Recent changes to content topology have left the nodes in eqiad with "extra" data, data which is no longer associated with them according to the current partitioning. The prescribed solution is a nodetool cleanup. Normally I'd recommend waiting until the final topology is in place to avoid double-handling, but space on these nodes is becoming quite tight.
I propose we initiate concurrent cleanups, one per rack (to limit any additional compaction-induced latency impact).
Host | Rack | Sequence | Status |
---|---|---|---|
restbase1002.eqiad.wmnet | a | 1 | Cancelled (see: https://phabricator.wikimedia.org/T121535#1887497) |
restbase1001.eqiad.wmnet | a | 2 | Started ~2015-12-17 18:06:00 (https://tools.wmflabs.org/sal/log/AVGxFO_i1oXzWjit6-Du) |
restbase1007-a.eqiad.wmnet | a | 3 | Freshly boostrapped |
restbase1003.eqiad.wmnet | b | 1 | Complete |
restbase1004.eqiad.wmnet | b | 2 | Decommissioning (hands-off) |
restbase1008-a.eqiad.wmnet | b | - | Freshly bootstrapped |
restbase1005.eqiad.wmnet | d | 1 | Complete; Complete |
restbase1006.eqiad.wmnet | d | 2 | Complete |
restbase1009-a.eqiad.wmnet | d | 3 | Complete |
Edit:
Resequence 1002 above 1001, as disk space is higher there.