TECHZEN Zenoss User Community ARCHIVE  

Control Center and backup/recovery details sought

Subject: Control Center and backup/recovery details sought
Author: Larry Virden
Posted: 2016-08-09 12:37

We have Zenoss Core 5.1.1 installed here at the moment.

We built a dev machine, got it working relatively well, and now want to create a production machine cloned from that dev machine.

We tried just cloning the VM. While some bits seemed to start up, not everything was working.

So I asked on the #zenoss IRC Chatroom and a couple of people indicated they had never successfully gotten that to work. Instead, they recommended, I should use Control Center's backup and recovery to create the new machine.

So, we strugged getting the backup to work. each time we use it, we get an error that says there is insuffiicent space. This despite the fact that we have over 100 GB available. Interestingly enough, however, is that after waiting quite a long time with no indication that anything had changed, we found a *.tgz file that was complete.

So, we rcp the file to the same location on the new machine. Since that machine already has a version of Zenoss Core installed, we go into Control Center and there is the file. We shot down the instance and services, then tell the Control Center software to recover. It warns us that once the process starts it can't be interrupted. We say go ahead.

After about 30 minutes of not seeing any progress indicated on the page, I hit refresh and notice that now my login has a mail message that says that recovery completed.

So then we go back to Control Center and tell it to start up Zenoss.core and the 40 children. We wait for 30 minutes. We look around - nothing has completed start up. I see no pointers to log files so that I can see what is going wrong.

Anyone have additional instructions that were forgotten in the Admin's guide

Thank you



Subject: Something that seems like it kind of works
Author: Larry Virden
Posted: 2016-08-09 14:04

So it turns out that a step that isn't documented is to add the new machine's host name to the hosts tab. I'm still not certain whether that is going to be sufficient - after I added the new host, I deleted the old host, and the application and services began, very slowly, to start up with the new host name for the endpoints, etc.

Along with this, we are contacting our Linux admin to add our new additional host aliases (zenoss5.newhost , etc.) into /etc/hosts and have that info pushed out everywhere it is needed, so that we can get into the system and start testing things out.

Any additional docs people have would be appreciated.

Anyone have a script that you run to search for strings across a Zenoss Core machine and through all its dockers I'd like something like that to look for other places that I need to update the host name...



< Previous
New Zenoss5 install getting 502 bad gateway when browsing to public enpoint
  Next
No graph data for Linux Server devices
>