.2-STABLE
Message-ID:  <199908301758.KAA16067@apollo.backplane.com>
References:   <Pine.GSO.3.96.990830103058.21640A-100000@hermosa.cinenet.net>

next in thread | previous in thread | raw e-mail | index | archive | help
:a) 
:
:This time I was able to break into DDB.  Here's the output (BTW, I had to
:transcribe this, is there any way to dump the output to a file?)

    No, you have to transcribe it or set the console up on a serial port
    (which can be annoying) and log it.

:b)
:trace shows the following:
:
:panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4
:bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1
:nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f
:lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1
:namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133
:stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44
:syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107
:Xint0x80_syscall() at Xint)x80_syscall + 0x4c
:
:--- 
:c)
:ps shows a few http and perl processes, some in nfsrecv state.   The
:output was too long to copy by hand.  Hope this helps.
:
:--
:d)
:
:On one of the machines, I now also see this message in the syslog:
:
:xl0: command never completed 
:
:(at random intervals)
:...
:Thanks for your help.
:
:Suresh

    Hmm.  This is very odd.  It seems unlikely to be a bug in NFS if 
    turning off SMP fixes the problem.

    If the machine didn't crash and ctl-alt-esc breaking it shows an
    active stack frame from a running process, it could be that the
    process is getting stuck in an endless loop somewhere or somehow.

    I'm at a loss at the moment.  It sounds like an SMP problem of some
    sort.  I'm hoping one of the SMP guys will have a brainfart :-)

    If it cannot be resolved soon I would back both machines down to
    a single-cpu and wait... having production machines crash is not fun.

    Current ought to work better for SMP then stable, but I hesitate to
    suggest it for a production machine.

					-Matt
					Matthew Dillon 
					<dillon@backplane.com>


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-smp" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199908301758.KAA16067>