Can you post the client log files? Avati On Tue, Jun 28, 2011 at 7:42 PM, Pat Haley <phaley at mit.edu> wrote: > Hi Anand Babu Periasamy, > > Yes, I can read > > > /gdata/projects/philex/OAG/**2011/May16/mdor3km07/coast_**den2.in<http://coast_den2.in> > > directly from the backend disk (specifically: logged onto > machine containing the disk and did a more). > > Furthermore I can read (i.e. "more") that file other nodes > in our cluster going through gluster. > > Following up on another lead from this list, the TSM software > is not using direct i/o (O_DIRECT). > > > On Mon, Jun 27, 2011 at 12:08 PM, Pat Haley <phaley at mit.edu <mailto: >> phaley at mit.edu>> wrote: >> >> >> Hi >> >> We have been trying back-up a glusterfs (v3.1.4) area using >> the Tivoli TSM software to an off-site area. The back-up >> keeps failing with the following typical error messages >> >> 06/14/2011 22:22:58 ANS1587W I/O error reading file attributes for: >> /gdata/projects/philex/OAG/__**2011/May16/mdor3km10/coast___**den2.in<http://coast___den2.in> >> <http://coast_den2.in>. errno = 22, Invalid argument >> >> >> 06/14/2011 22:22:59 ANS4007E Error processing >> '/gdata/projects/philex/OAG/__**2011/May16/mdor3km07/coast___**den2.in<http://coast___den2.in> >> <http://coast_den2.in>': access to the object is denied >> >> >> We originally thought the problem was bad meta-pointer files, >> but we still experience the failures after deleting all meta-pointer >> files. >> >> Has anyone else experienced/solved this problem? >> >> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=**__-=-=-=-=-=-=-=-=-=-=-=-=-=-=** >> -=__-=-=- >> Pat Haley Email: phaley at mit.edu >> <mailto:phaley at mit.edu> >> >> Center for Ocean Engineering Phone: (617) 253-6824 >> <tel:%28617%29%20253-6824> >> >> Dept. of Mechanical Engineering Fax: (617) 253-8125 >> <tel:%28617%29%20253-8125> >> >> MIT, Room 5-222B http://web.mit.edu/phaley/www/ >> 77 Massachusetts Avenue >> Cambridge, MA 02139-4301 >> ______________________________**___________________ >> Gluster-users mailing list >> Gluster-users at gluster.org <mailto:Gluster-users at gluster.**org<Gluster-users at gluster.org> >> > >> >> http://gluster.org/cgi-bin/__**mailman/listinfo/gluster-users<http://gluster.org/cgi-bin/__mailman/listinfo/gluster-users> >> <http://gluster.org/cgi-bin/**mailman/listinfo/gluster-users<http://gluster.org/cgi-bin/mailman/listinfo/gluster-users> >> **> >> >> >> >> Hi Pat, Let me get some one from the engineering team to coordinate with >> you. For now, you can check this: >> >> '/gdata/projects/philex/OAG/__**2011/May16/mdor3km07/coast___**den2.in<http://coast___den2.in>< >> http://coast_den2.in/> >> >> This file must exist on one of the backend disks. See if you can directly >> read from the backend disks. If it is not a bad backend disk, then it is a >> GlusterFS issue. >> >> If you provide us remote access via reverse-ssh / screen or webex, we can >> quickly narrow down. We don't have a TSM environment in our labs. You can >> help us validate compatibility. >> Also I am not sure if TSM allows scripting. If we can make TSM tools use >> GlusterFS 3.2 marker APIs, then you can achieve continuous backup >> functionality (backup data as and when it changes, without crawling the name >> space). It will significantly improve your backup performance (no more >> backup windows). Marker APIs can even be invoked via shell scripting using >> attr tools. >> >> -- >> Anand Babu Periasamy >> Blog [http://www.unlocksmith.org <http://www.unlocksmith.org/>] >> >> >> Imagination is more important than knowledge --Albert Einstein >> > > > -- > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=**-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=**-=-=- > Pat Haley Email: phaley at mit.edu > Center for Ocean Engineering Phone: (617) 253-6824 > Dept. of Mechanical Engineering Fax: (617) 253-8125 > MIT, Room 5-222B http://web.mit.edu/phaley/www/ > 77 Massachusetts Avenue > Cambridge, MA 02139-4301 > ______________________________**_________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://gluster.org/cgi-bin/**mailman/listinfo/gluster-users<http://gluster.org/cgi-bin/mailman/listinfo/gluster-users> > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://gluster.org/pipermail/gluster-users/attachments/20110628/bae21fdc/attachment-0001.htm>
