Description of problem: Created a mirror: [root@link-13 ~]# lvs -a -o+devices LV VG Attr LSize Origin Snap% Move Log Copy% Devices mirror_vol mirror_vg mwi-ao 100.00G mirror_vol_mlog 0.00 mirror_vol_mimage_0(0),mirror_vol_mimage_1(0) [mirror_vol_mimage_0] mirror_vg iwi-ao 100.00G /dev/sdc1(0) [mirror_vol_mimage_1] mirror_vg iwi-ao 100.00G /dev/sdb1(0) [mirror_vol_mlog] mirror_vg lwi-ao 4.00M /dev/sdb2(0) Started IO from all nodes in a four node cluster before the mirror was in sync. Observed this in the logs: Jul 20 21:11:50 link-13 kernel: device-mapper: Reason :: 1 Jul 20 21:11:50 link-13 kernel: device-mapper: unable to get server (1) to mark region (1713) Jul 20 21:11:50 link-13 kernel: device-mapper: Reason :: 1 Jul 20 21:11:50 link-13 kernel: device-mapper: unable to get server (1) to mark region (1713) Jul 20 21:11:50 link-13 kernel: device-mapper: Reason :: 1 Jul 20 21:11:50 link-13 kernel: device-mapper: sync_count (204940) > region_count (204800) - (mjIFcdjY)! Jul 20 21:12:02 link-13 last message repeated 954 times Jul 20 21:12:02 link-13 kernel: device-mapper: sync_count (204941) > region_count (204800) - (mjIFcdjY)! Jul 20 21:12:02 link-13 last message repeated 34 times Jul 20 21:12:02 link-13 kernel: device-mapper: sync_count (204942) > region_count (204800) - (mjIFcdjY)! Jul 20 21:12:05 link-13 last message repeated 287 times Jul 20 21:12:05 link-13 kernel: device-mapper: an error occured on the server while processing our request Jul 20 21:12:05 link-13 kernel: device-mapper: sync_count (204942) > region_count (204800) - (mjIFcdjY)! Jul 20 21:12:37 link-13 last message repeated 3068 times Jul 20 21:12:46 link-13 last message repeated 706 times The IO is still running a day later, and the logs are still showing the message. IO run was: b_iogen -o -m random -s read,write,readv,writev -t 1000b -T 100000b -d /dev/mirror_vg/mirror_vol | b_doio Translated that states: Do overlapping read(), write(), readv(), and writev(). Stride randomly on device. Generate each request with a minimum transfer of 1000 blocks, and a maximum transfer of 100000 blocks. Version-Release number of selected component (if applicable): lvm2-cluster-2.02.06-5.0.RHEL4 lvm2-2.02.06-5.0.RHEL4 device-mapper-1.02.07-4.0.RHEL4 cmirror-kernel-2.6.9-10.2
This bug should only be in cmirror-kernel, so all necessary fixes should be in.
These messages have not been seen in some time, marking verified.
Assuming this VERIFIED fix got released. Closing. Reopen if it's not yet resolved.