[SERVER-4165] Too Many attempts to update config error... Created: 27/Oct/11  Updated: 11/Jul/16  Resolved: 28/Oct/11

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

Type: Bug Priority: Major - P3
Reporter: Matthew Osterhaus Assignee: Greg Studer
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Sharded mongod cluster. Consists of 2 shards. Each shard consists of 3 server replica set. we have 2x mongos process running on 2 seperate servers.


Operating System: ALL
Participants:

 Description   

Running a sharded cluster, consisting of 2 shards. We are using 2 mongos processes.

MONGO ACCESS PROFILE:
We start a job that runs for about an hour and half and makes a lot of reads to mongo (no writes). reads are of the order a several thousand per seconds). I observe that the errors occur at the beginning of the job rather than being spread over the entire time period that the job runs

We are running 2x mongos processes. I'm seeing errors like the following:

ERRORS

2011-10-27 18:30:26,133 [btp-5] ERROR com.gilt.service.user.recommendation.SaleRecommendationHandlerDefault - error getting recs for userguid: e1e3b450-3491-012c-c778-0015177451fa
com.mongodb.MongoException: too many attempts to update config, failing
at com.mongodb.MongoException.parse(MongoException.java:82)
at com.mongodb.DBApiLayer$MyCollection.__find(DBApiLayer.java:303)
at com.mongodb.DBCursor._check(DBCursor.java:356)
at com.mongodb.DBCursor._hasNext(DBCursor.java:486)
at com.mongodb.DBCursor.hasNext(DBCursor.java:511)
at com.gilt.algo.commons.data.dao.mongo.UserBrandActivityDao.getBrandActivityProfileAsList(UserBrandActivityDao.java:71)
at com.gilt.service.user.recommendation.service.UserProfileServiceImpl.loadProfile(UserProfileServiceImpl.java:61)
at com.gilt.service.user.recommendation.service.UserProfileServiceImpl.getProfile(UserProfileServiceImpl.java:41)
at com.gilt.service.user.recommendation.service.SaleRecommendationServiceImpl.salesByBrands(SaleRecommendationServiceImpl.java:51)
at com.gilt.service.user.recommendation.SaleRecommendationHandlerDe



 Comments   
Comment by Aayush [ 12/Jan/12 ]

Thanks for the quick response Eliot.

Comment by Eliot Horowitz (Inactive) [ 12/Jan/12 ]

I would recommend upgrading the whole cluster to 2.0.2

Comment by Eliot Horowitz (Inactive) [ 12/Jan/12 ]

I would recommend upgrading the whole cluster to 2.0.2

Comment by Aayush [ 12/Jan/12 ]

we are running into this similar issue, and this exception from mongo is causing 500 errors on our Api.
What I want to know is, where exactly is the problem. Upgrading the entire cluster is a more time consuming task.
Is the issue with the router ( mongos ) or is it with the configserver ( mongod ) is it with the latest Java driver?

We want to upgrade the entire cluster, but due to time constraints, I was hoping to get a response about which part of the cluster is broken, and if we upgrade only that, will it solve the problem. ALso, it would be great to get a definitive answer about the fix version?
A quick response would be appreciated.

Thanks.

Comment by Eliot Horowitz (Inactive) [ 28/Oct/11 ]

I think its fixed in 2.6.5
If not - can follow up in JAVA

Comment by Matthew Osterhaus [ 28/Oct/11 ]

Looks like the update fixed the errors we were having. Jobs running against mongo today look much better on that front.

Looks like we are also seeing some null pointer exceptions like folks were seeing here:
https://github.com/aboekhoff/congomongo/issues/30

You know if this was fixed in 2.6.5? (was a bit unclear to me from the responses in that thread.)

Original issue is fixed. Thanks again Eliot!

Comment by Eliot Horowitz (Inactive) [ 27/Oct/11 ]

I believe you are seeing SERVER-3889
Can you try upgrading to 2.0.1

Generated at Thu Feb 08 03:05:09 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.