[SERVER-3851] With 2.0.0, mongostat in a shared environment reports incorrect "delete" numbers Created: 14/Sep/11  Updated: 30/Mar/12  Resolved: 15/Sep/11

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: 2.0.0
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Didier Ernotte Assignee: Spencer Brody (Inactive)
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Centos 6.0


Operating System: Linux
Participants:

 Description   

I have a multi shared environment (3 shards), and I'm doing load test on it. I have an infinite loop inserting and deleting small document in a collection. I have a random integer shard key that allow me to spread the load on the 3 shards, and that works pretty well, except that when I monitor the mongos with mongostat I have a 12k/s insert and 12k/s delete rate, and each mongostat on shard gives me a 4k/s insert and 12k/s delete rate. The 12k/s on each does not make sense. My script does 1 insert and 1 delete in sequence in a infinite loop.



 Comments   
Comment by Eliot Horowitz (Inactive) [ 15/Sep/11 ]

non-shard key deletes are sent in parallel to all shards, so this seems to be working as expected

Comment by Didier Ernotte [ 14/Sep/11 ]

The delete is made on the index, which is not the shard key

Comment by Eliot Horowitz (Inactive) [ 14/Sep/11 ]

Are you doing the delete on the shard key?

Generated at Thu Feb 08 03:04:14 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.