Wednesday 7 September 2011

DBS CRASHED OR SESSIONS RESET. RECOVERY IN PROGRESS or WSA E HOST UNREACH

This seems to be a fairly common stumbling block for Teradata users. I understand that most (if not all) applications connecting to Teradata interface using TPA or Trusted Parallel Application. I do not know if there any desktop applications which don't, so here goes a summary of the problem and a probable solution.

You are able to ping the Teradata node. When you try to login using BTEQ or SQL Assistant, you get an error message. For BTEQ the error message is this:

Warning: DBS CRASHED OR SESSIONS RESET. RECOVERY IN PROGRESS

For SQL Assistant, its WSA E HOST UNREACH (see the earlier blog for screenshot)

The simple solution might be start the TPA daemon on the server.

/etc/init.d/tpa start

Here's the full text:

(none):~ # /etc/init.d/tpa stop
PDE stopped for TPA shutdown
(none):~ # netstat -an | grep 1025
(none):~ # bteq

 Teradata BTEQ 12.00.00.00 for LINUX.
 Copyright 1984-2007, NCR Corporation. ALL RIGHTS RESERVED.
 Enter your logon or BTEQ command:
.logon suse/partha

.logon suse/partha
Password:

 *** Warning: DBS CRASHED OR SESSIONS RESET.  RECOVERY IN PROGRESS>
 *** Warning: Exiting because of three BREAKs!

 *** Exiting BTEQ...
 *** RC (return code) = 2
(none):~ # netstat -an | grep 1025
(none):~ # /etc/init.d/tpa start
Teradata Database Initiator service is starting...
Teradata Database Initiator service started successfully.
(none):~ # netstat -an | grep 1025
(none):~ # bteq

 Teradata BTEQ 12.00.00.00 for LINUX.
 Copyright 1984-2007, NCR Corporation. ALL RIGHTS RESERVED.
 Enter your logon or BTEQ command:
.logon suse/partha

.logon suse/partha
Password:


 *** Logon successfully completed.
 *** Teradata Database Release is 12.00.03.12
 *** Teradata Database Version is 12.00.03.14
 *** Transaction Semantics are BTET.
 *** Character Set Name is 'ASCII'.

 *** Total elapsed time was 9 seconds.

 BTEQ -- Enter your DBC/SQL request or BTEQ command:
select current_date;

select current_date;

 *** Query completed. One row found. One column returned.
 *** Total elapsed time was 2 seconds.

    Date
--------
11/09/07

 BTEQ -- Enter your DBC/SQL request or BTEQ command:
.quit

.quit
 *** You are now logged off from the DBC.
 *** Exiting BTEQ...
 *** RC (return code) = 0
(none):~ # netstat -an | grep 1025
tcp        0      0 0.0.0.0:1025            0.0.0.0:*               LISTEN
tcp        0      0 127.0.0.1:42528         127.0.0.1:1025          TIME_WAIT
tcp        0      0 127.0.0.1:42527         127.0.0.1:1025          TIME_WAIT
(none):~ # netstat -an | grep 1025
tcp        0      0 0.0.0.0:1025            0.0.0.0:*               LISTEN
(none):~ #

and here's the screenshot:

6 comments:

Ritesh said...
This comment has been removed by the author.
Ritesh said...

Hi Partha,

Thanks a lot for writing this page. I am totally new to Teradata and I was able to solve the RDBMS crash problem on Linux using your post.

However when I enter last command as netstat -an | grep 1025 I get something different than what you have showed on your screen shot.

tcp 0 0 :::1025 :::* LISTEN

Anonymous said...

valuable post.
Become an PGD in business analytics.
online aptitude training
learn core java online
MBA in marketing management
MBA in event management
MBA in Business analytics


siva said...

nice explaaanation Teradata dba Online Training

Afzal said...

Hi Partha,

It worked for me, but I want to know through this command what exactly teradata do at backend ?

Thank you,
Afzal

Afzal said...

It worked for me on vmware terminal but when i am trying to connect bteq in my window through CMD or in bteq utility it is giving me same error.