Bug #13916
closederasure code profile name
0%
Description
I've been experimenting with ceph erasure pools. In my naivety I created an erasure pool with a dash in the name. Example:
ceph osd erasure-code-profile set myprofile-wide k=3 m=2 ruleset-failure-domain=rack
The pool fails to create because the pg's never get created. They get stuck in incomplete-creating. I did a query on a few of the pgs and it's not apparent what is wrong. They don't appear to be stuck or waiting on anything. When I delete the pool and recreate the same thing without the dash in the name it goes through just fine. I think the ceph tool should deny invalid names. Here's what the command returns when you query the erasure profile after creating it.
ceph osd erasure-code-profile get ceph-wide directory=/usr/lib/x86_64-linux-gnu/ceph/erasure-code k=10 m=3 plugin=jerasure ceph-wide= ruleset-failure-domain=osd technique=reed_sol_van
Updated by Loïc Dachary over 8 years ago
I tried the following and the problem does not show. It would be great if you could provide me with a sequence of commands to reproduce the problem. Thanks :-)
loic@fold:~/software/ceph/ceph/src$ ceph osd erasure-code-profile set myprofile-wide k=3 m=2 ruleset-failure-domain=rack loic@fold:~/software/ceph/ceph/src$ ceph osd erasure-code-profile get myprofile-wide jerasure-per-chunk-alignment=false k=3 m=2 plugin=jerasure ruleset-failure-domain=rack ruleset-root=default technique=reed_sol_van w=8 loic@fold:~/software/ceph/ceph/src$ ceph osd pool create mypool 1 1 erasure myprofile-wide loic@fold:~/software/ceph/ceph/src$ ceph osd dump epoch 14 fsid e509fb9d-2119-400b-b3ee-e595827698a5 created 2015-12-14 13:17:26.718295 modified 2015-12-14 13:21:28.586998 flags sortbitwise pool 0 'rbd' replicated size 3 min_size 1 crush_ruleset 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 1 flags hashpspool stripe_width 0 pool 1 'mypool' erasure size 5 min_size 3 crush_ruleset 1 object_hash rjenkins pg_num 1 pgp_num 1 last_change 14 flags hashpspool stripe_width 4128 max_osd 3 osd.0 up in weight 1 up_from 4 up_thru 10 down_at 0 last_clean_interval [0,0) 127.0.0.1:6800/22215 127.0.0.1:6801/22215 127.0.0.1:6802/22215 127.0.0.1:6803/22215 exists,up 7a5471f8-514e-4b6c-8647-63a8654e61e6 osd.1 up in weight 1 up_from 6 up_thru 10 down_at 0 last_clean_interval [0,0) 127.0.0.1:6804/22415 127.0.0.1:6805/22415 127.0.0.1:6806/22415 127.0.0.1:6807/22415 exists,up 2755a563-fcd9-47a4-89bc-4033e74f855d osd.2 up in weight 1 up_from 9 up_thru 10 down_at 0 last_clean_interval [0,0) 127.0.0.1:6808/22668 127.0.0.1:6809/22668 127.0.0.1:6810/22668 127.0.0.1:6811/22668 exists,up acce5a40-b018-42d2-96f5-ccc563df4b6b loic@fold:~/software/ceph/ceph/src$ ceph osd pool get mypool erasure_code_profile erasure_code_profile: myprofile-wide
Updated by Loïc Dachary over 8 years ago
- Status changed from New to Need More Info
Updated by Samuel Just about 8 years ago
- Status changed from Need More Info to Can't reproduce
I'm closing this on the assumption that the original problem was an issue with the reporter's crush map. Reopen if I'm missing something!