Problem log into DataStage Server

Archive of postings to DataStageUsers@Oliver.com. This forum intended only as a reference and cannot be posted to.

Moderators: chulett, rschirm

Locked
admin
Posts: 8720
Joined: Sun Jan 12, 2003 11:26 pm

Problem log into DataStage Server

Post by admin »

This is a topic for an orphaned message.
admin
Posts: 8720
Joined: Sun Jan 12, 2003 11:26 pm

Post by admin »

Without having seen the actual environment my guess would be that the uvrpcd daemon failed to start. Did they see this on the ps report? Re-starting DataStage server will have issued the command to start this daemon, which is the process that handles connections from DataStage clients. This is why I always advise checking that uvrpcd is running after starting DataStage server.

Another possibility is that connections had not been shut down properly and a TCP/IP socket could not be bound. However, re-starting DataStage server is not likely alone to cure this condition.

It would have made diagnosis easier if you/they had reported the error number that came with the error message, as well as the exact error message. For example:
14551 network is unreachable
39121 licence has expired
39122 client and server running different versions
39123 comms software wrong version
81011 unknown host (host name misspelled or could not be resolved)
81012 uvrpc failed to fork service correctly
81013 cant find or open uvrpcservices file
81014 cant find service executable for (uvapi_server)
81016 connection refused - uvrpcd not running
81017 failed to initialize network interface (socket)

Finally note that, with effect from release 3.6 of DataStage, the names uvrpcd and uvrpcservices in the above are replaced with unirpcd and unirpcservices. This primarily affects the grep filter you use on ps to see whether the daemon is running.

-----Original Message-----
From: Melvin Thong [mailto:melvin@adtel.esb.com.my]
Sent: Tuesday, 15 May 2001 16:12
To: informix-datastage@oliver.com
Cc: Ray Wurlod
Subject: Problem log into DataStage Server


Hi,

About 10 days ago, there was a power failure in one of my customers Data Centre. The machine that has DataStage server installed and which is also a production server was brought down abruptly due to the power failure. For your information, the DataStage server was running on a Sun Solaris 2.6. The DataStage release is a version 3.5.

When the power was restored, the machine was brought up and DataStage was running fine again. However, as for today there is a problem facing by the Customers DataStage operators. They could not log into the DataStage through the DataStage client. The error message was something like "Failed to connect to host".

The operators told me that they did not try to do anything silly or mess around with the DataStage since it is a production server. I have asked them to check on the permission and seems no problem. I also suggested to them to issue a "ps -ef | grep uv" command to check on the DataStage server whether it is still running and it seems everything is still up and running. Finally, I advised them to bring down the DataStage server and re-start the server again. After the DataStage server has been re-started, they managed to log in again to the DataStage server via the DataStage client.

To all the DataStage Gurus out there, what is the possible cause that led to the above problem? My customer is demanding for an explanation that led to the problem. I have no idea what to explain to them about this problem since I am still considered a newcomer to DataStage world.

My concern here is that, whether DataStage server will take up or lock any resources which may lead to the mentioned problem? Please advise me on this matter and would appreciate your help.

Thanks.

Regards,
Melvin
Locked