0

I have run the Elasticsearch service for quite long time, but suddenly encountered the following

Caused by: org.elasticsearch.index.translog.TranslogCorruptedException: translog from source [d:\elasticsearch-7.1.0\data\nodes\0]indices\A2CcAAE-R3KkQh6jSoaEUA\2\translog\translog-1.tlog] is corrupted, expected shard UUID [.......] but got: [...........] this translog file belongs to a different translog.

I executed the GET /_ca/shards?v and most of the indexes are UNASSIGNED state.

Please help!

I went through the log files and saw the error message "Failed to update shard information for ClusterInfoUpdateJob within 15s timeout", could this error message cause most of the shards turn to UNASSIGNED?

xxestter
  • 441
  • 7
  • 19

1 Answers1

0

You can try to recover using elasticsearch-translog tool as explained in the documentation

  1. Elasticsearch should be stopped while running this tool
  2. If you don't have replica from which data can be recovered, you may lose some data by using the tool.
  3. Reason is mentioned that drive error or user error.
Gibbs
  • 21,904
  • 13
  • 74
  • 138
  • Hi, I am using Elasticsearch version 7.1, couldn't use the elasticsearch-translog – xxestter Aug 28 '20 at 02:25
  • @xxestter for Elasticsearch 7.1 you can use `elasticsearch-shard` which is what `elasticsearch-translog` was renamed to. – Nick Jul 30 '22 at 14:35
  • Following up on my previous comment I believe this naming change happened in version 6.5 because it is the first time it shows up under the "command line tools" section of their documentation. https://www.elastic.co/guide/en/elasticsearch//reference/6.5/shard-tool.html – Nick Jul 30 '22 at 14:46