-
Notifications
You must be signed in to change notification settings - Fork 4.7k
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
HIVE-28626: Display MoveTask/StatsTask duration on the query summary #5544
base: master
Are you sure you want to change the base?
Conversation
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.
Leftover
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.
no, for me, an empty starting line is not expected in case of any methods, that's why I simply removed :)
@@ -569,7 +569,8 @@ public int execute() { | |||
} | |||
releaseLocks(tbd); | |||
} | |||
|
|||
long moveFilesDuration = perfLogger.getDuration(PerfLogger.FILE_MOVES); |
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.
Two questions about this:
1.: This method already contains a perfLogEnd call:
// if _blob_files_kept is present, use it to move the files. Else fall back to normal case.
if (moveFilesUsingManifestFile(fs, sourcePath, targetPath)) {
perfLogger.perfLogEnd("MoveTask", PerfLogger.FILE_MOVES);
return;
}
Does it stops the perfLogger? If yes, how that change will behave in that case?
2: perfLogEnd is already called at line 196. Can it affect the output of this getDuration call?
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.
- it doesn't stop, just registers and endTime for the duration of FILE_MOVES
- yeah, depending on the outcome, the ending time should still be registered anyways
possible refactoring is to fully encapsulate the whole movetask behavior to have a single begin + end call
fdda590
to
e83ab6b
Compare
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.
Thanx Laszlo for the PR, minor nits, rest LGTM
e83ab6b
to
6e17233
Compare
Change-Id: Ib37772938db5fdecf6507cdd43717a89f1e6250a
6e17233
to
692a5c3
Compare
Quality Gate passedIssues Measures |
What changes were proposed in this pull request?
Display MoveTask's and StatsTask's duration on the console.
Why are the changes needed?
Better observability.
Does this PR introduce any user-facing change?
Yes, query output on the console slightly changes.
Is the change a dependency upgrade?
No.
How was this patch tested?
On cluster. Unfortunately, due to HIVE-28627, the console message is broken, works only intermittently, but in the logs, the very same messages appear.