This means you have configured Enterprise Search to listen on port 9200 and there is something on your server already using that port (I suspect Elasticsearch). I’d recommend checking your configuration file and making sure you didn’t mix up listen host/port and Elasticsearch host/port in enterprise-search.yml.
@oleksiy-elastic I had to configure it to 9200 and moved elastic to 9201 as we only have 9200 open to our AWS VMs and it takes a long time to get that changed.
That issue has resolved itself after a few restarts (or maybe when I changed the user for the filebeat log??) and got replaced by
NoMethodError: undefined method `join' for nil:NilClass
Which I have found some posts for already and it turned out that when AppSearch started up, it killed ElasticSearch - and pretty much the whole VM.
I restarted the VM and it's actually up and running now, but I suspect failures will start again when the log rotates as the owner will be root again instead of the enterprise-search user.
Any suggestions on how to resolve that?
Logs appear to be rotating fine so I'm now a very happy owner of a working setup, thank you!
Can you setup Enterprise Search as a service if you did not install from a package? I downloaded the enterprise-search-7.16.2.tar.gz and did a manual install. How can I keep Enterprse Search running all the time?
env ENT_SEARCH_DEFAULT_PASSWORD=X nohup bin/enterprise-search& doesn't work.
I had an issue installing from RPM for v.7.16... would run fine from bin/enterprise-search, but not from systemctl start. After reading the log in journal, I set permissions to the log directories as:
chown -R enterprise-search:enterprise-search /usr/share/enterprise-search/log
chown -R enterprise-search:enterprise-search /var/log/enterprise-search/
tail -f /var/log/enterprise-search/app-server.log.
Once done, it ran fine.