After redis dropping tile expiry lists for months, the tiles are out of data. We should regenerate them from tileratorui. This needs doing in both datacenters.
Description
Description
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Gehel | T175123 tileshell does not honor redis configuration in /etc/tilerator/config.yaml | |||
| Resolved | Pnorman | T176252 Regenerate tiles |
Event Timeline
Comment Actions
Mentioned in SAL (#wikimedia-operations) [2017-10-03T20:29:00Z] <pnorman> Start regenerating map tiles on codfw for z0-z12 - T176252
Comment Actions
Mentioned in SAL (#wikimedia-operations) [2017-10-03T23:46:22Z] <pnorman> Start regenerating map tiles on eqiad for z0-z12 - T176252
Comment Actions
Mentioned in SAL (#wikimedia-operations) [2017-10-04T19:38:07Z] <pnorman> Start regenerating map tiles on eqiad for z13-z14 - T176252
Comment Actions
Mentioned in SAL (#wikimedia-operations) [2017-10-04T20:18:05Z] <pnorman> Start regenerating map tiles on codfw for z13-z14 - T176252
Comment Actions
This is done. It took approximately 2h for z0-z11, 3h for z12, 15h for z13, 58h for z14. So 3-4 days to regenerate using 50% of capacity, which is pretty good.