Fix crash in HcalDigisClient (again)

[mark-grimes]

This bug shows up as a seg fault in HcalDigisClient during harvesting:

#5  0x00002b7c734c1a74 in MonitorElement::Fill(double, double) () from /afs/cern.ch/cms/sw/ReleaseCandidates/vol0/slc5_amd64_gcc472/cms/cmssw/CMSSW_6_2_X_SLHC_2014-03-18-0200/lib/slc5_amd64_gcc472/libDQMServicesCore.so
#6  0x00002b7c831fbcb4 in HcalDigisClient::HcalDigisEndjob(std::vector<MonitorElement*, std::allocator<MonitorElement*> > const&, std::string) () from /afs/cern.ch/cms/sw/ReleaseCandidates/vol0/slc5_amd64_gcc472/cms/cmssw/CMSSW_6_2_X_SLHC_2014-03-18-0200/lib/slc5_amd64_gcc472/pluginValidationHcalDigis.so
#7  0x00002b7c831fc8d8 in HcalDigisClient::runClient() () from /afs/cern.ch/cms/sw/ReleaseCandidates/vol0/slc5_amd64_gcc472/cms/cmssw/CMSSW_6_2_X_SLHC_2014-03-18-0200/lib/slc5_amd64_gcc472/pluginValidationHcalDigis.so
#8  0x00002b7c64b7d330 in edm::EDAnalyzer::doEndRun(edm::RunPrincipal const&, edm::EventSetup const&, edm::CurrentProcessingContext const*) () from /afs/cern.ch/cms/sw/ReleaseCandidates/vol0/slc5_amd64_gcc472/cms/cmssw/CMSSW_6_2_X_SLHC_2014-03-18-0200/lib/slc5_amd64_gcc472/libFWCoreFramework.so

Histograms are only booked if an un-intialised boolean is true. The histograms are filled if the boolean is true, but this is after the variable has been properly initialised. Hence there is a 50/50 chance that the code will either run as expected or seg fault, depending on the previous memory contents of the variable.
This was previously fixed in #2483 but was then broken again by #2833.

[cmsbuild]

A new Pull Request was created by @mark-grimes (Mark Grimes) for CMSSW_6_2_X_SLHC.

Fix crash in HcalDigisClient (again)

It involves the following packages:

Validation/HcalDigis

@ojeda, @danduggan, @rovere, @cmsbuild, @nclopezo, @deguio, @Degano can you please review it and eventually sign? Thanks.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
@andersonjacob, @mark-grimes you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[mark-grimes]

merge

Tests 10000, 10200, 10400, 11200, 11400, 11600, 11800, 12000, 12800 and 13000 pass all steps.
Tests 12200, 12400, 12600, 13200 and 13400 fail in step 2 with known errors.