xK6 delete S3 job removes all objects in bolt database but one. #153
Labels
No labels
P0
P1
P2
P3
good first issue
Infrastructure
blocked
bug
config
discussion
documentation
duplicate
enhancement
go
help wanted
internal
invalid
kludge
observability
perfomance
question
refactoring
wontfix
No milestone
No project
No assignees
4 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: TrueCloudLab/xk6-frostfs#153
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Expected behaviour:
All objects in bolt database are deleted
Existing behaviour:
One object in bolt database is not deleted.
Steps to repoduce:
/home/service/opt/k6/s3_128k.bolt
after removal contains one object, the same object is stores in bucket:(screenshot of boltDB browser and S3 browser with the object is attached)
Use latest K6 version:
Howto reproduce issue:
s3_registry.db
for entries.Looks like we are interested in OneShortSelector creation.
Here is where we iterate over registry.
The selectLoop seems to be working correctly. I suspect the issue is with k6. After updating to version 0.52.0, all objects were successfully deleted.Update:
Other changes I made along with the k6 update resolved the issue.
The problem occurs when using multiple deleters: the deleters processing the
last object+1
will complete and terminate the test prematurely.https://github.com/grafana/k6/issues/2804
k6 does not support runner synchronization.
Potential solutions:
test.Abort
. This approach mimics the loop mode.Oneshot
mode.If we know how much workers are using selector, we can set a shared counter and abort the test if all workers have received empty pointer from the selector.
Oneshot
mode #158