FlexGroup volume intermittently goes offline
Applies to
- ONTAP 9
- FlexGroup Volumes
Issue
- A FlexGroup volume becomes intermittently inaccessible when the constituent goes full.
- One FlexGroup constituent volume is going full even though multiple other constituents are not utilized:
Cluster1::*> vol show FlexGroup1*
Vserver Volume Aggregate State Type Size Available Used%
--------- ------------ ------------ ---------- ---- ---------- ---------- -----
Svm1 FlexGroup1 - online RW 6.01TB 981.9GB 16%
Svm1 FlexGroup1__0001 aggr1 online RW 241.3GB 241.1GB 2%
Svm1 FlexGroup1__0002 aggr2 online RW 241.6GB 241.5GB 0%
Svm1 FlexGroup1__0003 aggr1 online RW 5.30TB 257.9GB 95%
Svm1 FlexGroup1__0004 aggr2 online RW 241.9GB 241.3GB 0%
- The following errors are reported in the event logs:
[Node-01: wafl_spcd_main: monitor.volume.nearlyFull:error]: Volume FlexGroup1__0003@vserver:63xxxx6-76dc-11e9-9094-00xxxxxxb34 is nearly full (using or reserving 95%% of space and 0%% of inodes).
- The FlexGroup comes back online after the space is balanced using elastic sizing:
[Node-01: wafl_exempt09: fg.member.elastic.sizing:notice]: Constituent 1550 in FlexGroup FlexGroup1 was resized from 1378003688 to 1378032431.
[Node-01: wafl_exempt11: fg.member.elastic.sizing:notice]: Constituent 1550 in FlexGroup FlexGroup1 was resized from 1378032431 to 1378042397.
[Node-01: wafl_spcd_main: monitor.volumes.one.ok:debug]: Volume FlexGroup1__0003@vserver:63xxxx6-76dc-11e9-9094-00xxxxxxb34 is OK.