Page 1 of 1

Lkp_1,0: Operator terminated abnormally: received signal SIG

Posted: Fri Oct 30, 2009 7:21 am
by Thanh Dalton
Hello,

a lookup works fine. But as soon as a condition (the range) is added to the lookup stage, got the error at run time:
Lkp_1,0: Operator terminated abnormally: received signal SIGBUS.

Condition:
Key Expression: Range (DSLink2.lower, ">=", DSLink2.upper, "<=", 0)
Key Type: a..z
Column Name: INITIAL_ENTRY_DATE (all I want to show here it's a DATE).

Any help is highly appreciated.

Posted: Fri Oct 30, 2009 7:27 am
by chulett
What version do you have? I vaguely recall issues with the Range Lookup that may require a patch from your official support provider. :?

Posted: Fri Oct 30, 2009 7:37 am
by Thanh Dalton
chulett wrote:What version do you have? I vaguely recall issues with the Range Lookup that may require a patch from your official support provider. :?

chulett,

IBM Infromation Server 8.1.

Thanks a lot.

Posted: Fri Oct 30, 2009 7:45 am
by algfr
Hmm

Don't know much how to help but here are my two cents...

The range function should be done this way (checking one code + 1 date against list of codes and values interval such as dates for instance)

1) On the code : Set "=" and join with the code

On the start/end values, set "a..z" and join you value to check on the main stream to both bounds and check the range box (the zone will become red meaning you must insert an expression).

2) In the range expression use the following

Either :
value_to_check >= start_value AND value to check < end_value

OR

value_to_check > start_value AND value to check <= end_value

This is VERY important as intrervals must not overlap, I already had issues before.

3) When combining a range lookup (start/end) and a regular lookup, you must set "Multiple values returning".

Posted: Fri Oct 30, 2009 8:10 am
by Thanh Dalton
Thanks, algfr. I'll check the job according to your advice after completing this writing.

I forgot to mention that the exact same job works fine in Windows server. But it doesn't work on Unix (Solaris 10).

Posted: Fri Oct 30, 2009 8:29 am
by chulett
All the more reason to check for a patch, IMHO - which the Linux version (I'm finding) has more than its fair share of. :wink:

Posted: Fri Oct 30, 2009 6:25 pm
by ray.wurlod
Can you compile/run jobs with Transformer stages?

Range lookups are rewritten as transform operators (that is, as Transformer stages) and therefore have the same compiler and linker requirements as Transformer stages.

Posted: Fri Oct 30, 2009 8:08 pm
by chulett
Interesting, was not aware of that little tidbit. :wink:

Posted: Fri Oct 30, 2009 8:44 pm
by ray.wurlod
You (everyone) should review the score more often, especially when new functionality creeps in. It can be quite revealing!

Posted: Thu Nov 05, 2009 9:48 am
by Thanh Dalton
ray.wurlod wrote:Can you compile/run jobs with Transformer stages?

Range lookups are rewritten as transform operators (that is, as Transformer stages) and therefore have the same compiler and linker requirements as Transformer stages.
Hi Rayand others,

Thank you for your help.

Sorry for the late reply. I was off for a few days and just came back.

Yes, I can compile and run a Transformer stage. In fact, the same lookup wothout the range (condition) works fine. But as soon as the range is included, the error appears at run time.

Posted: Thu Nov 05, 2009 12:51 pm
by Thanh Dalton
chulett wrote:All the more reason to check for a patch, IMHO - which the Linux version (I'm finding) has more than its fair share of. :wink:

chulett,

Thank you. I couldn't find any patch relating to "parellel range lookup" for V8.1. What I found is for V8.0 at the following link.

http://publib.boulder.ibm.com/infocente ... 01fp2.html.

Maybe this is what you mentioned?

My version is 8.1.0.0. I just found there is a Version 8.1 Critical Patch, which I haven't installed yet. But this patch doesn't mention anything about range lookup or SIGBUS error.

Posted: Thu Nov 05, 2009 1:45 pm
by Kryt0n
chulett wrote:I vaguely recall issues with the Range Lookup that may require a patch from your official support provider. :? ...
I'm with Craig here... in particular, the vague bit...

We had the same problem a few months ago... think the final response from support was that our version of the C++ compiler was unsupported but as we jumped ship on to a new server that worked fine, we never pursued further.