[SERVER-34513] Caught WriteConflictException doing plan execution on collection Created: 17/Apr/18  Updated: 04/Sep/18  Resolved: 06/Aug/18

Status: Closed
Project: Core Server
Component/s: Logging, Querying, Write Ops
Affects Version/s: 3.6.3
Fix Version/s: None

Type: Bug Priority: Minor - P4
Reporter: Eldar Hrelja Assignee: Nick Brewer
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: File diagnostic.tar.gz     File logs.tar.gz     HTML File mongo_error    
Operating System: ALL
Participants:

 Description   

Dear Mongo,

I'm writing regarding mongo WriteConflictException. We're using this server in production and problem happens to often.

Here is what we got:

  • OS --> Debian Stretch 9.4
  • RAM --> 125 GB
  • CPU --> Intel Xeon CPU E5-2620 v4 2.10GHz (32 cores)
  • HDD --> 916 GB (we're not using XFS, instead we're using EXT4)
  • MongoDB --> Version 3.6.3
  • Installation is done following this manual (https://docs.mongodb.com/manual/tutorial/install-mongodb-on-debian), package libssl1.0.0:amd64 (1.0.1t-1+deb8u7) is also installed.
  • Mongo service is started using systemd service with numactl

In attachment I'm providing you a part of log that looks suspicious to me.

If you need any additional information, please let me know.

Thank you very much



 Comments   
Comment by Nick Brewer [ 06/Aug/18 ]

eldarhrelja It looks like our troubleshooting of this ticket fell through the cracks - I'm sorry about that. Looking at the diagnostic data and logs you've provided, you're correct that there doesn't appear to be any cause for concern, as the conflicting operations should retry without issue, and the conflicts are not causing any meaningful impact on performance. 

-Nick

Comment by Eldar Hrelja [ 03/May/18 ]

Dear Bruce,

 

Thank you very much for your quick support and sorry for my late reply. For now we're concern only with the message in the mongod log. Now I'm using logrotate setup to rotate mongo log files every 1 hour and this error occurs every few minutes per one log file.

We've been setup everything from scratch, new installation of our platform without any upgrades including mongo service. As we can notice for now, this doesn't have an impact on our application and most mongo connections are related with Java Android applications through API.

 

Here is diagnostic.data and logs in attachment

 

Kind regards

 

 

 diagnostic.tar.gz

logs.tar.gz

Comment by Bruce Lucas (Inactive) [ 17/Apr/18 ]

Hi Eldar,

Thanks for your report. Can you clarify what impact this issue has on your application, or is your concern only with the message in the mongod log? How often does this occur? Were there any changes, such as upgrade of mongod or changes in the application, when you started seeing this issue?

So that we can investigate further, can you please upload to this secure private portal the following for the affected mongod node:

  • the archived content of $dbpath/diagnostic.data
  • the most recent few days or more worth of mongod log files, including several instances of the error message.

Thanks,
Bruce

Generated at Thu Feb 08 04:36:55 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.