New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
CPU utilization #25
Comments
definitely not normal, can you get a core dump? or stack trace? strace? you can tmate me in if you'd like me to help on the machine |
Is there an easy way to determine stale or hung tmate-slave services, such that we can know which can be safely terminated and which might be in use by a client? |
I really don't understand your to see what are the active sessions, you could |
I can fully shutdown the daemon and forcibly terminate any remaining tomorrow, after communicating the shutdown to everyone who might be using it. If after we restart tmate-slave we see the same situation, what do you suggest we try next? |
give me the stack traces (or let me in that box through tmate so I can debug) (gdb -p , etc.) |
Private box inside the firewall. I can certainly ask tomorrow, but I wouldn't expect to see permission granted for you to connect.
Apologies if I'm not giving you all you need: I'm a sysadmin, not a developer. gdb is new and scary to me. |
this process is fine (it's the listener) how about how of these process that is stuck? |
Here's the backtrace output from a rather random selection of PIDs:
PID 9107:
PID 9125:
PID 9283:
PID 32258:
PID 32607:
|
So there's two problems:
|
Try that |
FYI: upgrading the tmate-slave daemon, and restarting it will not affect any current running sessions. The new connections will be served by the restarted daemon, while the other connections stay on the old code. |
Thanks! Just to be safe, we shutdown all the existing tmate-slave services. We built and installed a new RPM from the latest code. Things look okay right now. If it goes south again, we'll open a new ticket. 👍 |
👍 |
We're running tmate-slave on a RHEL7 server. Load on this system is extremely high, and CPU is pegged in nice.
Is this normal behavior? Do we need to allocate more CPU to tmate-slave hosts?
The text was updated successfully, but these errors were encountered: