Repeated UFS panics?

[Ivan Voras]

Hi,

Has anyone experienced frequent UFS panics on VMWare on fresh current?
I've started seeing ufs_dirbad panics almost daily. For a while I could
cause them simply by running make installworld to a freshly newfs-ed
file system - the panic message points to the *new* file system.
Unfortunately, kgdb cannot process the generated vmcores (though they
are on a separate file system that has never crashed). The repeatable
panics on new file systems have stopped when I rebuilt the kernel, but
now they are happening again, this time on my /usr/ports. I don't think
file system options have an influence - they first happened on a
"normal" UFS (noasync), and now they are happening on gjournaled, aync
mounted UFS.

Any ideas on how to provide more information on this? The panics usually
happen when I'm in X11, so I can't use the kernel debugger, but
sometimes they will happen when I'm in console.

I can, at least in theory, provide the VMWare machine for download, but
it's 10 GB+.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 250 bytes
Desc: OpenPGP digital signature
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20070826/174fb169/signature.pgp