RS primary still connects to old replica set members

Ops and Admin Replication
1

So I have the following mongo deployment running in aws:
FIrst replica set: mongors1x, mongors1y, mongors1z.
Second replica set: mongors2x, mongors2y, mongors2z.
There is also a config replica set and a mongos but it’s not important.

My current mission is to make a snapshot of mongors1x, launch the host from this snapshot and reconfigure the mongodb service on this host so it runs as a single node in it’s replica set separate from my regular deployment.

What I did:

  1. Launch a host from a snapshot of mongors1x. Let’s call it mongorestorers1.
  2. Launched server on it.
  3. Logged in and executed the following:
    cfg = rs.conf()
    cfg.memebers = [ { _id: 1, host: “mongorestorers1:27017”}]
    rs.reconfig(cfg, {force: true})

After reconfig it became PRIMARY. However when I look at the log, I can see that it still connects to mongors1x, mongors1y and mongors1z.

Is that normal?

3

Here are the entries from the log that tell me that my newly launched host talks to my main deployment that should be separate.
2021-11-05T06:04:29.738+0000 I ASIO [NetworkInterfaceASIO-ShardRegistry-0] Connecting to mongors2y:27017
2021-11-05T06:04:29.745+0000 I ASIO [NetworkInterfaceASIO-ShardRegistry-0] Successfully connected to mongors2y:27017, took 7ms (1 connections now open to mongors2y:27017)

4

Is this a typo or is that what was used while reconfigure

5

This is a typo. The rs.reconfigure() returns “ok” as 1.

1 Like