Discussion Locked

Arcserve Problem

0 helpful
19 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.
I am having a problem at a client site, the backup fails to run nightly. The server is Netware 4.11, running ArcServe 6.10. Whenever a scheduled job runs to backup the server volumes it fails with error "E3123 FALIED TO LOGIN TO TSA... ffdffd7" Ihave looked this up with Computer Associates and have tried there fix which was that the TSA directory did not exist, I deleted this dir and recreated. I have rebooted the server several times, the admin account is allowed to login multiple times (no restrictions)still no resolution. Would like to know if anyone has experienced this or can help.

19 replies

by jcparkerconsulting

I have seen this before. What is usually happening is that the TSAx.nlm is not loading or the process is failing to free enough SPX connections for the remote server to load. Increase your SPX connections (SErver Manager - communications), next place the proper TSA agent in the autoexec.ncf file to auto load. For 4.11 I beleave it is TSA411.NLM.

If the above fails, use the config utility (available from support.novell.com/misc/patlst.htm) and use the Config Reader (available at same location) to check the dates of the installed files. Some of the NLM's such as clib being outdated or not at the same level as the TSA .NLM can cause this problem.

by p. dickason, cne, mcse, cca

Hi,
In addition to the first answer, don't forget to load TSANDS.NLM also if you want to back up directory services. TSA411 is the target service agent for the file system only.
Hope this helps
Pete

by kuiphoff

It's still true that the MOST common reason for this failure is the password to the admin account has been changed. Every job that is set up has the password stored in it so the jobs now fail to authenticate. If this is the case, you need to modify all of the jobs with the new password.

The most common way to avoid this problem is to create a seperate user ID (with appropriate supervisor rights) that is only used for administering backups and recreate all of the jobs using this ID. Set the password to never expire and guard that password with your life. If this server only backs up itself then set concurrent connections to at least 4, if it backs up other servers set it to at least 10.

by boak

Do you have TSA.NLM and/or TSANDS.NLM Loaded. TSA is Target Service Agent, Arcserve uses that to login and backup the NDS. If that does not work try loading SMDR.NLM, Netware Data Requestor.

Hope it works.

by miller6010

I have seen a couple of problems regarding backing up remote servers with Arcserve that might help.
1. The TSA nlm's TSA410.nlm and TSANDS.nlm shold be the same version on both the backup server and remote server.
2. If SMDR.NLM does not load youcan get this
error. SMDR requires a specifix SAP type when backing up across segments. Check CA's website for the proper SAP number and verify it is not filtered out.

by cogtek

The TSA and TSANDS NLM's are the first place to look. If they are not loaded, not loading correctly or the wrong version, you will have these problems.
The second place to look is the actual Backup job you are running. You have to provide a Login account and password for each device to be backed up. Multiple volumes, multiples servers, Directory Services, all are considered different devices and must be verified seperately. Check the Job Queue tab in ArcServe Manager and you will notice that for each item you highlight in the Source tab, you must provide authentication.
The NDS authentication is often overlooked.
Try running a job that does not include the NDS and see if the error still occurs.

by jeff_hurst

It may be to easy.
the login for admin must be dinstinguished
CN=Admin.O=whatever

CA never states this anywhere and it caused me a problem with backups failing because of improper login.

by pdavis

I've also ran into this problem. It took a while, but we figured out that the password had changed for the user creating the job.

by joshg

All of the jobs submitted were submitted by the Admin user, the admin users password got changed at some point and the jobs wouldn't run.

This discussion is closed to new replies.