[SERVER-44923] update is too slow when the chunk migration Created: 03/Dec/19  Updated: 27/Oct/23  Resolved: 04/Dec/19

Status: Closed
Project: Core Server
Component/s: Admin
Affects Version/s: 3.4.23
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: ck wild Assignee: Dmitry Agranat
Resolution: Community Answered Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: Zip Archive diag.zip     Text File mongos156.log     Text File mongos192.log     Text File mongos195.log     Text File mongos196.log     Text File mongos241195.log     Zip Archive shard2019112713301430slow.log.zip    
Backwards Compatibility: Fully Compatible
Operating System: ALL
Participants:

 Description   

hi  

    i found the update operation would be too slow when the balancer is running( from  2019-11-27T13:40:00+0800  to  2019-11-27T14:50:00+0800     ), so i want to  check the chunk migration would  influence the performance of the sharding cluster;

    after stopping the balancer, the update operation will be ok. 



 Comments   
Comment by Dmitry Agranat [ 04/Dec/19 ]

Hi hmyyj126@126.com,

Though the data you've uploaded does not cover the time slot you've mentioned (from 2019-11-27T13:40:00+0800 to 2019-11-27T14:50:00+0800), some operations latency might sometimes be impacted during chunk migration. In fact, I was able to see such occurrences during other times on November 27th, for example, between 09:12 and 10:32 UTC.

I've also noted the MongoDB version of the cluster is 3.4 which will be EOL in January. We recommend using the latest MongoDB version, which of today is 4.2.1. If you suspect a bug or have a feature request, please upgrade your cluster to the latest MongoDB version and we'll be happy to assist.

Regards,
Dima

Generated at Thu Feb 08 05:07:22 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.