A library and a binary for replaying MongoDB oplogs at a multiple of the original speed. Useful for stress testing databases using real world data.
First build Oplog Replay and put it on your GOPATH with:
go get github.com/Clever/oplog-replay/cmd
Then run it and pipe in a bson file containing the oplog you want to replay:
oplog-replay < oplog.rs.bson
You can also specify the following flags:
| flag | default | description |
|---|---|---|
--speed |
1 |
Multiplier for playback speed. |
--host |
localhost |
Host that the oplog will be replayed against. |
--path |
/dev/stdin |
Oplog file to replay |
Include it in your code: include "github.com/Clever/oplog-replay/replay"
And call it as follows: replay.ReplayOplog(r io.Reader, float64 speed, host string)
You can get an oplog dump by specifying the collection directly:
mongodump --db local --collection oplog.rs
A --query flag can be specified to get only certain oplog entries.
Please view the dev-handbook for instructions.