What is the best way to move hyperic to a new server?

[tacheeze_hyperi]

I have at present got Hyperic installed on a vmware server for evaluation, and now we wish to put it onto its own dedicated server. Which is the best way of transferring the settings from the VMware installation to the new dedicated server?

Both servers are running WIN2k3 with the default Hyperic DB installation

[tacheeze_hyperi]

I got a full answer from a contact at hyperic, so posting it here if anyone else is wondering how to do it.

HQ runs PostGres SQL for its back end data source. Migrating all your server settings + data to a new server is simply a matter of dumping and re-importing the database. Of course if the new system you are migrating to has a different network address, all of the agents will have to be reset up to point to the new server ***

Detailed instructions on how to do this:
1. Stop current HQ Server
2. Take database dump*
3. Install HQ on new Server
4. Import database dump from step #2**
5. Install your license.xml in the new server.
6. Start new HQ Server

In step #3 it is ok to install a clean new instance of the HQ Server. This is because we will be importing the data from the old database into the newly installed database.

* Detailed instructions for step #2 (again, these only apply to the built-in HQ database):
1. Start the database by executing '<HQ Server>/bin/db-start.sh'
2. Take a database dump '<HQ Server>/bin/db-archive.sh'
3. Stop the database '<HQ Server>/bin/db-stop.sh'

In step #2, note the location of the database archive and copy that onto the new server.

** Detailed instructions for step #4 (built-in DB only):
1. Start the database '<HQ Server>/bin/db-start.sh'
2. Unzip the archive from dump process above 'gunzip $archive.gz'
3. Import the archive '<HQ Server>/bin/db-psql.sh < $archive'
4. Stop the database after successful import '<HQ Server>/bin/db- stop.sh'

*** Detailed instructions for reseting up agents.
1. The agent does not have to be manually reset up, instead you can edit the agent.properties file and pre-populate the values for agent- >server communication, push this file out to all your servers and restart the agents.
2. The agent can also be setup to connect to the server via a FQDN instead of an IP.

[Bri1]

I'm pretty sure you could also just copy the server-#.#.# directory. Hyperic is designed to be self-contained, so all the required files and configs are inside that folder.

[Bob_hyperic]

Hello,

I am interested in this operation as well. We have version 3.0.5 (open source)installed on C: drive. I need to move it to 😧 drive due to disk space issue, or to another server.

My question is:

1. I don't see db-archive.sh or similar script on the Windows box, but there are pg_dump or pg_dumpall available, can I use them to dump the database to a file, and restore it later on, the database name seems to be hqdb.
2. All files are included in c:\...\server-3.0.5, if simply copying this folder to anohter drive or server, does it work? HQ runs as a Windows service, it will look for the program to run in the original location.

Please help, thanks very much.

Bob

[jroh]

Just to add to your script.. as part of step 4

Under the new installation I had to drop the old hqdb, re-create it, then import.. in order for my old settings to be imported correctly.

server-3.0.5/bin/db-start.sh
server-3.0.5/bin/db-dropdb.sh hqdb
server-3.0.5/bin/db-createdb.sh hqdb
server-3.0.5/bin/db-psql.sh</tmp/hqdb-07.13.07.dump
server-3.0.5/bin/db-stop.sh

[ceefour]

Awesome!

Thank you tacheeze_hyperic !