crate
crate copied to clipboard
crate
WARN: `failed to turn off translog retention` on startup
CrateDB version
5.0.0
CrateDB setup information
No response
Steps to Reproduce
After upgrading a cluster from 4.7 to 5.0.0 we see lots of the following error messages on CrateDB startup
2022-07-26 17:18:09 | [2022-07-26T15:18:09,879][WARN ][o.e.i.s.IndexShard ] [data-hot-0] [schema.table][0]failed to turn off translog retention
-- | --
| | 2022-07-26 17:18:08 | at java.lang.Thread.run(Thread.java:833) [?:?]
| | 2022-07-26 17:18:08 | at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635) [?:?]
| | 2022-07-26 17:18:08 | at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136) [?:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.common.util.concurrent.AbstractRunnable.run(AbstractRunnable.java:37) [crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard$3.doRun(IndexShard.java:1861) [crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard.trimTranslog(IndexShard.java:1050) ~[crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard.getEngine(IndexShard.java:2453) ~[crate-server.jar:?]
| | 2022-07-26 17:18:08 | org.apache.lucene.store.AlreadyClosedException: engine is closed
| | 2022-07-26 17:18:08 | [2022-07-26T15:18:08,977][WARN ][o.e.i.s.IndexShard ] [data-hot-0] [.partitioned.schema.table2][0]failed to turn off translog retention
| | 2022-07-26 17:18:08 | at java.lang.Thread.run(Thread.java:833) [?:?]
| | 2022-07-26 17:18:08 | at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635) [?:?]
| | 2022-07-26 17:18:08 | at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136) [?:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.common.util.concurrent.AbstractRunnable.run(AbstractRunnable.java:37) [crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard$3.doRun(IndexShard.java:1861) [crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard.trimTranslog(IndexShard.java:1050) ~[crate-server.jar:?]
| | 2022-07-26 17:18:08 | at org.elasticsearch.index.shard.IndexShard.getEngine(IndexShard.java:2453) ~[crate-server.jar:?]
| | 2022-07-26 17:18:08 | org.apache.lucene.store.AlreadyClosedException: engine is closed
Expected Result
no errors / warnings on startup after migrating from 4.x to 5.0
Actual Result
9 | [2022-07-26T15:18:09,879][WARN ][o.e.i.s.IndexShard ] [data-hot-0] [schema.table][0]failed to turn off translog retention
-- | --
Those are WARN messages, not errors.
Did the ugprade run smoothly nevertherless?
If the update runs smoothly, then the issue is only the WARN logs which I wouldn't consider changing anything regarding those logs, as it could be helpful to troubleshoot real issues during upgrade.
If it causes much trouble to users, maybe we could define another logger for those, or mention them in an upgrade guide/how-to
@proddata The warning log message should not be written in some cases, fixed by #12982, but if so it can be ignored. Thank you for reporting.
