[sheepdog-users] not all vdi are shown after cluster restart

[Valerio Pachera]

2013/8/9 Liu Yuan <namei.unix at gmail.com>:
> After some tests with current master, I think this problem is solved already.
> Please update to the latest master, it is kind of stable and is the release
> candicate for v0.7.0

Great Yuan!

0.7.0_rc0_12_g2b526ef

# parallel-ssh -i -h etc/pssh.conf 'collie vdi list | grep -c backup_data'
[1] 15:11:13 [SUCCESS] sheepdog004
21
[2] 15:11:13 [SUCCESS] sheepdog003
21
[3] 15:11:13 [SUCCESS] sheepdog002
21
[4] 15:11:13 [SUCCESS] sheepdog001
21

I've been counting all the vdi (and snapshot) named 'backup_data' and
they match!
Right now it's recovering something.

In sheep.log there are lot's of

Aug 09 15:09:01 [main] get_vdi_copy_number(103) No VDI copy entry for
c8d144 found
Aug 09 15:09:01 [main] get_vdi_copy_number(103) No VDI copy entry for
c8d144 found
Aug 09 15:09:01 [main] get_vdi_copy_number(103) No VDI copy entry for
c8d144 found
Aug 09 15:09:01 [main] get_vdi_copy_number(103) No VDI copy entry for
c8d144 found


then lot's of
Aug 09 15:09:09 [main] recover_object_main(625) done:6 count:414446,
oid:2e51a00009068
Aug 09 15:09:09 [main] recover_object_main(625) done:7 count:414446,
oid:c8d12f00035981
Aug 09 15:09:09 [main] recover_object_main(625) done:8 count:414446,
oid:c8d14400002647
Aug 09 15:09:09 [main] recover_object_main(625) done:9 count:414446,
oid:c8d13900034f

before them
Aug 09 14:33:46 [main] main(797) shutdown
Aug 09 15:09:00 [main] md_add_disk(142) /mnt/sheep/dsk01/obj, nr 1
Aug 09 15:09:00 [main] md_add_disk(142) /mnt/sheep/dsk02, nr 2
Aug 09 15:09:00 [main] md_add_disk(142) /mnt/sheep/dsk03, nr 3
Aug 09 15:09:00 [main] send_join_request(774) IPv4 ip:192.168.6.41 port:7000
Aug 09 15:09:00 [main] for_each_object_in_stale(384) /mnt/sheep/dsk01/obj/.stale
Aug 09 15:09:00 [main] for_each_object_in_stale(384) /mnt/sheep/dsk02/.stale
Aug 09 15:09:00 [main] for_each_object_in_stale(384) /mnt/sheep/dsk03/.stale
Aug 09 15:09:00 [main] init_vdi_state(188) failed to read inode header
800e4aa600000000 0
Aug 09 15:09:00 [main] init_vdi_state(188) failed to read inode header
80c8d12e00000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80c8d13700000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80f131b700000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80c8d13e00000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80c8d13600000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80c8d12800000000 0
Aug 09 15:09:01 [main] init_vdi_state(188) failed to read inode header
80c8d14400000000 0
Aug 09 15:09:01 [main] main(869) sheepdog daemon (version
0.7.0_rc0_12_g2b526ef) started

Once the recovery is done I'll try to check some disk and run some vm.

I hope everything to be fine.

Thank you.