[SERVER-9319] Allow to disable secondaries when they are behinded for a long time. Created: 11/Apr/13  Updated: 11/Apr/13  Resolved: 11/Apr/13

Status: Closed
Project: Core Server
Component/s: Replication
Affects Version/s: 2.2.3
Fix Version/s: None

Type: New Feature Priority: Major - P3
Reporter: stronglee Assignee: Unassigned
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Linux


Issue Links:
Duplicate
duplicates SERVER-4935 Mark node Recovering when replication... Closed
Participants:

 Description   

I deployed a sharding cluster with 4 replica sets. Each rs contains a primary and a secondary and an arbiter. I insert data into primary and make queries on secondaries by using read preferrence "SecondaryPreferred".
However, in some circumstance,my secondaries lag behind for a long time(eg. serveral hours) because of heavy load or hardware maintenance. In this circumstance, I actually don't want my app to read data from secondries becuse data in them are quite old. I prefer them to read from primary instead.

So. Is it possible to add a config to specify a value, when the secondary find that it's
behind more than this time span, it will turn to a "CATCH UP" status so that it's not a valid node temporary. And then "SecondaryPreferred" reads will be routed to primaries.



 Comments   
Comment by Scott Hernandez (Inactive) [ 11/Apr/13 ]

This is on the roadmap and here SERVER-4935

Generated at Thu Feb 08 03:20:03 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.