Describe the issue:
We are experiencing a lot of circuit breaker exceptions while indexing our daily logs.
My main questions are what could causes this issue and what are some essentials to monitor what is exactly in memory.
I have a hard time to understand what is in memory to know what can causes this issues.
The log is telling you that the size of the bulk update is larger than the limit in which the circuit breaker is configured to handle. The best approach to prevent this going forward is to optimise for smaller bulk updates or add the required resources needed. I suggest also reading this document - Circuit breaker settings - OpenSearch Documentation
Hey @Leeroy,
many thanks for your answer. I just read through that documentation and it helped.
I just have one last question. What are some essentials to monitor in opensearch?
I use the zabbix template for ElasticSearch on our cluster but there are probably a lot more information that are useful. If you don’t know it’s no big deal, i just wanted to ask.
Again many thanks for your response.
No problem, and good question. Opensearch has lots of different features, so finding exactly what best to monitor and alert on depends highly on your cluster setup and what it is you’re using Opensearch for.
I reviewed the template I believe you’re referring to. A couple of points stand out, but first I be curious to know which triggers you’re using? Do prototype triggers enable by default?
I would first review the triggers in which exist, note down what it is you’re checking on. You can report back and we can discuss further.