Randomize the id with GITHUB_RUN_NUMBER for scheduled test #7392
Conversation
Signed-off-by: Young Bu Park <[email protected]>
youngbupark
temporarily deployed
to
functional-tests
GitHub Actions
Inactive
Radius functional test overview
Click here to see the list of tools in the current test run
Test Status⌛ Building Radius and pushing container images for functional tests... |
Signed-off-by: Young Bu Park <[email protected]>
youngbupark
had a problem deploying
to
functional-tests
GitHub Actions
Failure
|
For the detailed report, please go to
|
Signed-off-by: Young Bu Park <[email protected]>
Signed-off-by: Young Bu Park <[email protected]>
youngbupark
had a problem deploying
to
functional-tests
GitHub Actions
Failure
|
For the detailed report, please go to
|
Signed-off-by: Young Bu Park <[email protected]>
youngbupark
had a problem deploying
to
functional-tests
GitHub Actions
Failure
youngbupark
had a problem deploying
to
functional-tests
GitHub Actions
Failure
Signed-off-by: Young Bu Park <[email protected]>
youngbupark
temporarily deployed
to
functional-tests
GitHub Actions
Inactive
Radius functional test overview
Click here to see the list of tools in the current test run
Test Status⌛ Building Radius and pushing container images for functional tests... |
| UNIQUE_ID=$(echo $BASE_STR | sha1sum | head -c 10) | ||
| if [ "$GITHUB_EVENT_NAME" == "schedule" ]; then | ||
| # Add run number to randomize unique id for scheduled runs. | ||
| BASE_STR="${GITHUB_RUN_NUMBER}|${BASE_STR}" |
There was a problem hiding this comment.
is there any reason we only do this for scheduled runs? is there a downside to adding the run number on all runs?
There was a problem hiding this comment.
If we use GITHUB_RUN_NUMBER for PR event, workflow will create resource groups and resources whenever we trigger functional tests. This results in unnecessary resource creation in our subscription. To avoid this issue and save the cost, we do not use GITHUB_RUN_NUMBER for every event.
There was a problem hiding this comment.
Even though we have purge workflow, it cleans up only 3 days+ old resources. So we still need to pay 3 days resoruce usage. so it is better not to use GITHUB_RUN_NUMBER for every event.
|
For the detailed report, please go to
|
Description
We generate unique id by concatenating GitHub Actions' system variables, hashing the string, and taking 10 characters. This way will reduce the chance to create multiple resource for the same PR. In scheduled test, the most of variables are same, so there is a little chance to face the conflicts. particularly, when resource group is being in progress triggered by the previous action runs like below failures. This change 1) adds GITHUB_RUN_NUMBER, which is updated for each action run, to the raw id before hashing and 2) adds the prefix to each hashed string to avoid the conflicts between different workflows.
#7383
#7382
#7380
Type of change
Fixes: #7388