napp-it omnios active directory connection refresh

[2cents]

Sometimes there is an issue with the DC not being online during a OmniOS reboot (after power outage) OmniOS is up before DC is ready. This leaves SMB active directory non-operational until either a AD rejoin, or a OmniOS reboot. [napp-it 18.12q1 / OmniOS v11 r151022cj / Server 2016].

• Is there a way to delay the AD connection on OmniOS boot to wait for a DC?
• Is there a way to run a service job or cron job that would reconnect without administrator involvement?
• What specific commands are needed short of the brute force reboot or manual rejoin?
• How can I simulate/force the lost connection for testing purposes without killing all the VMs and disconnecting the DC during reboot?

See this explanation using SmartOS: (Would this do the trick?)

Joining SmartOS to an Active Directory domain - SmartOS Documentation - SmartOS Wiki

 

[gea]

You may initiate a restart of the SMB service. This should re-connect the AD.
If this is an AiO (ESXi), you may control bootup order.
For testing purposes you should use ESXi with two vnics (one connected to AD), the other for clients. This allows a disable of the AD connection with user connectivity enabled.

You can also check the napp-it menu script for Services > SMB > AD join.
You may use this (with Expect) to re-join via script.

 

[2cents]

You may initiate a restart of the SMB service. This should re-connect the AD.
If this is an AiO (ESXi), you may control bootup order.
For testing purposes you should use ESXi with two vnics (one connected to AD), the other for clients. This allows a disable of the AD connection with user connectivity enabled.

You can also check the napp-it menu script for Services > SMB > AD join.
You may use this (with Expect) to re-join via script.

 

[2cents]

I've tried the SMB service restart (client/server for good measure) per your suggestion in another thread. I did this both from napp-it, and CLI multiple times and it does not re-establish the connection. Services > SMB > AD Join works immediately every time. However, it is not a long term solution and running it as a script is also not a long term solution since the password for domain join rights has to be stored somewhere.

It is/isn't an AIO. It is because the 2nd DC is run from a VM but it is in the ZFS data store under OmniOS (no room on ESXi datastore [RAM Disk]). However, the main DC is standalone hardware. Thus, the 2nd DC is never available on reboot prior to OminOS, and the problem only happens when there is an issue of the main DC not being available (not often, but it has/does happen).

Any idea why SMB service restart is not rejoining? If that worked, I could just run a job to do that.

Thanks for your input.