Bruno Bressi

Just wanted to say thanks for the commit @chinghua0731, you saved me more than some time trying to solve this myself :heart:

Did you manage to get around to implementing this? Or does someone have a workaround/ an idea on how to enforce a single instance on a Linux system?

Ah ok, and then it would wrap the fman binary. My aim is to use only fman as the default file manager to handle "file-manager-related" MIME Types. I'll try that,...

We also had this issue in Rancher 2.6.9 - thanks for the workarounds! To other people using the code for helm3, please note the ┆ symbol in the bash script:...

That didn't solve the issue unfortunately. Could someone from rancher take a look in the logs to at least try and point us in some direction? This issue is still...

We have been doing some debugging on the issue the last couple of weeks. We've build own own cluster-agent image with a debug log on the `writeData` function of the...

What I forgot to mention, is that we also built a dev image of the cluster-agent with golang's race detector. The cluster-agent has multiple race conditions happening in multiple parts...

@JonCrowther Crash still happens, once per day with the debounce set to 60s. We'll keep on monitoring for a couple of days to get a better feeling of the mean...

@JonCrowther We've been running your fix for the last week, and it works like a charm. We just installed rancher 2.8.3 today and were surprised to the new `steve` version...