Recovery runs done inside an error step

[comunian]

Dear Didier,
we are using some remote computers for doing the errors study on TraceWin.
When we do a single step of errors study, or in the meanwhile of an error step, if we get a stop on the remote computer, we lose the runs done.
For example, if we do a single step with 1000 runs, if the remote machine stops to work after 300 runs we lose this partial work done and we need to restart the errors study with again 1000 runs.
Is There any way to recover the partial runs done inside the step?
Sometimes also happens that a remote machine stops to works, and the TraceWin continue to try to restart the jobs on that machine.
Is there any way to exclude the stopped remote machine after, say, 10 tentative of jobs restart?
Best Regards,
Michele Comunian

[Didier]

Dear Michele,

I understand your first request, but clearly it is very complicated to develop as the code is done today and I could not do that in the near future.
For the machines that don't respond anymore, normally they are only queried regularly but with a longer and longer time gap. But in any case, I don't understand what the problem is here. They don't answer, but it's without consequence, isn't it?

Regards,

Didier

[comunian]

Dear Didier,
we get a lot of errors on our remote computer system. As you can see from the snapshot.
We are using the latest version of TraceWin.
May you can understand why ?
Best Regards,
Michele Comunian

cloud_prob.PNG (112.98 KiB) Viewed 2427 times

[comunian]

Dear Didier,
I post here a very simple example to reproduce the remote runs errors.
May be the problem is connected with the adjust command ?
Best Regards,
Michele Comunian

[Didier]

Dear Michele,

I think, it's fixed in the last TraceWin version.

Regards,

Didier

[Didier]

Dear Michele,

I'm coming back to this post to let you know that a new feature has been added that allows statistical error studies to be resumed after a voluntary or involuntary interruption.
I think you might be interested

Regards,

Didier