weajun opened a new issue #7368:
URL: https://github.com/apache/incubator-doris/issues/7368


   ### Search before asking
   
   - [X] I had searched in the 
[issues](https://github.com/apache/incubator-doris/issues?q=is%3Aissue) and 
found no similar issues.
   
   
   ### Version
   
   0.14.13
   
   ### What's Wrong?
   
       
   in my doris cluster, tablet frequent balance, a tablet are added to the 
balance pending queue every second, but most of the balance tasks failed, few 
tasks can be successful. when success, the redundant replication will be 
deleted, but  the next second will still enter the pending balance queue. When 
I delete the table to which this tablet belongs, other one tablet will appear 
in the balance pending queue every second.   and this situation has been going 
on for many days. 
   
   I analyzed the cause of the problem. In my doris cluster, each be has two 
kinds of disk medium, ssd and hdd, and most of the data is used ssd. For 
historical reasons, there are very few tables in hdd, and these  hdd table 
occupies a very small space, only at the k level, but the hdd disk space is 
very large, at the T level. When the amount of data is small, the balance 
strategy is difficult to balance the data between all be, which means difficult 
to let all be are in the "MID" state. Check SHOW 
PROC'/cluster_balance/cluster_load_stat/HDD', there are always some bes in the 
"HIGN" state, and some bes in the "MID" or "LOW" state.
       
   
![image](https://user-images.githubusercontent.com/6142924/145541304-9479ed31-e33f-4d90-b84a-58fe01e2212a.png)
   
   
![image](https://user-images.githubusercontent.com/6142924/145545006-5428881c-277e-4e27-b480-c5b783e31846.png)
   
   
   ### What You Expected?
   
   A tablet, don’t need to be frequently balanced
   
   ### How to Reproduce?
   
   _No response_
   
   ### Anything Else?
   
   _No response_
   
   ### Are you willing to submit PR?
   
   - [X] Yes I am willing to submit a PR!
   
   ### Code of Conduct
   
   - [X] I agree to follow this project's [Code of 
Conduct](https://www.apache.org/foundation/policies/conduct)
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to