[DOCS-12474] Update movePrimary and dropDatabase warnings to also include all shard nodes Created: 14/Feb/19  Updated: 30/Oct/23  Resolved: 19/Aug/19

Status: Closed
Project: Documentation
Component/s: manual, Server
Affects Version/s: None
Fix Version/s: Server_Docs_20231030

Type: Bug Priority: Major - P3
Reporter: Randolph Tan Assignee: Kay Kim (Inactive)
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Participants:
Days since reply: 4 years, 25 weeks, 1 day ago
Epic Link: DOCS: 4.2 Server/Tools
Story Points: 0.25

 Description   

Description

Now that $out can be doing inserts on a remote shard, it can also be affected by the issues with movePrimary and dropDatabase. Note that the flushRouter command will also need to be sent to secondaries as well.

Scope of changes

  • Investigate behavior mentioned above
  • Update documentation for 4.2 only

 



 Comments   
Comment by Kay Kim (Inactive) [ 20/Aug/19 ]

No worries. Hope you are doing well. Kay

Comment by 章 黒鉄 [ 20/Aug/19 ]

Thank you Kay. I appreciated that was worked out so quickly, and in this week following the GA release no less.

Akira

Comment by Kay Kim (Inactive) [ 19/Aug/19 ]

Updated. I did a quick cursory check, and I think this is part of SERVER-32198

Comment by Githook User [ 19/Aug/19 ]

Author:

{'name': 'Kay Kim', 'email': 'kay.kim@10gen.com', 'username': 'kay-kim'}

Message: DOCS-12474: 4.2 clarify dropDatabase/movePrimary cache clearing change
Branch: master
https://github.com/mongodb/docs/commit/61ecc5819da2ec2fc753db912e21984f123cacbd

Comment by Kay Kim (Inactive) [ 19/Aug/19 ]

I'll double check. I would imagine that $merge as well but will find out if broader issue.

Comment by 章 黒鉄 [ 19/Aug/19 ]

Thanks, appreciated Kay. I hope that the issue is less broad than I thought when I first read just the release notes.

May I also ask is it just the $out agg operator that can possibly be mislead by the cache after a db namespace is reused? I did some JIRA searching but failed to find a match for the underlying cause.

Akira

Comment by Kay Kim (Inactive) [ 19/Aug/19 ]

Hey akira – I think it's just that I shortened it. I'll reopen this issue to add clarification.

Comment by 章 黒鉄 [ 19/Aug/19 ]

Hi Kay.

The two pages for the dropDatabase and movePrimary command only warn of the issue, and recommend the workaround of restarting all shard mongod and mongos nodes, if dropping and then re-using the same database name.

But the 4.2 release notes page removes that 'only if re-using the same db namespace' clause and just says the workaround is needed anytime the dropDatabase or movePrimary are used.

What is the issue that makes it necessary to say that any dropDatabase or movePrimary command requires the workaround? Or is it just that the "dropDatabase and movePrimary" section in the 4.2 release note page shortened the information a bit too much?

Cheers,

Akira

Comment by Githook User [ 05/Aug/19 ]

Author:

{'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}

Message: DOCS-12474: 4.2 dropDatabase/movePrimary post-step
Branch: master
https://github.com/mongodb/docs/commit/dda19b8d4f70464002560a48fb893d4b0f801c8b

Generated at Thu Feb 08 08:05:19 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.