Re: [SCARED] Is ext2 unreliable?

From: Michael Gerdts (gerdts@cae.wisc.edu)
Date: Tue May 16 2000 - 15:05:08 EST


On Tue, May 16, 2000 at 10:21:59AM -0700, Linda Walsh wrote:
> Just tested lsof/fuser. VMware deletes an open file in /tmp
> that it uses for some purpose while it is up. An fuser on /tmp gives:

This is really a common practice for things that need temporary files, or
so it seems from tmpfile(3S) on Solaris:

SYNOPSIS
     #include <stdio.h>

     FILE *tmpfile(void);

DESCRIPTION
     The tmpfile() function creates a temporary file and opens a
     corresponding stream. The file will automatically be
     deleted when all references to the file are closed. The
     file is opened as in fopen(3S) for update (w+).

This does seem to be a ways away from what the original poster was
complaining about... As I see it the issue with his (or was it her...)
problem was that there were no running processes that should have been
using the partition, and as such the kernel should have cleaned up any file
descriptors that were left over.

If the non knfsd portion of the kernel-based NFS server had files opened,
should lsof or fuser be able to see that? If a system gets into this
state, is it possible to "rmmod nfsd"? Does that make it so that the file
system can be unmounted?

It seems as though this may be a question worth asking over on the nfs list
too. Archives and subscription information are available at
http://lists.sourceforge.net/mailman/listinfo/nfs/.

Mike

-
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 : Tue May 23 2000 - 21:00:11 EST