-
Type: Task
-
Resolution: Gone away
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: Guides
-
Labels:None
-
1
Documentation on how to do point in time restore for a replicaset using mongodump is missing
The case is:
- I have a filesystem snapshot from time 1
- something bad happens at time 3
- my oplog cover operations starting from time 0
I can perform a point in time restore:
- dump the current oplog (mongodump -d local -c oplog.rs)
- mv dump/local/oplog.rs.bson dump/oplog.bson
- rm -rf dump/local
- mongorestore --oplogReplay --oplogLimit=<seconds>
Step 1 can be optimized in order to dump only the interested time range using mongodump option
--query { "ts" : { "$gte" : start_sec }, "ts" : { "$lte" : end_sec } }