I am working on an analysis of the 200 GeV pAu Run 15 data. In general run QA , I have made a TProfile of the average number of vertices reconstructed in each run. There is a marked jump from an average of ~6.2 to ~7.2 between runs on day 133 (i.e. 16133### on May 5, 2015) and the following day, and then back again to ~6.2 during day 149 (runs #'s 16148###) on May 28th. There is not a corresponding shift in the ZDC rate.

Attached is a PDF with the TProfiles of both he n-Vertices and ZDC rates as a function of run number.

The shift logs appear to note the change in the following links.

https://www.star.bnl.gov/devcgi/qa/QAShiftReport/showRun.php?reptype=FastOffline&repnum=1748

https://www.star.bnl.gov/devcgi/qa/QAShiftReport/showRun.php?reptype=FastOffline&repnum=1765

Given that my analysis only uses the single "best" vertex, it is not obvious that this will impact my analysis at all. However, it does affect about half the date. I'm not sure of what this is a symptom, and if that unknown will prove important. Thanks for the input!

Update (3/5/2017):

The following TProfiles have been requested:
1. number of global tracks, 
2. number of primary tracks,
3. number of tof hits/ matched tracks,
4. global track dE/dx
5.  (number of tracks)/(primary vertex)
I ran through the data again, and got 1,2, and 4, but 2 and 5 are missing. I will debug and update again when I get it to work. I did see, however, that there is a discontinuity into the phi distribution of the tracks, so it appears likely that it was an effect from a single sector.