Question about reproducing an error

[Mag0007]

Hello all,

this is a general computer question, not unix related. If there is an application,x, and it is constantly crashing. we have another empty server, and load application x onto it. How hard will it be to reproduce this error?

TIA

 

[KenCunningham]

I guess it depends on whether the other machine is configured (hardware/software) exactly the same as the 'bad' one? However, it could be a fault on the bad machine, which will probably not be reproduceable on the new one, without introducing the same fault too.

A bit of a 'how long is a piece of string?' I think!

 

[marrow]

Why do you want to reproduce the error? Surely you want to eliminate it or do you? As Ken says is the "new" system exactly identical (CPU/?Memory/Diskspace) If so and you have incorrect code then it should fail on both systems. Unless of course you have a hardware problem.

 

[Mag0007]

well at the place where I work, we have production and dev. When there is a problem in production, they try to reproduce this in a 'lab' env. I don't see why this would be helpful...thats why I am asking

 

[SamBones]

The purpose of trying to reproduce a production problem in a lab environment is because you can stop everything to try to diagnose the problem in the lab. You can't stop the production environment to do diagnostics.

The lab system does need to be identical. You just need to duplicate enough of the factors that went into creating the problem. But an identical system gives you the best chance of doing that.

Although I find if you have a smaller machine (i.e. less memory, fewer CPUs, etc), that can sometimes make a problem easier to reproduce. Especially if it's a memory starvation issue, or something like that.

 

[Mag0007]

Thanks for everyone's input...