Hi Kevin,

Thank you for reporting this issue and for the further information you
provide.

On Tue, 2006-09-05 at 16:46 -0400, kcheek wrote: 
> To follow up some more, I have a few suggestions for things that could
> be done to help resolve the problems I encountered:
> 
>   - mondoarchive should check if filesystems are read-only before
>   trying to use them for scratch or temp

Hm, interesting point. I thought it did but might be wrong. Need to look
into this.

>   - mondoarchive should exclude afs filesystems when option "-N" is
>   used

Agreed & possible with relatively little effort. Obviously not specific
to AFS, so everyone would benefit.

>   - mondoarchive should exclude paths from the "find" command ("Making
>   catalog of files to be backed up") if the path has been excluded
>   with the "-E" option

The behaviour you described in your previous message does indeed appear
to be wrong, so this would need fixing. Obviously not specific to AFS,
so everyone would benefit.

>   - mondoarchive should prefer /tmp and/or /var/tmp over other paths
>   for temp and scratch (if there is enough space) - maybe even prompt
>   the user if insufficient space is available in /tmp or /var/tmp

Even more so, it should behave in an FHS compliant manner - see here for
more details: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=312546.

>   - mondoarchive should try to avoid network filesystems for temp and
>   scratch (give preference to local filesystems if there is enough
>   space)

Yes, agreed and it does that for other network filesystems, so this
should be relatively easy to fix.

> -Kevin

It's about eight years since I last used AFS. ;-)

What does the output of mount say on your system, what is in fstab?
And/or how do you mount an AFS filesystem? Would you be aware of any
publicly accessible AFS volumes that I could mount for test purposes?

Best regards & thanks a lot,
Andree
-- 
Andree Leidenfrost
@ Debian Developer
Sydney - Australia

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to