-
Notifications
You must be signed in to change notification settings - Fork 75
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
Additional workloads: cyclictest, oslat and testpmd add #264
Conversation
bold=$(tput bold) | ||
uline=$(tput smul) | ||
normal=$(tput sgr0) | ||
python3 -m pip install -r requirements.txt | grep -v 'already satisfied' |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should probably deploy a venv first and then install/run from there to avoid any potential issues
bold=$(tput bold) | ||
uline=$(tput smul) | ||
normal=$(tput sgr0) | ||
python3 -m pip install -r requirements.txt | grep -v 'already satisfied' |
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.
Use a venv?
@@ -14,6 +14,21 @@ fi | |||
export TOLERATIONS="[{key: role, value: workload, effect: NoSchedule}]" | |||
export UUID=${UUID:-$(uuidgen)} | |||
|
|||
export baremetalCheck=$(oc get infrastructure cluster -o json | jq .spec.platformSpec.type) |
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.
@rsevilla87 can you vet these changes given they impact all kube-burner e2e tests
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.
@rsevilla87 re-ping
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, seems like it checks if the test runs in a BM environment to install python 3.8 later. However I think it should be a software requirement and we shouldn't do it on runtime (we could say that oc and kubectl are also requirements.)
…from clashing / overlapping benchmarks
@mohit-sheth @sjug @rsevilla87 Can we get another set or two of eyes on this one please? |
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!
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, tested on a 120 node BM cluster.
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!
* Updated workloads to run on baremetal * Added oslat, cyclictest and testpmd workloads * combination of 39 commits, squashed * no more echoing empty lines * changes for the PR / removing unnecessary files * checking for the existance of id_rsa for the kni user * reverting changes * Missing bracket * changes to check only the benchmark with the current UUID to prevent from clashing / overlapping benchmarks * Turned the time to wait for the perfprofile into a variable, fixed a typo * removing the UUID part of the benchmark name in the templates * fixing a few minor issues, reverting UUID changes * fixed cyclictest index * last bits for the merge * removed a double entry * final cleanup * indentation error Co-authored-by: jdowni000 <[email protected]> Co-authored-by: Marko Karg <[email protected]>
Description
Three more workloads added:
Fixes
n/a
This PR depends on the work @jdowni000 has done for the baremetal clusters #248