T.J. Yang

I am following this PR: https://github.com/saltstack/salt/pull/66335 , not really know the details. but this issue looks like a publish port issue that 66335 trying to fix.

- EventPublisher process use 1.6% of 16G, this is the time port 4505 failed. ``` [me@salt01 srv]$ egrep '286129|MEM' salt-2024-04-24-06\:56.txt PID STARTED TIME %MEM RSS SZ VSZ SIZE 286129 01:00:02...

- This is my effort using makefile to automate the tracking of EventPublish memory usage when it stop working. ie , putting "make memory" into hourly cronjob to record the...

my 2nd salt-master failed to test.ping itself, restart salt-master doesn't work anymore. I had to reboot the RL8 OS. this definitely smell like kernel network resource depletion. - OS restart...

After downgrade from 3007 to 3006.8 version, my salt-master servers, the 4505 port is now stable and able to send out salt commands.

@compi-tom, Thanks for the confirmation. Hopeful when 3007 turned into LTS from STS. This issue can be acked by core team and resolved.

I have another centos 7 on vmware workstation which behave better. I was able to break out the haning test phase by control+C. The other one is on ESXi 5.5...

@doesitblend , thanks for the oneliner script to cleanup the .pyc. I am still getting this error message from minion startup. Any suggestion ? salt-minion[1567838]: [ERROR ] Publish server binding...

@z900collector , the one-liner worked for me to remove *.pyc files and I don't have your message above. I only have following error message from "systemctl status salt-minion" per another...

rocky9t01a test VM upgraded from 3006.6 to 3007.0. I thought next version 3006.8 will be latest one with this bug fix but latest one is 3007 without this bug fix....