Geoff,
Posting it here as could not get any response on ICM forum.
I would like to understand in details on how the syncing process happens in a setup where there are 2 HDS servers, 2 loggers.  Also what can cause one of the HDS server to go out of sync even though loggers and one of the HDS is in sync.  I have had this happen and do not have a good explanation yet.
I want to look at this process in detail and also what order the tables get updated.
Thanks,
Hemal

Thanks Bill. I if you have a situation where the Logger A has up todate date and HDSA incorrect and Logger B incorrect data and also HDS B has incorrect data. What would be the best way to correct it ? Would you restart process on A to kick of the replication and update HDSA and also export A to B logger and do the same on B ?
Hemal

From: Cisco Developer Community Forums [mailto:cdicuser@developer.cisco.com]
Sent: Thursday, December 20, 2012 3:40 PM
To: cdicuser@developer.cisco.com
Subject: New Message from Bill Webb in Customer Voice Portal (CVP) - CVP - All Versions: RE: HDS Sync process

Bill Webb has created a new message in the forum "CVP - All Versions": -------------------------------------------------------------- The Logger to HDS sync process is between a single Logger and a single HDS, and is exclusive of the Logger sync process between duplexed Loggers. There is a separate process (replication - rpl or hrpl now, I think) that controls the synchronization of data between a Logger and HDS.
So the standard setup for a duplexed Logger and HDS setup is that you have Logger A and HDS "A" set up to replicate, and then Logger B and HDS "B" to replicate. The HDSs know nothing of each other, and have no synchronization between them. This means the situation you describe is entirely possible, that is: both Loggers in "synch" and one HDS in synch, but the other HDS out of synch. It simply means the historical replication between the HDS and the Logger it is synching to has failed somehow.
As for the order of replication of the tables, I used to know that, but I think it is pretty simple in theory (like alphabetical by table name), but I also believe it cycles through each table in this order in smaller time chunks until it is fully replicated, then it performs differential synchs using the same method.
- Bill
--
To respond to this post, please click the following link: http://developer.cisco.com/web/cvp/forums/-/message_boards/view_message/9626782 or simply reply to this email.