I am building a system with a redundant MDS, that is two MDS sharing a set of disks, one being Active, the other Standby. If I put the MGS and MDS on the same system, it appears that they must be on the same partition as well. Otherwise, when there is a failover, the MGS will not fail over. Is that true? Assuming that it is true, then I mount the MGS/MDT partition first, then all the OSTs. When I unmount, I usually want to unmount the MDT prior to unmounting the OSTs, because the MDT is a client of the OSTs. Is that possible here? In other words, I want to stop the MDT service first, then unmount the OSTs, then unmount the MGS. How can I stop the MDT service? If I don''t do this, that is, if I just unmount the OSTs followed by the MGS/MDT, then I get errors like this: Lustre: raid6-OST0002-osc: Connection to service raid6-OST0002 via nid 10.2.46.74 at o2ib <mailto:10.2.46.74 at o2ib> was lost; in progress operations using this service will wait for recovery to complete. . . . LustreError: 8964:0:(lov_obd.c:418:lov_disconnect_obd()) Target raid6-OST0000_UUID disconnect error -5 LustreError: 8964:0:(lov_obd.c:418:lov_disconnect_obd()) Target raid6-OST0002_UUID disconnect error -5 Lustre: Request x105 sent from raid6-OST0003-osc to NID 10.2.46.75 at o2ib <mailto:10.2.46.75 at o2ib> 5s ago has timed out (limit 5s). . . . Lustre: MGS has stopped. Lustre: Mount still busy with 7 refs, waiting for 330 secs... So, my question is: Is there a way to stop just the MDT function PRIOR to unmounting the OSTs? Thanks. -Roger Roger Spellman Staff Engineer Terascala, Inc. 508-588-1501 www.terascala.com <http://www.terascala.com/> -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.lustre.org/pipermail/lustre-discuss/attachments/20080910/d3d42618/attachment.html
On Wed, 2008-09-10 at 16:23 -0400, Roger Spellman wrote:> I am building a system with a redundant MDS, that is two MDS sharing a > set of disks, one being Active, the other Standby. > > If I put the MGS and MDS on the same system, it appears that they must > be on the same partition as well.No.> Otherwise, when there is a failover, the MGS will not fail over. Is > that true?Not true. b. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.lustre.org/pipermail/lustre-discuss/attachments/20080910/1225d738/attachment.bin
Brian J. Murrell wrote:> On Wed, 2008-09-10 at 16:23 -0400, Roger Spellman wrote: >> I am building a system with a redundant MDS, that is two MDS sharing a >> set of disks, one being Active, the other Standby. >> >> If I put the MGS and MDS on the same system, it appears that they must >> be on the same partition as well. > > No. > >> Otherwise, when there is a failover, the MGS will not fail over. Is >> that true?If the MDT and MGT are separate partitions, then you will have to fail them over as separate services, as each partition will be mounted separately. The separate partitions can be on one system. Of course any decent HA tool will allow you to failover multiple services with one action. Should note- the MGS is very small, and only used for configuration changes, and mount information. If all your clients are already mounted, an MGS failure is quite transparent - you can run for quite some time with a dead MGS. For a very robust system, I would suggest moving the MGS to a small machine (heck, a cheap laptop would work for all but the biggest sites) replicate the MGT disk and put your failover dollars on the MGS. You could build a very robust failover MGS for the cost of two cheap whitebox PC''s (modulo network hardware cost). Also, a separate MGS is recomended when you have more than one filesystem. cliffw> > Not true. > > b. > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss at lists.lustre.org > http://lists.lustre.org/mailman/listinfo/lustre-discuss