Dataset corruption, SIGSEGV while reading

niremy · Post by **niremy** » Thu Feb 24, 2011 5:53 am

chulett wrote:Have you involved your official support provider yet?

Not yet,
We are in a phase that prevents us from reinstalling the DataStage project or the server, as they will probably ask us

(or I may be a bit dishonest towards the offical support and its capability to analyse the problem)

PaulVL · Post by **PaulVL** » Thu Feb 24, 2011 6:04 am

Well, try looking at your column definitions and make sure your NLS mappings are the same between server and enterprise for that job.

I've seen seg faults due to that type of mismatch.

sarjushah · Post by **sarjushah** » Thu Feb 24, 2011 4:07 pm

I was receiving the exact same error today in my job. Check if you handled all your nulls.

Once I took care of some nulls the error got away.

Best of luck

Sarju Shah

niremy · Post by **niremy** » Fri Feb 25, 2011 11:29 am

Thanks PaulVL and sarjushah,

I'll try to have a look on your tips on monday !

clarcombe · Post by **clarcombe** » Tue Mar 01, 2011 5:15 am

niremy wrote:Thanks PaulVL and sarjushah,

I'll try to have a look on your tips on monday !

If I were you I would get the original programmers bqck to fix it. Sounds like they are a bunch of cowboys

niremy · Post by **niremy** » Wed Mar 16, 2011 4:37 am

clarcombe wrote:If I were you I would get the original programmers bqck to fix it. Sounds like they are a bunch of cowboys

We don't need cowboys because we don't have cows

I'm glad to announce that we found the problem : APT_AUTO_TRANSPORT_BLOCK_SIZE was set to true.

When we put it to it's default value (false) the SIGSEGV problem was gone.

Thanks for your help and support, problem's resolved

ajay.vaidyanathan · Post by **ajay.vaidyanathan** » Wed Mar 16, 2011 12:26 pm

Hi,
Delete your DataSet from the location using DataSet Management and then try running your job again from the beginning.

This should solve your problem. If this resolves your problem i will let you know what was the problem.