Increasing memory usage
awcheney opened this issue · comments
Hi folks,
We're running kubernetes-event-exporter in all our clusters, using it to export events to ElasticSearch. In just one of six clusters, we're seeing what looks like a memory leak - memory usage gradually increases over time until the container is OOMKilled and the cycle starts again. This cluster is seeing a lot more events than the others (around 30* as many, at roughly 18k-20k events per hour) - it runs our Selenium grid.
We're running the latest image (v0.10) as of yesterday - bumping the version made no difference.
It's not super-critical - we probably lose a few events when the container is killed - but it would be good to understand if there's anything we can do about this.
Ngā mihi nui!
I also had this problem. I counted the logs before OOM, and the number of events was about 30/s, which lasted 19 minutes.