-
Notifications
You must be signed in to change notification settings - Fork 3.7k
[fix](partition rebalancer) fix migrate tablets between backends back and forth #39333
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
[fix](partition rebalancer) fix migrate tablets between backends back and forth #39333
Conversation
|
run buildall |
TPC-H: Total hot run time: 39903 ms |
TPC-DS: Total hot run time: 185809 ms |
ClickBench: Total hot run time: 30.24 s |
|
run buildall |
fe/fe-core/src/main/java/org/apache/doris/catalog/TabletInvertedIndex.java
Show resolved
Hide resolved
|
run buildall |
1 similar comment
|
run buildall |
deardeng
left a comment
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
|
PR approved by anyone and no changes requested. |
TPC-H: Total hot run time: 37411 ms |
TPC-DS: Total hot run time: 189504 ms |
ClickBench: Total hot run time: 30.99 s |
|
run buildall |
1 similar comment
|
run buildall |
TPC-H: Total hot run time: 37578 ms |
TPC-DS: Total hot run time: 184812 ms |
ClickBench: Total hot run time: 30.85 s |
|
run cloud_p0 |
|
run cloud_p1 |
2 similar comments
|
run cloud_p1 |
|
run cloud_p1 |
0cbc171 to
f863a2c
Compare
|
run buildall |
TPC-H: Total hot run time: 37633 ms |
TPC-DS: Total hot run time: 184864 ms |
ClickBench: Total hot run time: 30.29 s |
dataroaring
left a comment
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
|
PR approved by at least one committer and no changes requested. |
… and forth (apache#39333) BUG: partition rebalancer migrates tablets back and forth: move from A to B, then B to A, then A to B, ... . The reason is the counting tablet num of backends is incorrect. It doesn't considering the pending and running sched tasks. After these tasks finished, the tablet num will change. Fix: when calcuting the tablet num of backend, it should consider the in-progress moves which will change tablet num later.
… and forth (#39333) BUG: partition rebalancer migrates tablets back and forth: move from A to B, then B to A, then A to B, ... . The reason is the counting tablet num of backends is incorrect. It doesn't considering the pending and running sched tasks. After these tasks finished, the tablet num will change. Fix: when calcuting the tablet num of backend, it should consider the in-progress moves which will change tablet num later.
BUG: partition rebalancer migrates tablets back and forth: move from A to B, then B to A, then A to B, ... . The reason is the counting tablet num of backends is incorrect. It doesn't considering the pending and running sched tasks. After these tasks finished, the tablet num will change.
Fix: when calcuting the tablet num of backend, it should consider the in-progress moves which will change tablet num later.