[SERVER-13692] db wide lock while performing mongorestore on a sized collection Created: 23/Apr/14  Updated: 30/Nov/14  Resolved: 30/Nov/14

Status: Closed
Project: Core Server
Component/s: Performance
Affects Version/s: 2.4.9
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Andrey Nikitin Assignee: Unassigned
Resolution: Incomplete Votes: 0
Labels: performance, replication
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Duplicate
Operating System: Linux
Steps To Reproduce:

just ran a mongorestore on the problematic database.
i cant share the dump file as in contains customer sensitive data.

Participants:

 Description   

Hi,

while preforming a mongorestore with a database of roughly 4GB, there was a big collection there (audit table) that when restoring it (and the indexes on the collection afterwards), the process took a lot of time - about 45 minutes and it made a huge impact and all of the nodes during that time thus making all of the mongo nodes respond extremely slow rendering the service unable to work with.
so my question is, is this a normal mongodb behavior? should i expect db-wide lock on every big restore ill be doing? if not what can cause the issue?

one thing to note is that we are using NFS storage, and i know mongodb does not work very well with it, but again should a restore of 4gb db take 45 minutes and lock the entire replica set even if its NFS?

logs attached - the restore took place on 17/04 around 2am and again later that day on 8-10am.



 Comments   
Comment by James O'Leary [ 23/Apr/14 ]

Hi Andrey,

Closing as per your request. An engineer will contact you presently on the new ticket.

Regards,
-Jim

Generated at Thu Feb 08 03:32:34 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.