asked on

Location of replication errors on Domino server

We have a few users using Notes 6.5.4 on their Windows XP SP2 laptops, replicating over the Internet using their cable or DSL connection to our Domino 6.5.4 FP1 server. Their replication receiving e-mails always works with no problems, however they get occaisonal timeouts when replicating large e-mails back to the server. It doesn't seem to happen at specific times of the day. They get the timeouts during the middle of the day, or late at night when no one is on the network.

I don't think this is a Domino problem, but I'm wondering if there is a lot file somewhere in Domino where I can get an error message. Does this exist? Is it something that I might need to enable?

Thanks, and please let me know if you need more information.

SysExpert

This info is stored locally on the users log.nsf file, so I do not see any easy way to get to it.

Other options.
1) Increase the TCPIP timeout in the User preferences= ports - from 5 to 15 seconds.

2) Check the MTU size on the users having issues.

You are probably correct that it is a network issue , but the above may help.

I hope this helps !

marilyng

I have this problem from my virus scanner on large e-mails. If it is that, you can disable the virus scanner ability to check outbound e-mails.

Reducer01

ASKER

I was able to have the user send me a copy of his log.nsf file. Not much there, just a generic error saying unable to replicate his database. Is there a way to turn up the logging on the client to get more information?

Thanks again.

Reducer01

ASKER

I figured out how to up the logging capabilites on the Notes client. I captured lots of info, but this seems to be the problem:

(244-204 [242]) WRITE_OBJECT(REP86256B4B:005A7B00-RRV0001ECD6,0x8000 at 0x68000): 1157 ms. [32854+16=32870]
(245-206 [243]) WRITE_OBJECT(REP86256B4B:005A7B00-RRV0001ECD6,0x8000 at 0x70000): 1281 ms. [32854+16=32870]
(246-207 [244]) WRITE_OBJECT(REP86256B4B:005A7B00-RRV0001ECD6,0x8000 at 0x78000): 2516 ms. [32854+16=32870]
(247-209 [245]) WRITE_OBJECT(REP86256B4B:005A7B00-RRV0001ECD6,0x8000 at 0x80000): 1125 ms. [32854+16=32870]
(248-210 [246]) WRITE_OBJECT(REP86256B4B:005A7B00-RRV0001ECD6,0x8000 at 0x88000): 828 ms. [31920+0=31920] (Remote system no longer responding)

SysExpert

Can't tell for sure, but it looks like a slow connection and maybe TCPIP timed out.

I would increas the timeout in the ports section if the ser preferences to 15 seconds

I hope this helps !

marilyng

Agreed. It looks like it's trying to collect a stream in blocks. Or its the virus checker taking too long to step through the file, which is why it's trying to read it in blocks.

86256B4B:005A7B00 << this is the database replica ID
RRV0001ECD6 << this should be the note ID, but this is the wrong number for a note id ( it should look something like this: NT000020FA.

So, I wonder if you can work backwards and find the database, check the user sessions, and maybe narrow down what the user is trying to download or receive.

Reducer01

ASKER

We changed the timeout to 15 seconds, but that didn't help.

I'm hesistant to start looking at the AV software because we've been on the same version for ages with zero problems, but if we can't narrow it down in Domino/Notes, then I'll be glad to start looking there.

The database is the user's mail.nsf file. How do I go about checking user sessions? I'm probably familiar with the process, but I haven't heard that term for it before. The user is trying to replicate an e-mail with a PDF attachment from his laptop over his DSL connection to our Domino server. After taking with him some more, it seems that it fails most on PDF files, while other largish attachments (.TIF, .DOC, etc.) go through just fine.

Thanks again for the help!

SysExpert

1) That could be an AV issue.

It is worth turning off scanning for email or in general, just to see if the problem goes away.

marilyng

Agreed. Anytime this happened to me, it was the virus scanner timing out on the local client. And I would get a "replication error - replication failed, unable to connect... " error. Are you client virus scanners on laptop linked to the virus server?

It would depend on the virus software to change the timeout, or the preferences, i.e. don't scan .pdf documents.

Reducer01

ASKER

I disabled Symantec on the client PC. Same error. :(

marilyng

Ok, so it's probably the tcp/ip timeout.

From Domino Administrator:

TCP/IP error messages -- Client or server

The establishment of the connection took longer than the expected default of 5 seconds. This can happen when the connection is over a dial-on-demand ISDN modem connection, remote bridge, or router. From the Port Setup dialog box, increase the TCP/IP connection-time-out interval. On a normal LAN, it is best to enter a value of no greater than 10 seconds, as the client or server won't retry the connection until the timer has expired.
To access the Port Setup dialog on a Notes client, use File - Preferences - User Preferences and click Ports. To access this dialog box for a Domino server, use the Domino Administrator's Configuration tab and select Server - Setup Ports from the Tools pane.
Once in the Port Setup dialog box, select the TCP/IP port and click the port name Options button.

SOLUTION

SysExpert

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Reducer01

ASKER

I think we're getting there! I changed the timeout to 10 seconds and the replication went fine. I'll be trying it on a couple other machines that are having problems and then report back.

Thanks for everyone's help.

marilyng

Yippee! Glad it seems to be working.

Reducer01

ASKER

I think we got our hopes up too soon. The user began having problems again this afternoon.

ASKER CERTIFIED SOLUTION

marilyng

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Reducer01