Login Scripts Appear different across WAN

[zowie2]

Help!

Our Network consists of about 20 NW 5.0/5.1 servers across a WAN. Recently two of our techs in different offices were discussing the container login script for a particular location. When they pulled up the script in NWAdmin, it was not exactly the same on their screens!! They checked and double checked to make sure it was the same container, but thte script appeared differently. Needless to say, we're a little freaked here. The scripts had not been changed in several months. The only major change was a loss of an IPX link between three of nine officee about three weeks ago. We were running TCP/IP across our VPN for server communication long before that link disappeared.

 

[DrGreen26]

You may want to choose the server (NDS Replica) that you know to be correct and do a send all objects, which will update the remote NDS replicas. GOod catch, and good thing your admins login to different servers. I must tell you that if you all logged into the same server each time, you would of never of found this problem. Also, run dsrepair after you do a send all updates. When you run it, run it on every server in the tree wether it has a replica or not. That will the local Database information on each server will be checked for errors.

You may also want to login to other servers in your tree before sending out all the updates just to make sure that no other replica in the tress is showing the wrong information: ie login scripts, object counts per continer etc. Just to be on the safe side.

HTH


Mark C. Greenwood, CNE
m_jgreenwood@yahoo.com

CNE 4.11 and CNE 5 certified. BS Degree in MIS. Working in the industry for 8 years.

I work with NT servers, NDS for NT as well.

 

[TheLad]

Definately sounds like a lack of NDS replication to me..... -----------------------------------------------------
"It's true, its damn true!"
-----------------------------------------------------

 

[salserocito]

Is TIMESYNC syncronized across the Servers???

Good Luck!

 

[zowie2]

Thank you for all your responses. We checked timesync and performed dsrepairs across all servers but did not find any unusual errors that affected the problem. We had text-file backups of all of our login scripts and applied a cut-and-paste to the problems container. Upon saving, the login appeared consistent in all of locations. Too weird to figure out... our tree is in pretty good health otherwise.
Any of you ever have problems syncing when some of your office-to-office links change from IPX to IP?