[SERVER-43375] MongoS should differentiate between temporary DNS failures and permanent authentication failures Created: 19/Sep/19  Updated: 29/Oct/23  Resolved: 14/Jun/22

Status: Closed
Project: Core Server
Component/s: None
Affects Version/s: None
Fix Version/s: 5.0.0

Type: Improvement Priority: Major - P3
Reporter: Ivan Grigolon Assignee: [DO NOT USE] Backlog - Sharding NYC
Resolution: Fixed Votes: 1
Labels: sharding-nyc-subteam2
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
related to SERVER-53287 Improve cluster/mongos health observa... Closed
Assigned Teams:
Sharding NYC
Backwards Compatibility: Fully Compatible
Participants:
Case:
Story Points: 5

 Description   

Opening this to improve how the drivers handle connections to a mongos that is experiencing external authentication problems (such as networking problems to connect to a KDC)

The MongoDB Driver should try a different MongoS if it has contacted a MongoS that cannot speak to the external authentication backend.



 Comments   
Comment by Andy Schwerin [ 15/Jun/22 ]

Do I remember correctly that mongod and mongos do not connect to the KDC themselves, but that the driver does all communication with the kdc? As such, changes to mongod and mongos cannot be used to detect network issues with the kdc.

Comment by Lamont Nelson [ 14/Jun/22 ]

Closing since this was addressed with the Health Manager: https://www.mongodb.com/docs/manual/administration/health-managers/

Comment by Garaudy Etienne [ 17/Dec/20 ]

Made SERVER-53287 to investigate this type of issue. 

Comment by Ted Gulati [ 21/Aug/20 ]

We had a similar issue again and I was wondering if you are planning to schedule this ticket in the near future.

Generated at Thu Feb 08 05:03:01 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.