Daniel Leaberry
2007-Feb-11 10:25 UTC
[Lustre-discuss] 1.6b7 can''t mount more than one ost
I''m running the 64bit lustre rpms on Centos 4u4 over a channel bonded (mode 0) pair of interfaces. After formatting the MDS I bring it up following the directions on the mountconf wiki. It comes up and looks fine. I then bring up one of my ost''s (doesn''t matter which one, I''ve tried them all.) and it connects just fine as well. I can mount a client and see my filesystem consisting of one ost. Any attempt to bring up a second ost results in the following message. mount.lustre: mount /dev/sdd at /var/mnt/ost02 failed: Operation already in progress The target service is already running. (/dev/sdd) dmesg on the mds shows this LustreError: The config log for lustre01-OST0000 already exists, yet the server claims it never registered. It may have been reformatted, or the index changed. writeconf the MDT to regenerate all logs. LustreError: 25772:0:(mgs_llog.c:1699:mgs_write_log_target()) Can''t write logs for lustre01-OST0000 (-114) LustreError: 25772:0:(mgs_handler.c:429:mgs_handle_target_reg()) Failed to write lustre01-OST0000 log (-114) LustreError: 25772:0:(mgs_handler.c:551:mgs_handle()) MGS handle cmd=253 rc=-114 LustreError: 25772:0:(ldlm_lib.c:1349:target_send_reply_msg()) @@@ processing error (-114) req@00000104247fd850 x434/t0 o253->4109e143-5a1a-bcb6-5ce6-bab2b7a2cc98@192.168.101.11@tcp:-1 lens 4672/4672 ref 0 fl Interpret:/0/0 rc 0/0 LustreError: 25871:0:(lov_obd.c:479:lov_add_target()) UUID lustre01-OST0000_UUID already assigned at LOV target index 0 LustreError: 25871:0:(obd_config.c:1021:class_config_llog_handler()) Err -17 on cfg command: Lustre: cmd=cf00d 0:lustre01-mdtlov 1:lustre01-OST0000_UUID 2:0 3:1 dmesg on the oss shows this kjournald starting. Commit interval 5 seconds LDISKFS FS on sdd, internal journal LDISKFS-fs: mounted filesystem with ordered data mode. LDISKFS-fs: file extents enabled LDISKFS-fs: mballoc enabled LustreError: 31514:0:(client.c:571:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err == -114 req@0000010037d3d400 x434/t0 o253->MGS@MGC192.168.101.31@tcp_0:26 lens 4672/4672 ref 1 fl Rpc:R/0/0 rc 0/-114 LustreError: 31514:0:(mgc_request.c:730:mgc_target_register()) register failed. rc=-114 LustreError: 31514:0:(obd_mount.c:948:server_register_target()) registration with the MGS failed (-114) LustreError: 31514:0:(obd_mount.c:1048:server_start_targets()) Required registration failed for lustre01-OSTffff: -114 LustreError: 31514:0:(obd_mount.c:1560:server_fill_super()) Unable to start targets: -114 LustreError: 31514:0:(mgc_request.c:150:config_log_find()) can''t get log lustre01-OSTffff LustreError: 31514:0:(obd_mount.c:1347:server_put_super()) no obd lustre01-OSTffff LustreError: 31514:0:(obd_mount.c:118:server_deregister_mount()) lustre01-OSTffff not registered LDISKFS-fs: mballoc: 0 blocks 0 reqs (0 success) LDISKFS-fs: mballoc: 0 extents scanned, 0 goal hits, 0 2^N hits, 0 breaks LDISKFS-fs: mballoc: 0 generated and it took 0 Lustre: server umount lustre01-OSTffff complete LustreError: 31514:0:(obd_mount.c:1911:lustre_fill_super()) Unable to mount (-114) It suggests writeconf''ing the logs. I''ve tried that dozens of times with the same result. It also says the registration with the MGS failed. I don''t know why since it obviously succeeded for the 1st ost on the same box. I''ve reformatted all the disks at least 5 times and I''ve reached a point where I have no idea why it won''t mount the second ost. Any insight would be appreciated. Thanks, Daniel
Hi, Ost will be able to be operated in one ? I was testing with MDS x1 OST x1. But, It has been freezed. How many are a minimum number of necessary OST? 2007/2/12, Daniel Leaberry <dleaberry@iarchives.com>:> I''m running the 64bit lustre rpms on Centos 4u4 over a channel bonded (mode 0) pair of interfaces. > > After formatting the MDS I bring it up following the directions on the mountconf wiki. It comes up and looks fine. I then bring up one of my ost''s (doesn''t matter which one, I''ve tried them all.) and it connects just fine as well. I can mount a client and see my filesystem consisting of one ost. Any attempt to bring up a second ost results in the following message. > > mount.lustre: mount /dev/sdd at /var/mnt/ost02 failed: Operation already in progress > The target service is already running. (/dev/sdd) > > dmesg on the mds shows this > LustreError: The config log for lustre01-OST0000 already exists, yet the server claims it never registered. It may have been reformatted, or the index changed. writeconf the MDT to regenerate all logs. > LustreError: 25772:0:(mgs_llog.c:1699:mgs_write_log_target()) Can''t write logs for lustre01-OST0000 (-114) > LustreError: 25772:0:(mgs_handler.c:429:mgs_handle_target_reg()) Failed to write lustre01-OST0000 log (-114) > LustreError: 25772:0:(mgs_handler.c:551:mgs_handle()) MGS handle cmd=253 rc=-114 > LustreError: 25772:0:(ldlm_lib.c:1349:target_send_reply_msg()) @@@ processing error (-114) req@00000104247fd850 x434/t0 o253->4109e143-5a1a-bcb6-5ce6-bab2b7a2cc98@192.168.101.11@tcp:-1 lens 4672/4672 ref 0 fl Interpret:/0/0 rc 0/0 > LustreError: 25871:0:(lov_obd.c:479:lov_add_target()) UUID lustre01-OST0000_UUID already assigned at LOV target index 0 > LustreError: 25871:0:(obd_config.c:1021:class_config_llog_handler()) Err -17 on cfg command: > Lustre: cmd=cf00d 0:lustre01-mdtlov 1:lustre01-OST0000_UUID 2:0 3:1 > > dmesg on the oss shows this > kjournald starting. Commit interval 5 seconds > LDISKFS FS on sdd, internal journal > LDISKFS-fs: mounted filesystem with ordered data mode. > LDISKFS-fs: file extents enabled > LDISKFS-fs: mballoc enabled > LustreError: 31514:0:(client.c:571:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err == -114 req@0000010037d3d400 x434/t0 o253->MGS@MGC192.168.101.31@tcp_0:26 lens 4672/4672 ref 1 fl Rpc:R/0/0 rc 0/-114 > LustreError: 31514:0:(mgc_request.c:730:mgc_target_register()) register failed. rc=-114 > LustreError: 31514:0:(obd_mount.c:948:server_register_target()) registration with the MGS failed (-114) > LustreError: 31514:0:(obd_mount.c:1048:server_start_targets()) Required registration failed for lustre01-OSTffff: -114 > LustreError: 31514:0:(obd_mount.c:1560:server_fill_super()) Unable to start targets: -114 > LustreError: 31514:0:(mgc_request.c:150:config_log_find()) can''t get log lustre01-OSTffff > LustreError: 31514:0:(obd_mount.c:1347:server_put_super()) no obd lustre01-OSTffff > LustreError: 31514:0:(obd_mount.c:118:server_deregister_mount()) lustre01-OSTffff not registered > LDISKFS-fs: mballoc: 0 blocks 0 reqs (0 success) > LDISKFS-fs: mballoc: 0 extents scanned, 0 goal hits, 0 2^N hits, 0 breaks > LDISKFS-fs: mballoc: 0 generated and it took 0 > Lustre: server umount lustre01-OSTffff complete > LustreError: 31514:0:(obd_mount.c:1911:lustre_fill_super()) Unable to mount (-114) > > > > It suggests writeconf''ing the logs. I''ve tried that dozens of times with the same result. It also says the registration with the MGS failed. I don''t know why since it obviously succeeded for the 1st ost on the same box. I''ve reformatted all the disks at least 5 times and I''ve reached a point where I have no idea why it won''t mount the second ost. Any insight would be appreciated. > > Thanks, > Daniel > > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss@clusterfs.com > https://mail.clusterfs.com/mailman/listinfo/lustre-discuss > >
JC.LAFOUCRIERE@CEA.FR
2007-Feb-11 14:08 UTC
[Lustre-discuss] 1.6b7 can''t mount more than one ost
Hello, there is a bug in 1.6b7 with the size of the lustre fs name (bugzilla 11564) you cannot use a name longer than 7 char and lustre01 is 8 try with a smaller name, it should work JCL -----Original Message----- From: lustre-discuss-bounces@clusterfs.com on behalf of Daniel Leaberry Sent: Sun 2/11/2007 6:24 PM To: lustre-discuss@clusterfs.com Subject: [Lustre-discuss] 1.6b7 can''t mount more than one ost I''m running the 64bit lustre rpms on Centos 4u4 over a channel bonded (mode 0) pair of interfaces. After formatting the MDS I bring it up following the directions on the mountconf wiki. It comes up and looks fine. I then bring up one of my ost''s (doesn''t matter which one, I''ve tried them all.) and it connects just fine as well. I can mount a client and see my filesystem consisting of one ost. Any attempt to bring up a second ost results in the following message. mount.lustre: mount /dev/sdd at /var/mnt/ost02 failed: Operation already in progress The target service is already running. (/dev/sdd) dmesg on the mds shows this LustreError: The config log for lustre01-OST0000 already exists, yet the server claims it never registered. It may have been reformatted, or the index changed. writeconf the MDT to regenerate all logs. LustreError: 25772:0:(mgs_llog.c:1699:mgs_write_log_target()) Can''t write logs for lustre01-OST0000 (-114) LustreError: 25772:0:(mgs_handler.c:429:mgs_handle_target_reg()) Failed to write lustre01-OST0000 log (-114) LustreError: 25772:0:(mgs_handler.c:551:mgs_handle()) MGS handle cmd=253 rc=-114 LustreError: 25772:0:(ldlm_lib.c:1349:target_send_reply_msg()) @@@ processing error (-114) req@00000104247fd850 x434/t0 o253->4109e143-5a1a-bcb6-5ce6-bab2b7a2cc98@192.168.101.11@tcp:-1 lens 4672/4672 ref 0 fl Interpret:/0/0 rc 0/0 LustreError: 25871:0:(lov_obd.c:479:lov_add_target()) UUID lustre01-OST0000_UUID already assigned at LOV target index 0 LustreError: 25871:0:(obd_config.c:1021:class_config_llog_handler()) Err -17 on cfg command: Lustre: cmd=cf00d 0:lustre01-mdtlov 1:lustre01-OST0000_UUID 2:0 3:1 dmesg on the oss shows this kjournald starting. Commit interval 5 seconds LDISKFS FS on sdd, internal journal LDISKFS-fs: mounted filesystem with ordered data mode. LDISKFS-fs: file extents enabled LDISKFS-fs: mballoc enabled LustreError: 31514:0:(client.c:571:ptlrpc_check_status()) @@@ type == PTL_RPC_MSG_ERR, err == -114 req@0000010037d3d400 x434/t0 o253->MGS@MGC192.168.101.31@tcp_0:26 lens 4672/4672 ref 1 fl Rpc:R/0/0 rc 0/-114 LustreError: 31514:0:(mgc_request.c:730:mgc_target_register()) register failed. rc=-114 LustreError: 31514:0:(obd_mount.c:948:server_register_target()) registration with the MGS failed (-114) LustreError: 31514:0:(obd_mount.c:1048:server_start_targets()) Required registration failed for lustre01-OSTffff: -114 LustreError: 31514:0:(obd_mount.c:1560:server_fill_super()) Unable to start targets: -114 LustreError: 31514:0:(mgc_request.c:150:config_log_find()) can''t get log lustre01-OSTffff LustreError: 31514:0:(obd_mount.c:1347:server_put_super()) no obd lustre01-OSTffff LustreError: 31514:0:(obd_mount.c:118:server_deregister_mount()) lustre01-OSTffff not registered LDISKFS-fs: mballoc: 0 blocks 0 reqs (0 success) LDISKFS-fs: mballoc: 0 extents scanned, 0 goal hits, 0 2^N hits, 0 breaks LDISKFS-fs: mballoc: 0 generated and it took 0 Lustre: server umount lustre01-OSTffff complete LustreError: 31514:0:(obd_mount.c:1911:lustre_fill_super()) Unable to mount (-114) It suggests writeconf''ing the logs. I''ve tried that dozens of times with the same result. It also says the registration with the MGS failed. I don''t know why since it obviously succeeded for the 1st ost on the same box. I''ve reformatted all the disks at least 5 times and I''ve reached a point where I have no idea why it won''t mount the second ost. Any insight would be appreciated. Thanks, Daniel _______________________________________________ Lustre-discuss mailing list Lustre-discuss@clusterfs.com https://mail.clusterfs.com/mailman/listinfo/lustre-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.clusterfs.com/pipermail/lustre-discuss/attachments/20070211/a4fa5171/attachment.html
Nathaniel Rutman
2007-Feb-12 15:08 UTC
[Lustre-discuss] 1.6b7 can''t mount more than one ost
JC.LAFOUCRIERE@CEA.FR wrote:> > Hello, > > there is a bug in 1.6b7 with the size of the lustre fs name (bugzilla > 11564) > you cannot use a name longer than 7 char and lustre01 is 8 > try with a smaller name, it should work > > JCL >I''ve attached a fix to this bugzilla ticket.> > > -----Original Message----- > From: lustre-discuss-bounces@clusterfs.com on behalf of Daniel Leaberry > Sent: Sun 2/11/2007 6:24 PM > To: lustre-discuss@clusterfs.com > Subject: [Lustre-discuss] 1.6b7 can''t mount more than one ost > > I''m running the 64bit lustre rpms on Centos 4u4 over a channel bonded > (mode 0) pair of interfaces. > > After formatting the MDS I bring it up following the directions on the > mountconf wiki. It comes up and looks fine. I then bring up one of my > ost''s (doesn''t matter which one, I''ve tried them all.) and it connects > just fine as well. I can mount a client and see my filesystem > consisting of one ost. Any attempt to bring up a second ost results in > the following message. > > mount.lustre: mount /dev/sdd at /var/mnt/ost02 failed: Operation > already in progress > The target service is already running. (/dev/sdd) > > dmesg on the mds shows this > LustreError: The config log for lustre01-OST0000 already exists, yet > the server claims it never registered. It may have been reformatted, > or the index changed. writeconf the MDT to regenerate all logs. > LustreError: 25772:0:(mgs_llog.c:1699:mgs_write_log_target()) Can''t > write logs for lustre01-OST0000 (-114) > LustreError: 25772:0:(mgs_handler.c:429:mgs_handle_target_reg()) > Failed to write lustre01-OST0000 log (-114) > LustreError: 25772:0:(mgs_handler.c:551:mgs_handle()) MGS handle > cmd=253 rc=-114 > LustreError: 25772:0:(ldlm_lib.c:1349:target_send_reply_msg()) @@@ > processing error (-114) req@00000104247fd850 x434/t0 > o253->4109e143-5a1a-bcb6-5ce6-bab2b7a2cc98@192.168.101.11@tcp:-1 lens > 4672/4672 ref 0 fl Interpret:/0/0 rc 0/0 > LustreError: 25871:0:(lov_obd.c:479:lov_add_target()) UUID > lustre01-OST0000_UUID already assigned at LOV target index 0 > LustreError: 25871:0:(obd_config.c:1021:class_config_llog_handler()) > Err -17 on cfg command: > Lustre: cmd=cf00d 0:lustre01-mdtlov 1:lustre01-OST0000_UUID 2:0 3:1 > > dmesg on the oss shows this > kjournald starting. Commit interval 5 seconds > LDISKFS FS on sdd, internal journal > LDISKFS-fs: mounted filesystem with ordered data mode. > LDISKFS-fs: file extents enabled > LDISKFS-fs: mballoc enabled > LustreError: 31514:0:(client.c:571:ptlrpc_check_status()) @@@ type == > PTL_RPC_MSG_ERR, err == -114 req@0000010037d3d400 x434/t0 > o253->MGS@MGC192.168.101.31@tcp_0:26 lens 4672/4672 ref 1 fl Rpc:R/0/0 > rc 0/-114 > LustreError: 31514:0:(mgc_request.c:730:mgc_target_register()) > register failed. rc=-114 > LustreError: 31514:0:(obd_mount.c:948:server_register_target()) > registration with the MGS failed (-114) > LustreError: 31514:0:(obd_mount.c:1048:server_start_targets()) > Required registration failed for lustre01-OSTffff: -114 > LustreError: 31514:0:(obd_mount.c:1560:server_fill_super()) Unable to > start targets: -114 > LustreError: 31514:0:(mgc_request.c:150:config_log_find()) can''t get > log lustre01-OSTffff > LustreError: 31514:0:(obd_mount.c:1347:server_put_super()) no obd > lustre01-OSTffff > LustreError: 31514:0:(obd_mount.c:118:server_deregister_mount()) > lustre01-OSTffff not registered > LDISKFS-fs: mballoc: 0 blocks 0 reqs (0 success) > LDISKFS-fs: mballoc: 0 extents scanned, 0 goal hits, 0 2^N hits, 0 breaks > LDISKFS-fs: mballoc: 0 generated and it took 0 > Lustre: server umount lustre01-OSTffff complete > LustreError: 31514:0:(obd_mount.c:1911:lustre_fill_super()) Unable to > mount (-114) > > > > It suggests writeconf''ing the logs. I''ve tried that dozens of times > with the same result. It also says the registration with the MGS > failed. I don''t know why since it obviously succeeded for the 1st ost > on the same box. I''ve reformatted all the disks at least 5 times and > I''ve reached a point where I have no idea why it won''t mount the > second ost. Any insight would be appreciated. > > Thanks, > Daniel > > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss@clusterfs.com > https://mail.clusterfs.com/mailman/listinfo/lustre-discuss > > > ------------------------------------------------------------------------ > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss@clusterfs.com > https://mail.clusterfs.com/mailman/listinfo/lustre-discuss >