Trouble running ElasticSearch with previous data after upgrade from 6.7 to 7.4 (through 6.8)

I had a local Ubuntu 18 on a Hyper-V machine with ES 6 installed.
The path.data & path.logs are configured on a mounted external drive.

The IT guys tried to upgrade it to 6.8, then to 7.4 but something went wrong (Kibana kept restarting).

This machine does not exist anymore.

I now have a new machine with ES & Kibana 7.4 that works with the default '/var/lib/elasticsearch'.

I mounted the external drive and changed the paths to the respective paths in the external storage.

When I start the services, I can get to ES from remote machines.

Kibana does not work.

When I try to start only ES, 'http://10.163.129.77:9200/_cat/health?v' returns:

status: red
unassign: 4899
pending_tasks: GROWING INDEFINITELY

Any ideas?
Other info\logs you need to help me?

Thanks in advance.

Hi @IdanLevi,

there are a couple of things to try out to see if the data can be rescued:

  1. Examine why the shards stay unassigned by using the allocation explain api.
  2. Examine the pending cluster tasks using the pending tasks api.
  3. Look into if you have any assigned shards at all or if all shards are red?

Also, just to be sure, your ES6 installation has never been running any prior versions of Elasticsearch? Indexes created on ES5 will need to be reindexed on ES6 before upgrading to ES7.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.