Re: rm -r sometimes produces errors under NFS

[Vincent Lefevre]

On 2007-03-10 12:41:27 +0100, Jim Meyering wrote:
> Vincent Lefevre <address@hidden> wrote:
> > On 2007-03-09 00:44:55 +0100, Jim Meyering wrote:
> >> Realize that for most people (everyone except you, afaik),
> >> rm works just fine.
> >
> > Yes, for most people, rm works fine. But the problem exists (I had
> > it on 3 different NFS servers in the past few years). And for your
> > information, other users have reported the same problem, e.g.
> >
> > http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=994291&admit=-682735245+1173400109463+28353475
> 
> That can't be relevant to any recent rewinddir-related change.

Yes, I noticed that: I got this problem with previous versions (and
various NFS servers) though it was quite rare in my case, but the
coreutils snapshot 6.8+ (containing the recent rewinddir-related
change) made things much worse.

> That report involves Redhat EL 3.0 U4 W, so they're probably
> using a Red Hat-patched coreutils-4.5.3.

Note that I also got the problem with the current Debian/testing.
So, it is not specific to Red Hat.

> The comments imply that at least one person there suspected that NFS
> caching was the root cause.

Yes, but IMHO, the coreutils behavior is suboptimal (even with other
file systems, in case of competing rm's). And I don't think that
ignoring NFS problems would be a good idea; it is well-known that
NFS caching is a problem (even when only one machine is involved),
in particular under Linux.

> You can avoid these "ENOENT" (No such file or directory) errors
> simply by using -f.

As I said, this is not a solution as the -f option will also remove
files without write permission, without any question.

> Perhaps it depends on a hard-to-reproduce NFS bug, too.
> That is why I am reluctant to make a significant change to solve
> your problem without first hearing that it affects more people.

You should make sure that the 6.8+ snapshot will be well tested, then.

Regards,

-- 
Vincent Lefèvre <address@hidden> - Web: <http://www.vinc17.org/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)