Date: Mon, 17 Jan 2000 18:26:41 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: arnee <arnee@geocities.com> Cc: current@FreeBSD.ORG, Gerald Abshez <gerald@manhattanprojects.com> Subject: Re: NFS and Samba (file transfer crashes system) Message-ID: <200001180226.SAA18343@apollo.backplane.com> References: <387FDF2D.575DC537@geocities.com> <200001150528.VAA95955@apollo.backplane.com> <3883B0F4.60A1623F@geocities.com>
next in thread | previous in thread | raw e-mail | index | archive | help
:-----------------
:kernel.0 vmcore.0
:panic: ffs_blkfree: bad size
:
:syncing disks... 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
:giving up on 1 buffers
:Uptime: 26m34s
:
:dumping to dev #ad/0x20021, offset 864256
:dump ata2: resetting devices .. done
:(kgdb) where
:#0 0xc01372b0 in boot ( )
:#1 0xc0137634 in poweroff_wait ( )
:#2 0xc01ceee3 in ffs_blkfree ( )
:#3 0xc01d1335 in ffs_indirtrunc ( )
:#4 0xc01d0e8d in ffs_truncate ( )
:#5 0xc01d896e in ufs_setattr ( )
:#6 0xc01dac45 in ufs_vnoperate ( )
:#7 0xc0195f5c in nfsrv_setattr ( )
:#8 0xc01b0bd1 in nfssvc_nfsd ( )
:#9 0xc01b04eb in nfssvc ( )
:#10 0xc020ee26 in syscall ( )
:#11 0xc0201b06 n Xint0x80_syscall ( )
:#12 0x80480f9 in ?? ( )
:....
:#9 0xc018c039 in ip_output ( )
:#10 0xc0193060 in udp_output ( )
:#11 0xc0193298 in upd_send ( )
:#12 0xc0156183 in sosend ( )
:#13 0xc01ad56a in nfs_send ( )
:#14 0xc01b4d81 in nfssvc_nfsd ( )
:#15 0xc01b4477 in nfssvc ( )
:#16 0xc0215336 in syscall ( )
:#17 0xc0207606 in Xint0x80_syscall ( )
:#18 0x80480f0 in ?? ( )
:
:The last two panics have the same infos as "kernel.1 vmcore.1." So, I will be
:"cvsup" this afternoon, hoping things work, and yes I will add DEBUG-g option
:next time I compile the kernel :-). Also, please let me know if I am writing any
:redundant infos or if I should include any other infos from kgdb when posting a
:panic message. Sorry for the ignorance but I am still familiarizing with how
:things work in -current.
:
:arnee
Hmm. Ok, a couple of questions:
* Are you running softupdates on the UFS filesystem on the NFS server?
* Are the NFS mounts NFSv2 or NFSv3? TCP or UDP mount? (I assume udp by
the traces).
And now: You indicate that you are getting crashes both when you try to
do file operations via NFS, *AND* when you try to do file operations
via Samba. I presume that Samba has nothing to do with NFS so the
question becomes: Is it samba or NFS that is causing the problem?
If you turn off NFS and just drag-and-drop from the windows box can
you crash the machine? Now if you turn off Samba and turn on NFS
can you crash the machine with 'cp' from the NFS clients?
There is another possibility. You've indicated that you have had to
fsck quite often, sometimes losing files to corruption. It is possible
that *neither* Samba nor NFS are at fault and instead UFS has gotten
itself corrupted in a way that fsck cannot detect or fix.
The only way to know for sure is to backup the entire filesystem, wipe
it clean with newfs, and then restore it.
I would also look at your 'dmesg' output to see if you are getting any
IDE errors.
And there is yet another possibility: The new ATA driver has lots of
problems and as far as I can tell none of the reported problems have yet
been fixed. I would recommend going back to the 'wd' driver and seeing
if that helps.
So, in summary: You are giving us lots of great debugging information
but your system setup is complex and that makes it difficult to focus
in on the problem. We need to reproduce the panics in a less complex
setup to figure out what is causing them.
-Matt
Matthew Dillon
<dillon@backplane.com>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200001180226.SAA18343>