Installed MID server but I'm getting a "duplicate name" error in the logs and the MID won't work

[tspitalnick]

I have a new, clean London dev instance and downloaded the mid, set up the install and it worked just fine. I gave my MID a totally unique name and it's logging in via my admin credential (I know, I know, but it's just a dev instance). I do the startup, the MID shows up in my instance, I validate, but then I get this:


10/30/18 09:14:51 (244) StartupSequencer WARNING *** WARNING *** Encountered error: [An active MID Server with a duplicate name detected.] in ensuring agent record on the instance. Retry...
10/30/18 09:14:51 (244) StartupSequencer Waiting to retry in 5 minutes. Attempt 1 of 3.
10/30/18 09:19:51 (879) StartupSequencer WARNING *** WARNING *** Encountered error: [An active MID Server with a duplicate name detected.] in ensuring agent record on the instance. Retry...
10/30/18 09:19:51 (880) StartupSequencer Waiting to retry in 5 minutes. Attempt 2 of 3.
10/30/18 09:24:52 (486) StartupSequencer WARNING *** WARNING *** Encountered error: [An active MID Server with a duplicate name detected.] in ensuring agent record on the instance. Retry...
10/30/18 09:24:52 (486) StartupSequencer Waiting to retry in 5 minutes. Attempt 3 of 3.

10/30/18 09:29:54 (424) StartupSequencer WARNING *** WARNING *** Encountered error: [An active MID Server with a duplicate name detected.] in ensuring agent record on the instance. Retry...
10/30/18 09:29:55 (210) StartupSequencer Stopping MID Server due to exceeding retry attempts.

Then it shuts down for good. I absolutely do not have another MID by the same name. I DO have another mid that still exists from a prior install, but it is confirmed that the service is inactive and definitely a different name.

Any ideas?

Edit: In the MID unresolved issues, this came up: "A MID Server with a duplicate name or sys_id was prevented from connecting. Install path: C:\agent-london. This issue must be manually resolved."

[Brandon Smith3]

Gotcha, try changing the name in the config file(put a -new on end) or something on it and then see if the record in the MID table updates. A name, is just that and nothing more, and should update the correct record in the table by sys_id. Once you see that its matched on the MID table, take the mid server back down, update the name in the config to the name you would like and start it back up. It should again modify the name in the MID table and stick. 

Just make sure you do not change the sys_id as this will create a new MID record. 

[tspitalnick]

So I just tried that. The MID "started" (meaning it's saying it's UP but the error still shows up) but the name did NOT change in the table.

[Chuck Tomasi]

Delete the table entry, restart the MID server and it should re-register with the instance with the correct name as put in the config.xml file.

[Brandon Smith3]

Interesting though, before you restart and delete the MID record, did the configuration parameter 'name' on the MID table record update to your new name? If this is a brand new mid with no other config parameters or other related records (ip ranges, clustering, app credentials, ect) I'd also agree with Chuck to delete and start a new record. Just be sure to fix the name and remove the sys_id in the config file in the agent folder. 

[tspitalnick]

Brandon: Changing the name parameter didn't update the existing record. I have to assume the error came in before that part happened.

Chuck: I stopped the service, deleted the table record, re-appended the "-new" to the "name" parameter, restarted the service. The "-new" name record showed up with a brand new sys_id on the table and that sys_id was updated on its own in the config for the sys_id parameter.

Error is still showing up.