gluster-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gluster-devel] Bug Reports


From: Krishna Srinivas
Subject: Re: [Gluster-devel] Bug Reports
Date: Tue, 10 Feb 2009 15:23:00 +0530

> Is there a way to specify --disable-direct-io-mode in fstab?


 volume fuse
   type mount/fuse
   option direct-io-mode 1
   option entry-timeout 1
   option attr-timeout 1
   option mount-point /mnt/storage
   subvolumes unify
 end-volume

you can give the options to fuse in the vol file.

On Tue, Feb 10, 2009 at 1:32 AM, Gordan Bobic <address@hidden> wrote:
> OK, just managed to clear this weird condition and return things to some
> appearance of normality. Stopping all the servers and explicitly deleting
> the offending symling/directory (it seems to have splitbrained - some
> servers had the symling (which wasn't viewable!) and others had the
> directory (which was also not viewable). Deleting them all from the backing
> store, and remounting seems to have fixed the problem. Really weird how it
> split-brained while all the servers were up, though. :-/


It should not happened. If you get any clues on the reproducing steps
it will help us.

Krishna


>
> Gordan Bobic wrote:
>>
>> And another one - and this one is quite WTF-worthy. It may be the one that
>> someone (Filipe, maybe?) reported and/or submitted a patch for recently, but
>> I'm not sure.
>>
>> OpenOffice seems to have decided to fall over when /home is mounted on
>> GlusterFS. It was working OK on qa1, but with rc1 it just dies. Removing
>> the ~/.openoffice.org2.0 directory makes things even worse because it
>> doesn't get re-created properly - some directories end up being created
>> as root, which is a neat trick consider OOo is running as an ordinary
>> user. Those directories also get created with 000 permissions, which is
>> pretty wrong, since directories have to be executable. This works correctly
>> if ~/ is on a non-gluster FS.
>>
>> This is where it gets _really_ weird. If I make a directory
>> /tmp/.openoffice.org2.0 and try to symlink it to ~/.openoffice.org2.0, what
>> happens instead is a _real_ directory ~/.openoffice.org2.0 appears, and the
>> symlink gets created inside it. Where that came from I have no idea. So, I
>> delete the said directory, and go to the backing store and create the
>> symlink there. That works fine, but then, open restarting all the servers
>> there is no directory nor is there a symlink when viewed with ls! But - if I
>> name the full directory (/home/gordan/.glusterfs.org2.0/), I can cd into it,
>> and the content is there, as OOo created it, in /tmp!
>>
>> And now my head really hurts. It seems that there is something quite
>> thoroughly broken when it comes to symlink and directory handling. :(
>>
>> But at least the glusterfs daemon notices that something is wrong:
>> 2009-02-09 19:45:29 W
>> [afr-self-heal-common.c:871:sh_missing_entries_lookup_cbk] home: path
>> /gordan/.openoffice.org2.0 on subvolume home1 => -1 (No such file or
>> directory)
>> 2009-02-09 19:45:29 E
>> [afr-self-heal-common.c:787:sh_missing_entries_create] home: conflicing
>> filetypes exist for path /gordan/.openoffice.org2.0. returning.
>> 2009-02-09 19:45:29 W
>> [afr-self-heal-common.c:436:afr_sh_missing_entries_done] home: aborting
>> selfheal of /gordan/.openoffice.org2.0
>>
>> Gordan
>>
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]