Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.

Database size discrepancy - replicated sites

Created: 22 Oct 2013 • Updated: 04 Nov 2013 | 12 comments
Brent.Noble's picture
This issue has been solved. See solution.

Hi guys,

Hoping someone can point me in the right direction here. Here's a high level view of the environment I'm dealing with:

- Large environment with 10000+ clients.
- Single domain
- Two sites with replication enabled.

Site 1
- Two SEPM servers. One primary and a standby (standby does not service clients unless there is a failure).
- Single SQL database

Site 2
- One primary SEPM server.
- Single SQL database.

The two primary SEPM servers are load balanced for clients, and are servicing roughly half each.

Here's my concern/issue. I've noticed the database for Site 1 (which has two SEPMs) is around 50GB, while the second site the database is around 90GB. Looking at the tables I can see that there are some differences, but the size difference seems to be spread amongst the whole database, so there's not any one table that appears to be bloated.

A couple of things I've noted so far:
- Content is being replicated
- Auto-replicate is enabled.

Are these two things likely to cause issues in terms of DB size? I know they can cause problems with replication itself but I'm not having any problems with replication.

Any other ideas?

I'm yet to log a case but I will if need be. Just after some advice first so I can attempt to self diagnose.

Thanks,
Brent

Comments 12 CommentsJump to latest comment

SMLatCST's picture

The main reason I can think of for the large difference in size would be if log replication is happening in only one direction.  Can you confirm if this is the case?

pete_4u2002's picture

you should also look at log retention at the each site.

also is the content revision same across the site?

Beppe's picture

By default the logs are not replicated, this should explain that difference.

Regards,

Giuseppe

Chetan Savade's picture

Hi,

Thank you for posting in Symantec community.

I tried to find out with senior technician what can be the possibilities & those are the same which are already suggested.

1) Replication of logs

2) Number of Content revision per SEPM.

3) Old packages if SEPM is upgraded.

In most of the scenarios Auto replication is not recommended though it's working fine in your case.

Chetan Savade
Sr.Technical Support Engineer, Endpoint Security
Enterprise Technical Support
CCNA | CCNP | MCSE | SCTS |

Don't forget to mark your thread as 'SOLVED' with the answer that best helps you.<

Brent.Noble's picture

Thanks for the replies guys, I'll investigate each of these and report back. At face value I couldnt see any differences in replication between the two sites.

Rafeeq's picture

Whats the number of content revisions stored? by default its 30, is it same?

Brent.Noble's picture

I can confirm that log replication is enabled for both sites and the content revision settings are also the same (30 days).

Anything else I can check? Is there a way to confirm that log replication is working correctly?

Brent.Noble's picture

I'll also add there are no errors in the replication log itself, however...

Entry taken from this morning (Replication-0.log) from SEPM server on Site_A:

2013-10-24 07:50:47.498 THREAD 68 FINE: ------------ Thread started --------------
2013-10-24 07:50:47.498 THREAD 68 WARNING: ReplicationHandler>> handleRequest: Begin...
2013-10-24 07:50:47.498 THREAD 68 WARNING: ReplicationHandler>> handleRequest: action replicateNow
2013-10-24 07:50:47.498 THREAD 68 WARNING: ReplicationHandler>> handleRequest: RpcData size=0
2013-10-24 07:50:47.498 THREAD 68 FINE: Borrow connection from pool.
2013-10-24 07:50:49.404 THREAD 68 FINE: Return connection to pool.
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationHandler>> replicateNow: Begin...
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationHandler>> replicateNow: doing replicateNow... siteName=Site_B, is2Way=false
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationManager>> replicateNow: Begin...
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationManager>> replicateNow: scheduling a thread...
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationTask>> ReplicationTask (2): local partner name=Site_A, id=C90F3D780A5D208A00EAE50CC661713D; remote partner name=Site_B, id=(replication now. so no id stored); twoWay=false, new thread name=ReplicationTask-Site_B-452; replicationNow=true
2013-10-24 07:50:49.404 THREAD 68 WARNING: (aa) LocalPartner>> subscribeToImmediatelyForOneTimeReplication: scheduling Replication Task - RemotePartner=Site_B, schedule=Schedule = [Day of Week = null; Frequency = null; Start Hour = 0; Start Minute = 0] This schedule starts in 0 days 2 hours 0 minutes and 0 seconds
2013-10-24 07:50:49.404 THREAD 68 WARNING: (aa) LocalPartner>> subscribeToImmediatelyForOneTimeReplication: scheduled Replication Task - RemotePartner=Site_B, schedule=Schedule = [Day of Week = null; Frequency = null; Start Hour = 0; Start Minute = 0] This schedule starts in 0 days 2 hours 0 minutes and 0 seconds, task=[name=ReplicationTask-Site_B-452 (29889550), isReplicationNow=true]
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationManager>> replicateNow: Done!
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationHandler>> replicateNow: Done!
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationHandler>> handleRequest: returning...
2013-10-24 07:50:49.404 THREAD 68 WARNING: ReplicationHandler>> handleRequest: Done!
2013-10-24 07:50:49.404 THREAD 68 FINE: ------------ Thread stopped --------------

Last (and only) entry from Replicatoin log on SEPM server in Site_B

2013-10-04 08:14:07.732 THREAD 84 WARNING: ReplicationHandler>> handleRequest: Begin...
2013-10-04 08:14:07.732 THREAD 84 WARNING: ReplicationHandler>> handleRequest: action replicateNow
2013-10-04 08:14:07.732 THREAD 84 WARNING: ReplicationHandler>> handleRequest: RpcData size=0
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationHandler>> replicateNow: Begin...
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationHandler>> replicateNow: doing replicateNow... siteName=Site_A, is2Way=false
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationManager>> replicateNow: Begin...
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationManager>> replicateNow: scheduling a thread...
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationTask>> ReplicationTask (2): local partner name=Site_B, id=583119570A5E20B4015A64E3595748C5; remote partner name=Site_A, id=(replication now. so no id stored); twoWay=false, new thread name=ReplicationTask-Site_A-2; replicationNow=true
2013-10-04 08:14:07.748 THREAD 84 WARNING: (aa) LocalPartner>> subscribeToImmediatelyForOneTimeReplication: scheduling Replication Task - RemotePartner=Site_A, schedule=Schedule = [Day of Week = null; Frequency = null; Start Hour = 0; Start Minute = 0] This schedule starts in 0 days 2 hours 0 minutes and 0 seconds
2013-10-04 08:14:07.748 THREAD 84 WARNING: (aa) LocalPartner>> subscribeToImmediatelyForOneTimeReplication: scheduled Replication Task - RemotePartner=Site_A, schedule=Schedule = [Day of Week = null; Frequency = null; Start Hour = 0; Start Minute = 0] This schedule starts in 0 days 2 hours 0 minutes and 0 seconds, task=[name=ReplicationTask-Site_A-2 (8140796), isReplicationNow=true]
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationManager>> replicateNow: Done!
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationHandler>> replicateNow: Done!
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationHandler>> handleRequest: returning...
2013-10-04 08:14:07.748 THREAD 84 WARNING: ReplicationHandler>> handleRequest: Done!

Based on these logs it looks like Site_B hasn't successfully replicated since the 4th of this month, where as Site_A is replicating regularly.

SEPM is reporting that replication succeeded for both sites. This suggests replication is not occuring from Site_B to Site_A, which might explain the database size difference (Site_B has the larger database)

Is my logic sound here or am I missing something?

SMLatCST's picture

That sounds like a fairly astute assessment of events to me (i.e. replication is only happening one way).

As a bit of background, this article attempts to explain how replication is meant to work:

http://www.symantec.com/docs/TECH152682

Perhaps you could try initiating a "Replicate Now" command from Site_B to see if this equals things out?

Brent.Noble's picture

Thanks for that article, always good to learn more about how these things work in the background.

Are you able to tell me what the difference is between the "Replication" and "ReplicationLocal" log file?

Is the ReplicationLocal file the replication log when replication is initiation from another site to the local site?

SMLatCST's picture

I've always understood the ReplicationRemote and ReplicationLocal logs to be more focussed versions of the Replication log, concentrating on the interactions with the remote partner and local processes, respectively.

Unfortunately, I can't find any reference to either in the SEP KB, so this is just my guess on the matter.

Brent.Noble's picture

To close this thread off, the database size has returned to normal, so I can only assume this was a SQL related issue (perhaps caused by some failed backups). Marking thread as resolved.

SOLUTION