Backup/Restore Upgrade issues

This is not meant as a criticism. The goal is to help save other admins time/energy/frustration in repeating troubleshooting. We’ve done just one backup --> restore upgrade, and the list below covers the unexpected issues we encountered.

It would be nice to have a pinned, non-update expiring post for everyone’s reference.

Here’s what we found:

  1. After the restore and reboot, endpoints were failing registration. Spent hours on this and finally opened a ticket. Two days later when we fired the VM back up to roll forward with support monitoring, the problem had gone away. Do not know what caused or fixed this.
  2. The new system was at GMT -0 after the restore, even though the source system was Pacific Time (currently GMT -7).
  3. Conferences had been set for participant PIN and leader wait. The GUI reflected this, but the conferences were allowing users with without a PIN and without the leader.
  4. UCP was inaccessible until I added a second directory in User Manager.
  5. There was a recurring Symlink notification on the dashboard until we removed the Digium module.

From others:

After an upgrade, external address for global settings was set to NONE, at least thats what it pulled up in EPM. Added the proper external IP, when editing an extension I no longer get the error.

Need the exact version of the backup module, and probably useful to know the framework version you restored from.

  1. Not much to be done with this. My guess is that SIP bind ports were changed during the restore and you didn’t restart asterisk.
  2. If there is no ticket already, create one
  3. I recall seeing this recently, there may be a ticket already, but if not
  4. No idea what this is about.
  5. I presume the notification had instructions on how to resolve without removing the module

Hi Lorne-

It wasn’t my intention to create a work list for anyone or complain. Just try to save other folks some time. I also recognize that some of this could be specific to the system from which we upgraded, which was v13 and I think had been upgraded years ago from v12 using an upgrade script.

Backup Module: I believe it was the current version, We put that module on Edge on the source system and, at the time, Edge and non-Edge were the same version.

  1. We restarted the system several times over hours trying to get endpoints to register. I was on the desperate side to get it fixed and didn’t do a good job taking notes. I did get one to register by changing it to PJSIP and doing a factory reset on the handset. I couldn’t do this for all of them because some endpoint are remote. Plus when I did this we lost a BLM key indicator. The site has a lot of SPA514G phones with a few Sx05 models.
    Forum post with error: After 13 to 15 upgrade all endpoints are offline
  2. Okay.
  3. Okay.
  4. Me neither. UCP blew up with an error that I posted here in the forums:
    V13 to v15 upgrade: Issue with UCP
  5. It did not, no. The person who handled the support ticket for the endpoints not registering recognized it and told me to remove the Digium module to resolve. I didn’t save the exact syntax, sorry.

I think the intent is clear. I was just pointing to resources so we can ensure that known issues get addressed.

Symlink notification came back after a module update just now:

retrieve_conf resolved a symlink with
This is a notice to let you know that the original file was moved to /etc/asterisk/backup, there is nothing more you need to do

I either mis-remembered or the Digium module had re-installed. I’ve removed it for sure now.

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.