I've been getting processes SIGBUS'd on my root-NFS'd firewall. The
cause seems to be NFS related or knfsd related. I've tracked the
problem throughout the mm layer on the client by following printk's:
filemap_nopage: readpage 1 (c0046c24) failed
do_no_page: vma->vm_ops->nopage (c00290d4) failed
do_page_fault: killing sshd(560)
System.map indicates that these addresses are:
c00290d4 t filemap_nopage
c0046c24 t return_EIO
So the inode has been marked bad.
I've been able to reproduce it by keeping the NFS server busy while the
client does a recursive ls on an NFS mount from this server, and
repeatidly doing 'ssh client ps aux' until it breaks.
Once it breaks, then that's it - it will always send a SIGBUS to the
processes, and the only way out seems to be a reboot of the client.
The server is running 2.2.13, knfsd, with the nfs-utils-0.1.5 package.
I've recently upgraded the client to 2.2.14, and noticed this problem.
Tomorrow, I'll try backing down the client to 2.2.13 and re-testing in
the same way (update will follow).
I have a feeling though that 2.2.13 will behave similarly, which would
point to knfsd as the cause.
_____
|_____| ------------------------------------------------- ---+---+-
| | Russell King rmk@arm.linux.org.uk --- ---
| | | | http://www.arm.linux.org.uk/~rmk/aboutme.html / / |
| +-+-+ --- -+-
/ | THE developer of ARM Linux |+| /|\
/ | | | --- |
+-+-+ ------------------------------------------------- /\\\ |
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Jan 15 2000 - 21:00:13 EST