I would strongly encourage you at this cluster size to have the metadata on SSD (or metadata and data). Especially if you don't have enough RAM to keep around the data in the metadata database. I…
Based on your informations above I would say that garage is stalling because the disk is simply too slow to handle your cluster size. Could you get some telemetry from the OS to confirm that?…
Layout avant mise à jour :
[root@df-ykl:~]# docker exec -it 0bc65f9b208e /garage layout show
2024-11-28T21:05:45.674060Z INFO garage_net::netapp: Connected to [2a02:a03f:6510:5102:6e4b:9…Vu les besoins en terme de latence qui doivent être relativement faible, je pense qu'il n'y a aucun prérequis à avoir woodpecker qui utilise forcément un postgres local.
3661a597fa
Merge pull request 'feat: add use_local_tz configuration' (#908) from ragazenta/garage:feat/local-timezone into main
e87942dad3
Merge pull request 'Deploy new postgres replica on pasteque@corrin' (#46) from migration-neptune into main
@ragazenta do you mind adding something to the reference configuration?
Well from my understanding workers might start at different times on different nodes depending on the local configured TZ
Small question if possible, there is no reason why really but does that value change if you restart the node?
This could lead to interesting behaviors if not all nodes share the same local TZ settings... Nothing dangerous, but maybe some unexpected?
Closing for now as the questions has been answered