Error while loading events, show in logs of one of clickhouse shard
Summary
Got error when loading events, but ingest, rotor and console work :
shard1 error:
And rest of services look OK :
shard0
bulker:
ingest:
│ time="2024-08-19 16:02:36" level=info msg="[metrics_server] Starting metrics server on 0.0.0.0:9091" │││ │ ││││││ │ │ │ │ │
│ time="2024-08-19 16:02:36" level=info msg="Starting http server on 0.0.0.0:3049" │││ │ ││││││ │ │ │ │ │───────────────────────┐───────────────┐────────┐
│ time="2024-08-19 16:04:01" level=error msg="[ch_events_log] Error preparing batch: code: 60, message: Table default.events_log does not exist"
System configuration and versions
k8s install of Jitsu v2.8.0
(https://github.com/stafftastic/jitsu-chart)
Seems like the table that for storing event logs is configured incorrectly. It should be created in newjitsu_metrics DB, see compose file for a reference
It might be because of https://github.com/stafftastic/jitsu-chart/blob/fb5d459c9c831d25c99a49981c92de43b174c1d8/templates/_helpers.tpl#L145 -> when we use the jitsu-chart, it uses by default the "default" db.
It changed recently ?
@hendrikheil After investigation, I found that this error of unknown table does not always occur (sometimes the data load well in my console). Then I went to my k8s clickhouse shards and find out that in one of the shard the db does not exist :
I got 1 replica and 2 shards :
This is the shard for which the table does not exist :
That shard does have it :
Any idea what I could do to fix this ?
This issue is stale because it has been open for 30 days with no activity.
This issue was closed because it has been inactive for 14 days since being marked as stale.