How to restart hosts that are down?

[MartinLoeper (Martin Löper)]

Just a quick question: Is it possible to force a restart of a specific replica member in Atlas?

We started a backup restore last Friday and it is stuck until now because one of the hosts did not come up after it was shut down for the restore process. A blue bar is displayed in Atlas which says “We are deploying your changes”. We are not able to change anything because the restore must be finished first.

However, the restore is apparently stuck waiting for the host which is down. Is it possible to force a restart or to get another AWS instance spawning up to replace the one which is down?

Best,
Martin

[Stennie]

Hi @MartinLoeper,

There currently is no API for directly restarting a cluster member - Atlas should take care of that automatically if one is unhealthy.

Please contact Atlas support for assistance with operational issues like this.

Regards,
Stennie

[MartinLoeper (Martin Löper)]

Hi @Stennie,

thanks for your answer.
I know that everything should be recovered automatically.
However, the host is still down and the organization unfortunately has no active support plan.

I assume there is no other option than subscribing to a developer plan in this case?
I understand that this is an operational issue and should be discussed with support in the first place.
However, under our current circumstances I wanted to check first if there isn’t an option how I can deal with it manually without support being involved.

Best,
Martin

[Stennie]

@MartinLoeper

I assume there is no other option than subscribing to a developer plan in this case?

Hi Martin,

You can still contact the Atlas support team for operational support questions via the in-app chat. The Atlas support link I provided earlier should show this as an option when you are logged into your Atlas account, but if you are having trouble with that link, the relevant instructions are also on Request Atlas Support.

Developer and Premium support plans include private case management with severities & SLA, broader scope for questions, and 24x7 coverage for cases with Blocker and Critical severity. For more information see the MongoDB Support Policy.

Regards,
Stennie