Hi,
I am looking to establishing a disaster recovery strategy as regards Rhythymx, this is for a client who wishes to set up a High Availability solution for thier website using an EVA( Enterprise virtual array).
They have two sites located in alternative geographical locations, they have the website binary files replicated across the EVA, and the databases are set up to do SQL log shipping. This seems to work well.
Now they would like to set up the WCM element. What they would like to do is set up rhythymx to to be fully replicated across the EVA, and be kept up to date. So in the event of fail over Rhythymx can continue to publish as normal, with out the need to do a full publish etc. However after initial investigation I found out that the only DR solution rhythymx supports is full database backup and restore. This doesn’t fit in with the clients needs, as like anything they want an instantaneous solution, and a fairly automatic process.
I was wondering if anybody has had any experience with setting up a DR solution for rhythymx, and what they managed to do.
Also does rhythymx have any white papers or documentation as to this regards?
Whilst Rhythmyx is running, no files (other than log files) will be updated on the Rhythmyx server. This means that in theory you can have a copy of RX in location A, and a copy of the same files in location B. By all means synchronise these copies occasionally using something like rsync.
If you are syncing, do however be aware of database connection settings between the server/database (there’s a handful of files that database settings are stored in - actually quite a few in Rhythmyx 5.x, and a lot less in 6.x).
Regarding databases - if you’ve a database in both locations and you use some sort of replication to keep them both in sync, as long as you’re not running Rhythmyx in each location at the same time and users connecting to it (you wouldn’t do that would you?) you’ll in effect be able to switch server “stacks” whenever you wish. There is nothing wrong with this configuration.
We don’t have any white papers (at least I’ve not seen any) about this, but Rhythmyx is fairly self-contained, and provided it can get a JDBC connection to its database you can replicate to your hearts content!
Hope this makes sense (and helps).
This question is addressed in the Rhythmyx Administration Manual, which was introduced in Rhythmyx Version 6.5. (The link is to the 6.5.2 release of the document.)
See Chapter 6, “System Management and Recovery”, specifically the section “Setting Up a Failover Server”. Contact your Percussion Software Sales Representative for details about licensing this server.
we are also planing for rhythmyx DR
we can only find a couple of config files with JDBC DB connection settings :
rxconfig/Installer/rxrepository.properties
rxconfig/Workflow/rxworkflow.properties
are these the only files we need to change if we move our rhythmyx repository from our live servers to our DR database servers ?
actually there are over 100 files with 1000 hardcoded references to the database repository so i think its time to start a new thread
http://forum.percussion.com/showthread.php?p=1495