[SERVER-36936] Mongos memory usage is too large(>7G) Created: 30/Aug/18  Updated: 25/Oct/18  Resolved: 25/Oct/18

Status: Closed
Project: Core Server
Component/s: None
Affects Version/s: 3.4.16
Fix Version/s: None

Type: Question Priority: Major - P3
Reporter: lkx Assignee: Kelsey Schubert
Resolution: Incomplete Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: Text File session_mem.log    
Participants:

 Description   

Test mongodb cluster performance with ycsb:

mongodb version: 3.4.0 or 3.4.16;

mongodb sharded cluster : 3 shardSver;

ycsb version:  0.12.0;

system : centos7;  memory: 48G; CPU: 24 core;

Operation sequence: 

  1. start sharded cluster;

  2. sh.enableSharding('ycsb');

  3. sh.shardCollection('ycsb.usertable',{_id:"hashed"});

  4. ycsb load data ( recordcount=20000,  operationcount=20000, threads=64);

  5. ycsb run scan (recordcount=20000, operationcount=20000, **scanproportion=1, threads=64),

At this time, it will occupy 7G of memory and will not be released for a long time.

 6. drop collection(ycsb.usertable),   and drop database(ycsb)

   Memory usage still has not decreased;



 Comments   
Comment by Kelsey Schubert [ 25/Oct/18 ]

Hi lkx,

We haven’t heard back from you for some time, so I’m going to mark this ticket as resolved. If this is still an issue for you, please provide additional information and we will reopen the ticket.

Regards,
Kelsey

Comment by Nick Brewer [ 01/Oct/18 ]

lkx Have you had a chance to collect the diagnostic.data with the heap profiler enabled?

Thanks,
-Nick

Comment by Nick Brewer [ 17/Sep/18 ]

lkx Based on the information you've provided, we're not seeing the sort of memory utilization that was described previously - memory usage appears to be roughly 2GB. To investigate this, we'll need the diagnostic.data information requested previously, once the heap profiler has been enabled on the mongos via:

mongos --setParameter heapProfilingEnabled=true

Thanks,
-Nick

Comment by lkx [ 15/Sep/18 ]

Yes!  I added the result of serverStatus to the attachment.

Comment by Nick Brewer [ 14/Sep/18 ]

lkx Are you still experiencing this issue? If so, we'll need the heap profiler information requested previously, so that we can better diagnose the cause.

Thanks,
-Nick

Generated at Thu Feb 08 04:44:30 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.