How to Set up Gerrit (Master/Slave) for ELastic Stack Monitoring

ritesh shetty

unread,

Jul 15, 2019, 12:58:57 PM7/15/19

to Repo and Gerrit Discussion

Hello Folks,

We got a new requirement to monitor complete Gerrit infra (more than 200 nodes). I felt Elastic Stack is the good use case to set up monitoring of Gerrit servers as well as Elastic APM to find any application related issues.

Is there any documentation on what needs to be monitored as well as best practices for setting up an effective monitoring solution.

For POC purpose, do we have any Gerrit repo to clone and set ElasticStack monitoring solution?

Let me know how your thoughts as well.

Thanks,

Ritesh

Doug Luedtke

unread,

Jul 15, 2019, 1:44:10 PM7/15/19

to Repo and Gerrit Discussion

You win! I have about 50 nodes at the moment. :)

We use the Elastic Stack for collecting the logs from all the nodes. I'm always requesting changes because I am not the Administrator for the Elastic Stack products.

The Elastic company seems to be very good with their support. They offered one local class and it was very informative. Elastic has been difficult for me because I'm not an Administrator for the Elastic products and I did not fully know the options available to me. So I make requests and in most cases they have been done. I'm sure there is a better way for me to use the Elastic Stack, but it's an evolving process at the moment.

I'd be very interested to know how everyone else uses it too.

David Pursehouse

unread,

Jul 15, 2019, 8:19:31 PM7/15/19

to ritesh shetty, Repo and Gerrit Discussion

I guess this is about monitoring logs etc (i.e via logstash or similar) rather than gerrit's internal metrics. But just in case it is about the latter, please note that the metrics reporting to elasticsearch no longer works. See issue 9516 [1] for details.

[1] https://crbug.com/gerrit/9516

Sven Selberg

unread,

Aug 9, 2019, 4:05:30 AM8/9/19

to Repo and Gerrit Discussion

We use the ELK stack for sshd, access, error logs (FileBeat -> Logstash -> Elastic) and gerrit-events (Gerrit rabbitmq plugin -> RabbitMQ -> Logstash -> Elastic).

On Monday, July 15, 2019 at 6:58:57 PM UTC+2, ritesh shetty wrote:

Hello Folks,
We got a new requirement to monitor complete Gerrit infra (more than 200 nodes). I felt Elastic Stack is the good use case to set up monitoring of Gerrit servers as well as Elastic APM to find any application related issues.
Is there any documentation on what needs to be monitored as well as best practices for setting up an effective monitoring solution.

I don't know of any such repository but we have some Logstash filters that might help.

Reply all

Reply to author

Forward