[OpenAFS] possible microsoft excel 2010 corruption issue on openafs
volumes
info@probanet.it
info@probanet.it
Wed, 14 Dec 2011 10:36:26 +0100
--=_469a2307fc70e3fe4b355e402ca92450
Content-Type: text/plain;
charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Hello!
We're experiencing the same issue on a large number of volumes=0D
(around
28000, just finished upgrading fileserver to 1.6.0 in order to use=0D
DAFS,
using Ubuntu 11.10 x86_64). Our volumes doesn't have a .backup, just=0D
RW
+ RO (1 cheap-replica and 1 remote replica). Where can we find a=0D
patch
and how should we use it? Thanks for your help! :)
Stefano
Fabio=0D
On Wed, 7 Dec 2011 19:23:50 -0500, Derrick Brashear wrote: there's an=0D
issue in the 1.6.0 fileserver but if you are also having it on 1.4.14=0D
that's new=20
Derrick =20
On Dec 7, 2011, at 19:19, Jonathan Nilsson wrote:
=0D
Hello, We've had some strange file corruption and missing data issues=0D
with Microsoft Excel 2010 on a Windows 7 OpenAFS 1.5.78 client. 1) A use=
r=0D
creates an Excel file on their local drive, then copies it to an AFS RW=0D
mount point (via mapped network drive). 2) several clients read the file=0D
(no changes made) 3) at some point the file is no longer able to be opene=
d,=0D
or it opens but warns of corruption and missing data When I recover the=0D
file from the backup on the day of the file's modification timestamp, the=
=0D
restored file is fine. Curiously, the restored file appears identical to=0D
the corrupted file: the file size (bytes) and timestamp are identical. T=
he=0D
fileservers housing volumes where this has happened are: CentOS 6.0=0D
2.6.32-71.29.1.el6.x86_64 with OpenAFS 1.6.0 RHEL 6.1=0D
2.6.32-131.2.1.el6.x86_64 with OpenAFS 1.4.14 We have upgraded the clien=
t=0D
to OpenAFS 1.7.2 (we did this just before the 1.7.3 announcement) hoping =
to=0D
fix the issue if it is a bug... Is there any known issue with 1.5.78 clie=
nt=0D
that could have caused corruption like this? --=20
Jonathan.Nilsson at uci=0D
dot edu Social Sciences Computing Services SSPB 1265 | 949.824.1536 =20
=0D
Links:
------
[1] mailto:jnilsson@uci.edu
--=_469a2307fc70e3fe4b355e402ca92450
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
charset=UTF-8
<p>Hello!<br /> We're experiencin=
g the same issue on a large number of volumes (around<br />28000, just fini=
shed upgrading fileserver to 1.6.0 in order to use DAFS,<br />using Ubuntu =
11.10 x86_64). Our volumes doesn't have a .backup, just RW<br />+ RO (1 che=
ap-replica and 1 remote replica). Where can we find a patch<br />and how sh=
ould we use it? Thanks for your help! :)<br /><br />Stefano<br />Fabio</p>
<p> </p>
<p>On Wed, 7 Dec 2011 19:23:50 -0500, Derrick Brashear <shadow@gmail.com=
> wrote:</p>
<blockquote style=3D"padding-left: 5px; border-left: 2px solid #1010ff; mar=
gin-left: 5px; width: 100%;"><!-- html ignored --><!-- head ignored --><!--=
meta ignored -->
<div>there's an issue in the 1.6.0 fileserver but if you are also having it=
on 1.4.14 that's new</div>
<div><br />Derrick
<div></div>
</div>
<div><br />On Dec 7, 2011, at 19:19, Jonathan Nilsson <<a href=3D"mailto=
:jnilsson@uci.edu">jnilsson@uci.edu</a>> wrote:<br /><br /></div>
<blockquote>
<div>Hello,
<div></div>
<div>We've had some strange file corruption and missing data issues with Mi=
crosoft Excel 2010 on a Windows 7 OpenAFS 1.5.78 client.</div>
<div></div>
<div>1) A user creates an Excel file on their local drive, then copies it t=
o an AFS RW mount point (via mapped network drive).</div>
<div>2) several clients read the file (no changes made)</div>
<div>3) at some point the file is no longer able to be opened, or it opens =
but warns of corruption and missing data</div>
<div></div>
<div>When I recover the file from the backup on the day of the file's modif=
ication timestamp, the restored file is fine. Curiously, the restored file =
appears identical to the corrupted file: the file size (bytes) and timestam=
p are identical.</div>
<div></div>
<div>The fileservers housing volumes where this has happened are:</div>
<div> CentOS 6.0 2.6.32-71.29.1.el6.x86_64 with OpenAFS 1.6.0</div>
<div> RHEL 6.1 2.6.32-131.2.1.el6.x86_64 with OpenAFS 1.4.14</div>
<div></div>
<div>We have upgraded the client to OpenAFS 1.7.2 (we did this just before =
the 1.7.3 announcement) hoping to fix the issue if it is a bug... Is t=
here any known issue with 1.5.78 client that could have caused corruption l=
ike this?</div>
<div>
<div></div>
-- <br />
<div><span style=3D"font-family: 'courier new',monospace;">Jonathan.Nilsson=
at uci dot edu</span></div>
<div><span style=3D"font-family: 'courier new',monospace;">Social Sciences =
Computing Services</span></div>
<div><span>SSPB 1265 | 949.824.1536</span></div>
</div>
</div>
</blockquote>
</blockquote>
<p> </p>
--=_469a2307fc70e3fe4b355e402ca92450--