[SERVER-4896] Mongo flush critical alert Created: 07/Feb/12  Updated: 14/May/14  Resolved: 09/Jul/12

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

Type: Question Priority: Minor - P4
Reporter: Aggie Filarska Assignee: Scott Hernandez (Inactive)
Resolution: Done Votes: 0
Labels: performance
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Red Hat Enterprise Linux Server release 5.7 (Tikanga). Three node replica set.


Backwards Compatibility: Minor Change
Participants:

 Description   

We have three node production replica set. We received critical flush alert when we were adding images to product catalog, what increased from ~150k MB to ~330KMB the product catalog collection.
The primary flush took 5ms at about that time where the secondary and tertiary took +500ms.
Currently, we have Mongo flush critical alert set to 400ms.

We would like to know why it takes much more time between Primary/master and slaves.
Should we increase current threshold for that alert?

Any information and help is appreciated.
Thank you,
Aggie Filarska



 Comments   
Comment by Eliot Horowitz (Inactive) [ 07/Feb/12 ]

That can be totally normal based on read/write patterns and generally not an issue unless overall io on a box is nearing limits.

Comment by Aggie Filarska [ 07/Feb/12 ]

Hi Scott,
By default Mongo ‘flushes to disk’ every 60 seconds.
Usually it takes about 5ms to flush and we noticed that when we added more data to our collection, the primary flush still took 5ms but secondary and tertiary flush took over 500ms.
We would like to know what is the possible impact on the performance when this happens and how to prevent it.

Thank you,
Aggie

Comment by Scott Hernandez (Inactive) [ 07/Feb/12 ]

I'm not sure what you mean byt "flush alert". Can you describe that?

It sounds like this is unrelated to java. I'm going to move this to the SERVER project now.

Generated at Thu Feb 08 03:07:17 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.