-
Notifications
You must be signed in to change notification settings - Fork 158
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
updates for ext.config.kdump.crash
test
#1708
Conversation
dustymabe
commented
May 2, 2022
•
edited
Loading
edited
@@ -37,3 +37,12 @@ packages: | |||
# Boost starving threads | |||
# https://github.com/coreos/fedora-coreos-tracker/issues/753 | |||
- stalld | |||
|
|||
postprocess: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
added here so we can share with RHCOS.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I ran this locally and the symlink to commonlib.sh
is broken. It's a relative symlink and the test is one level deeper.
Related PRs/Issues that can be closed if this merges |
c9c2a8f
to
c8073fb
Compare
Up until this point people have had to reboot once after setting up kdump so that if a crash happens the crash kernel doesn't get started with the ignition.firstboot kernel argument. If we add `ignition.firstboot` to the list of arguments in `KDUMP_COMMANDLINE_REMOVE=` in `/etc/sysconfig/kdump` then we can set it up on first boot without issues. Here we also switch the kdump.crash test to set up kdump via Ignition to prove out the change works as desired.
The kdump.crash test on the rawhide stream for aarch64 has started taking just longer than 10 minutes. Let's extend the timeout so those tests don't timeout and fail the entire run. Also added comments for the test's kola.json arguments.
In our documentation [1] we recommend setting it to 300M as a starting baseline. Let's update from 256M to 300M here. [1] https://docs.fedoraproject.org/en-US/fedora-coreos/debugging-kernel-crashes/
0be5546
to
f49f403
Compare
The irqpoll` karg that gets added when the crash kernel is started causes issues on AWS aarch64 instances. Let's workaround by removing that from `KDUMP_COMMANDLINE_APPEND` in `/etc/sysconfig/kdump` for now. See coreos/fedora-coreos-tracker#1187
This move the `arch-include` statments to `conditional-include`, which allows us to use `!=`. In this case what we were really trying to target was s390x IIUC.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
CI flaked on coreos/fedora-coreos-tracker#1190 |