[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ssic-linux-devel
Subject:    [SSI-devel] Re: unixnm.c oops
From:       Roger Tsang <roger.tsang () gmail ! com>
Date:       2005-04-21 4:48:28
Message-ID: 4982633505042021482495f4f0 () mail ! gmail ! com
[Download RAW message or body]

I'm doing a make distclean and rebuild of the kernel, and keep the
warnings.  Maybe the output would be of interest.

-Roger

On 4/20/05, Roger Tsang <roger.tsang@gmail.com> wrote:
> This bug is probably related to the unixnm.c oops previously reported
> and was thought to be kernel networking sockets without mmap IO.
> However this time the kernel networking sockets has mmap IO compiled
> in.  So this must be something else, and probably not shm or IPC.
> 
> -Roger
> 
> On 4/20/05, Roger Tsang <roger.tsang@gmail.com> wrote:
> > I'm using SSI-1.2.2-Lustre-1.2.4 with Laura's IPC patch and crashed
> > the whole cluster today.  My root filesystem is not NFS exported, but
> > seems like whenever the whole cluster crashes the CFS hard mount would
> > need a manual fsck even when it is a journaling fs.  This is a UP
> > kernel for P3/Coppermine with highmem.
> >
> > The initnode crashed so bad I didn't get any response on the console.
> > The failover node got into kdb however.  This is what I have on the
> > failover node, and haven't reboot the failover node yet.
> >
> > -Roger
> >
> > The following is from node 2.
> >
> > kdb> dmesg
> > <3>unixnmsvr_put: entry not found for node 2 and ino 33556058
> > <4>------------[ cut here ]------------
> > <4>kernel BUG at unixnm.c:696!
> > <4>invalid operand: 0000
> > <4>tun loop cls_u32 sch_sfq sch_htb softdog nfsd ip_vs_sed ipt_REJECT ipt_multip
> > ort ipt_state ip_conntrack ipt_TCPMSS iptable_filter ip_tables microcode ide-cd
> > s
> > <4>CPU:    0
> > <4>EIP:    0060:[<c020bb58>]    Not tainted
> > <4>EFLAGS: 00210246
> > <4>
> > <4>EIP is at unixnm_put [kernel] 0x188 (2.4.22-1.2199.nptl_ssi_9up)
> > <4>eax: 00000000   ebx: c0564768   ecx: 00000001   edx: d52b1f68
> > <4>esi: 00000014   edi: 0200065a   ebp: cbeeff2c   esp: cbeeff00
> > <4>ds: 0068   es: 0068   ss: 0068
> > <4>Process nautilus (pid: 133488, stackpage=cbeef000)
> > <4>Stack: 00000002 cbeeff1c 00000014 00000002 0200065a 00000000 cb943b00 fffffd4
> > 4
> > <4>       cd6f3b80 cb943c30 cb943b00 cbeeff44 c03419a3 cb943c30 00000286 cb943c3
> > 0
> > <4>       d7edf280 cbeeff54 c02f3df6 cb943c30 cb943b00 cbeeff6c c02f43d7 cb943c3
> > 0
> > <4>Call Trace:
> > <4>[<c03419a3>] unix_release [kernel] 0x43 (0xcbeeff30)
> > <4>[<c02f3df6>] sock_release [kernel] 0x56 (0xcbeeff48)
> > <4>[<c02f43d7>] sock_close [kernel] 0x37 (0xcbeeff58)
> > <4>[<c014f50f>] fput [kernel] 0xef (0xcbeeff70)
> > more>
> > Only 'q' or 'Q' are processed at more prompt, input ignored
> > <4>[<c014dfab>] filp_close [kernel] 0x4b (0xcbeeff90)
> > <4>[<c014e02f>] sys_close [kernel] 0x4f (0xcbeeffac)
> > <4>[<c010bae7>] system_call [kernel] 0x33 (0xcbeeffc0)
> > <4>
> > <4>Code: 0f 0b b8 02 c6 d7 39 c0 e9 3a ff ff ff 89 7c 24 10 8d 55 f0
> > <4>
> > kdb>
> > kdb> bt
> > Stack traceback for pid 133488
> > 0xcbeee000   133488        1  1    0   R  0xcbeee350 *nautilus
> > EBP        EIP        Function (args)
> > 0xcbeeff2c 0xc020bb58 unixnm_put+0x188 (0xcb943c30, 0x286, 0xcb943c30, 0xd7edf28
> > 0)
> >                                kernel .text 0xc0100000 0xc020b9d0 0xc020bba0
> > 0xcbeeff44 0xc03419a3 unix_release+0x43 (0xcb943c30, 0xcb943b00)
> >                                kernel .text 0xc0100000 0xc0341960 0xc03419d0
> > 0xcbeeff54 0xc02f3df6 sock_release+0x56 (0xcb943c30, 0xce89c700, 0x0, 0xce89c700
> > )
> >                                kernel .text 0xc0100000 0xc02f3da0 0xc02f3e00
> > 0xcbeeff6c 0xc02f43d7 sock_close+0x37 (0xcb943b00, 0xce89c700, 0xcb938580, 0xce8
> > 9c700, 0xd323e980)
> >                                kernel .text 0xc0100000 0xc02f43a0 0xc02f43f0
> > 0xcbeeff8c 0xc014f50f fput+0xef (0xce89c700, 0xd323e980, 0xce89c700, 0xe, 0x97c8
> > a40)
> >                                kernel .text 0xc0100000 0xc014f420 0xc014f530
> > 0xcbeeffa8 0xc014dfab filp_close+0x4b (0xce89c700, 0xd323e980, 0xcbeee000)
> >                                kernel .text 0xc0100000 0xc014df60 0xc014dfe0
> > 0xcbeeffbc 0xc014e02f sys_close+0x4f (0xe, 0x0, 0x4d1af34, 0xe, 0x97c8a40)
> >                                kernel .text 0xc0100000 0xc014dfe0 0xc014e040
> >            0xc010bae7 system_call+0x33
> >                                kernel .text 0xc0100000 0xc010bab4 0xc010baec
> > kdb>
> >
>


-------------------------------------------------------
This SF.Net email is sponsored by: New Crystal Reports XI.
Version 11 adds new functionality designed to reduce time involved in
creating, integrating, and deploying reporting solutions. Free runtime info,
new features, or free trial, at: http://www.businessobjects.com/devxi/728
_______________________________________________
ssic-linux-devel mailing list
ssic-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel

[prev in list] [next in list] [prev in thread] [next in thread]