Failed to create a replication Partner with embeded Databse

[Migration User]

HI,
I ha ve a problem to install a new SEPM as replication partner with embeded database.
this message is retrieved in the console of the primary server.
**************
May 14, 2009 10:16:14 AM CEST: Retrieval of local changed data for replication requested by remote server XX.XX.XX.XX failed during retrieving data! [reason: Retry to aggregate data for 4 times and LiveUpdate is still downloading the contents. Let's skip this round replication and wait for next round.] [Site: XXX] [Server: XXX]

**************
ProductVersion=11.0.4014.26

[Migration User]

As per my knowledge there is LUALL process is running on either one of the partner. Please stop that process or wait for it to be completed and then try to communicate with the Rep Partner. Make sure Both the manager has the same version installed on it.


Ajit

[Migration User]

Thanks Ajitjha for your response.
There is no Liveupdate process on both of the 2 servers and the 2 server has the same version.

[Migration User]

the problem is not in the replication
problem in retrieving the data by your "main" server from DB
also please tell what the kind of DB this server is using

please delete replication partner accounts and remote site on those 2 servers
!!! on main server delete replication partner accounts and remote site account of new SEPM

then add replication partner on new SEPM (replication partner) - it should help

[Migration User]

thanks Viachaslau,
the first server is using an SQL 2000 sp4 database.
I tried this solution but it dont help.

[Migration User]

did you installed the second SEPM as replication partner to the first one?

[Migration User]

The configuration in the management server configuration wizard ends with the error : "Unable to connect to the server specified" and i tried with no success the KB 2008061813510348.

[pete]

hi,

LiveUpdate is still downloading the contents. Let's skip this round replication and wait for next round.] [Site: XXX] [Server: XXX]

its becuase LUALL was running at the time of the error.

Run manual replication, if you do not see any LU process in task manager.

Pete!

[Migration User]

if firewall is on - make an exception for 8443 port

[Migration User]

In Task Manager->Processes, make sure that the LUALL, LuComs,LuCallback.exe are not running in SEPM.

Any process that starts with Lu for liveupdate.

Cheers,

[Migration User]

thanks all
there is no firewall between the 2 servers and no process that start with LU is running.

[Migration User]

Hi,

      Please go through the link below ..It might help you in resolving this issue.



http://service1.symantec.com/SUPPORT/ent-security.nsf/docid/2008101309471148

"Unable to connect to the server specified" error during the replication of Symantec Endpoint Protection Manager

Question/Issue:

In this document the "source Manager" will be the Symantec Endpoint Protection Manager (SEPM) connected to the database that is to be replicated and "destination Manager" the Symantec Endpoint Protection Manager whose database will be created as a replication of the source Manager Database. After a Replication Partner is added the error "Unable to connect to the server specified" occurs. There may be other errors.



Symptoms:

The following errors occur:

Error "Unable to connect to the server specified" during the first replication on the destination Manager, while trying to add the replication partner via the Management Server Configuration Wizard. After this error the destination Manager installation is damaged due to the database replication failure.
Error "Retrieval of local changed data for replication requested by remote server x.x.x.x failed during retrieving data! [reason: Failed to retrieve Metadata - Java heap space]" in Admin > Server > Notification Area of the source Manager.
Error "java.lang.OutOfMemoryError: Java heap space" in the scm-server-0.log of the source Manager.


Cause:

The Symantec Endpoint Protection Manager is a Java-based application much of it's performance is related to the size of the "Java Heap". Modifying the settings that control the size of the Java Heap will improve the performance of a wide variety of the functions performed by the Symantec Endpoint Protection Manager.

Solution:

To adjust the Symantec Endpoint Protection Manager heap size:


Click Start> Run.
In the Run dialog box, type regedit
Press Enter.
Locate the following registry key:
HKLM\SYSTEM\CurrentControlSet\Services\semsrv\Parameters\

Locate the following keys:
JVM Option Number 0
JVM Option Number 1

Adjust the key values upward, and for optimal performance, match the key values.
For example, to create a 1 GB static heap, set: "JVM Option Number 0" to "-Xms512m", and set "JVM Option Number 1" to "-Xmx1024m".
Close Regedit.
Click Start> Settings> Control Panel> Administrative Tools
In the Services dialog box, right-click Symantec Endpoint Protection Manager, and then click Restart.

[Migration User]

Hi,

        Please let us know the status of the issue.

[Migration User]

I have been working with support for several weeks on this, and no resolution.  I'm trying to add my 3rd site, and the replication fails.  The firewall between the boxes is open, there is plenty of disk space, the java heap has been adjusted.  At supports request, I have rebooted - even added memory to the boxes.

About 1/2 way through the initial replication, I'm asked to accept the key of the other server again.  The Configuration wizard claims it is "replicating database", but the process always ends with "Unable to connect to the server specificed."  This is a lie, because I can see the replication request come in on the other machine.

[Migration User]

Can you telnet from one machine to the other over port 8443?

Are both these servers on the same subnet?  Or are they on different subnets?  Remote site?
 
I saw you indicated that the firewall between the 2 sites was open...  So, my question to you is about the actual link between the 2 boxes.  What is the "pipe" between them?  Is it a dedicated fiber provided by your ISP (router to router) or is it going over an internet connection/VPN, etc.?

I have a user, here at the office, that wanted to VPN into the office, after 2 months of troubleshooting and inability to connect, no matter what we did, and against what he wanted, I called his ISP.  Turned out, the ISP blocked VPN and many other ports at their level.  They indicated an additional 3$ monthly charge for allowing the ports for the connection to be established. 

[Migration User]

My first server and second server are 1500 miles apart, with a dedicated WAN link.  They are on different ip networks.  Replication between these two servers WORKS GREAT.

The third server is in the same facility as the first server.  However, they are on difference IP networks, seperated by a firewall device.  I am fully confident there are no communication restrictions between the two machines.  SEP communication actually happens - I see it in the log - but REPLICATION FAILS.

To answer your question - the connection between the two boxes should be about 100Mb.


On a side:  That's totally not cool re: the $3 extra to get ports unblocked. 

[Thomas K]

Here is a Microsoft KB that talks about  "Replication setup over a firewall".

http://support.microsoft.com/kb/164667

I hope this is helpful,

Thomas

[Migration User]

Thanks but that KB relates to MS SQL databases.  I'm using the Symantec embedded databases.
(the firewall between the two boxes is wide open anyway)

-AA

[Migration User]

Assuming this hasn't been resolved, can you alert us to the status please?  Do you have a CASE ID #? (original poster?)

Eric

[Migration User]

 Would suggest to open a new thread as this is too old ( 38 weeks ) thread.
Would also suggest attaching The scm-server-x.log and replication-x.log.

[Migration User]

Yeah, if that will prove helpful.

[Migration User]

I'm almost facing the same issue while creating a Replication Site. I've logged a call with the Symantec, however, the way I see how things are moving here on this Forum, I'm sorry to say that I'm disappointed.

My issue is that I try to create a Replication Site  <Site C> for replication with Primary Server (Site) <Site A>, which already hosts a Replication Site <Site B>. "Site A" and "Site B" work fine, however,  when I try to create "Site C" things work fine initially and the "Site C" shows up on SE Management Console on "Site A". When the Initital Replication of the DB on "Site C" start, it fails after some hours. I tried it twice and while I'm typing the so-called Replication of DB on "Site C" is running, which I'm sure is gonna fail after several hours. All the Sites are connected through 2Mbps P2P link.

The Error that pops up is as follows:
*********************************************
Synchronization to the remote site failed.
JZ006: CaughtIOException:
comsybase.jdbc2.jdbc.SybConnectionDeadException: JZ0C0:
Connection is already closed
**********************************************
It might be worthy to mention here that the SEPM Server at "Site A" is using two NICs (Say NIC1 and NIC2) with two different IP Addresses configured for two separate Subnets. The Replication between "Site A" and "Site C" will happen through, let's say through NIC1. There's no firewall in between the servers at "Site A" and "Site C" and they can communicate with each other perfectly.

I would also like to mention here that the Replication Partner  Properties for Both Site B (B Working Fine) and C shows the Last Replication Server as the IP configured on NIC2. I beleive that shouldn't be a problem, keeping in mind that "Site B" is working fine with these settings and Replication succeeds between Site A and B. 

I would really be gateful to all who prove to be helpful regarding this Issue. Till now Symantec Tech Support is of absolutely no help.

To my amusement, the Symantec Tech Support guys seem to be so "Naive" in providing Error Resolution, after all I've to keep in mind the Standards that Symantec stands for.

No, offence meant.