[SERVER-27188] how to control the connections from mongos to shards Created: 28/Nov/16  Updated: 12/Dec/16  Resolved: 12/Dec/16

Status: Closed
Project: Core Server
Component/s: Internal Code
Affects Version/s: 3.2.8
Fix Version/s: None

Type: Question Priority: Critical - P2
Reporter: kimmy-github Assignee: Kelsey Schubert
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: Zip Archive mongos-log.zip    
Participants:

 Description   

1、there are 10 same programs in different PC server,and the config parameters of one program is below:
mongo.order.connectionsPerHost=100
mongo.order.threadsAllowedToBlockForConnectionMultiplier=20
mongo.order.connectTimeout=10000
mongo.order.socketKeepAlive=true
mongo.order.socketTimeout=15000
mongo.order.writeConcern=MAJORITY

2、when there are 10000 clients visit the 10 same programs at the same time,in order to get data; the 3 mongos get below 2000 connections in total;but
one secondary of one shard get more than 6000 collections;I think there has some problems about the connections from mongos to shards;

3 、I thried to modify parameter "connPoolMaxShardedConnsPerHost=10" in one mongos or one secondary of shard,but it doesn't work;so I want to know how to control the connections from mongos to shards

4、The file "questions_TaskExecutorPool.log" is the log of the mongos,The file "shard.log" is the log of shard,The file "metrics.*" are diagnostic data
Thank you very much



 Comments   
Comment by Kelsey Schubert [ 12/Dec/16 ]

Hi kimmy-github,

Thank you for clarifying. From your response, I do not see anything to indicate a bug in the MongoDB server. For MongoDB-related support discussion please post on the mongodb-users group or Stack Overflow with the mongodb tag. A question like this involving more discussion would be best posted on the mongodb-users group.

Kind regards,
Thomas

Comment by kimmy-github [ 29/Nov/16 ]

Hi,Thomas Schubert:
yes,I read from secondaries,but I think the collections from mongos to shards is too much,only the clients produce below 2000 collections,but collections from mongos to shards have up 6000,as you know, the collections has upper limits at one node;I hope the collections from mongos to shard has collection pool and I can control it;
The problems caused by too many collections are it's easy to reach to the upper limit,then It can not recieve new collections,so clients will get error when they cant get new data;
So far,there is no performance impact ,but the clients will get error when they cant get new data;

Thank you very much

Comment by Kelsey Schubert [ 28/Nov/16 ]

Hi kimmy-github,

Thanks for opening this ticket. So we can get a better understanding of what is going on here, would you please elaborate on this point?

when there are 10000 clients visit the 10 same programs at the same time,in order to get data; the 3 mongos get below 2000 connections in total; but one secondary of one shard get more than 6000 collections;I think there has some problems about the connections from mongos to shards

What behavior do you expect, and what behavior would you prefer? What is the root issue that you are seeing, is there a performance impact that you can quantify? Are you reading from secondaries?

Thank you,
Thomas

Generated at Thu Feb 08 04:14:24 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.