Histograms fail when running many Pythia events

[Frigorifico]

The university has a cluster they use for running simulations, they gave me a code that tells the cluster what to run, to run it x times (how many “jobs”), and in what queue.
At the end I end up with x output files, and they also gave me a code to merge those files, so far so good.

Now the problem is that sometimes when I open the histograms I get:

Error in TFile::ReadBuffer: error reading all requested bytes from file histos.root, got 218 of 300
Error in TFile::Init: histos.root failed to read the file type data.
(class TFile*)0x2ec0670

I’m not sure if the problem is Pythia or if the problem is Root. Sometimes I’ll see in the output that Pythia ran without problems, and sometimes I’ll see that the output stops before the even generation is supposed to end.

Regardless, the data in the histograms should be there even if the Pythia simulation had some problem, shouldn’t it?

[eguiraud]

It would seem that the file histos.root is malformed – presumably because of the problems in the event generation that you mentioned.

Can you open the corrupted histos.root “by hand”, e.g. with TBrowser? What does it show? Does it also print the same errors when you do?

In any case I think your options are either find out why the files are not generated correctly or leave out the malformed files from the merging step.