[Halld-offline] bad f250 pulse data

[Naomi Jarvis]

I think David was called or messaged about it at some point during the
following shift (owl).  He made this note
https://logbooks.jlab.org/entry/3608646

The problem was fixed when Sasha suggested rebooting the ROC at the start
of the following day shift. https://logbooks.jlab.org/entry/3608681

I don't know for sure that all the data are garbage, at present they are
certainly unreadable as hd_root crashes very near the start of the 000
file.  In some of the other files it keeps running for a bit longer.  I
tried rigging my own evio-reader to skip the error events  (I was only
interested in CDC gain calibrations so FCAL data weren't too important to
me, I just need the triggers and tracking) but was not entirely successful
:-( and instead I found another run to use for calibrations instead.  Not
every event contains the bad data, there were some fairly long gaps between
them.

Naomi.


On Tue, Nov 13, 2018 at 10:15 AM Shepherd, Matthew <mashephe at indiana.edu>
wrote:

>
> I was on shift when runs 50920 and 21 were taken.  I am certain we were
> looking at rootspy plots to verify data quality.  However, I do remember
> the rootspy processes dying frequently and having to restart them using the
> GUI.    It is a little unfortunate the data is unrecoverable.  :(
>
> There were some DAQ issues. We noticed erratic live time with no
> explanation for the reason.  The relevant shift summary entry is here:
>
> https://logbooks.jlab.org/entry/3608369
>
> Matt
>
> > On Nov 13, 2018, at 9:48 AM, Naomi Jarvis <nsj at cmu.edu> wrote:
> >
> >
> > Just FYI this affects runs 50920 to 50926.  The runs either side of
> those look ok.
> >
> > On Fri, Nov 9, 2018 at 2:27 PM Naomi Jarvis <nsj at cmu.edu> wrote:
> >
> > PlotBrowser incoming data is completely blank (not even an empty
> histogram) for 50924 and a few other runs.  That could be a clue.
> >
> > On Fri, Nov 9, 2018 at 2:10 PM Sean Dobbs <sdobbs at fsu.edu> wrote:
> > Hi Naomi,
> >
> > It does indeed looks like there's possibly some corrupted data in
> there.
> > We don't have any automatic monitoring that checks for something like
> this at the moment.
> >
> > Usually this is noted as we go through the runs and look at
> monitoring/etc. results.
> >
> > Cheers,
> > Sean
> >
> > On Fri, Nov 9, 2018 at 1:53 PM Mark Ito <marki at jlab.org> wrote:
> > Naomi,
> >
> > I opened an issue on this on GitHub, for the halld_recon repository, so
> it does not get lost.
> >
> >   -- Mark
> >
> > On 11/09/2018 01:45 PM, Naomi Jarvis wrote:
> >> Hi,
> >>
> >> I have run into a problem with the evio files in run 50924 - hd_root
> finds a bad f250 pulse and then throws an JException and crashes.  Files
> 000 and 001 crash instantly, file 002 proceeds for 2.1k events and then
> crashes.  I'm using halld_recon version 3.2.0.
> >>
> >> Does anyone know which runs do and don't have these errors in them?
> >>
> >> Was this already found by the automatic monitoring software?   (if not,
> why not?)
> >>
> >> Naomi.
> >>
> >>
> >> _______________________________________________
> >> Halld-offline mailing list
> >>
> >> Halld-offline at jlab.org
> >> https://mailman.jlab.org/mailman/listinfo/halld-offline
> >
> > --
> > Mark Ito,
> > marki at jlab.org, (757)269-5295
> > _______________________________________________
> > Halld-offline mailing list
> > Halld-offline at jlab.org
> > https://mailman.jlab.org/mailman/listinfo/halld-offline
> > _______________________________________________
> > Halld-offline mailing list
> > Halld-offline at jlab.org
> > https://mailman.jlab.org/mailman/listinfo/halld-offline
> > _______________________________________________
> > Halld-offline mailing list
> > Halld-offline at jlab.org
> > https://mailman.jlab.org/mailman/listinfo/halld-offline
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20181113/b361d466/attachment-0002.html>