Reason for change
We are adding a number of new monitoring tags (mostly, to monitor the distance to the SAA)would like to run for a few days the RHEL3 build of GlastRelease-v15r47p12gr06 (candidate release for the switchover to RHEL4), before moving on to RHEL4.
Plus, the staging for crumbs is now moved to xrootdwe are now Running the ScienceTools from our own installation on afs to improve robustness.
Test Procedure
We have processed runs in the DEV pipeline with this version of L1Proc.
...
We can easily switch back to the previous version of L1Proc.
CCB Jira
Details
L1Pipeline: L1Pipeline v1r77
- Stage crumbs (but not chunks) on xroot.
- Make more smaller crumbs.
- Fail fast if too many chunks.
- Don't make ft2Txt file.
GPLtools: fileOps3
- Support crumbs staging on xroot.
dataMonitoring/AlarmsCfg: AlarmsCfg-05-19-05
- Upper warning limit for AcdPedPedMeanDeviation_PMTB_TH1 (y_value) changed from 8 to 12. This is temporary change that will be reverted as soon as we have updated the claibration constants (GDQMQ-321).
- Added two more error conditions to the verify alarm file: checking the FSW compression level (GDQMQ-320), checking that periodic evts never have the TEM bug (GDQMQ-319).
- Limits on the digi trending quantity Mean_CompressionLevel moved by an "epsilon" as a workaround to a bug in the alarm handler (which is now fixed).
- num_sigma parameter set to zero for the alarm on digi trending quantity Mean_CompressionLevel (since we do not expect any single value lower than 8, we want to explicitely ignore the error on the average).
- Upper limits on digi trending OutF_Ratio_EvtSize_CompressedEvtSize changed from 3.25, 3.50 to 3.40, 3.60 to accommodate the B2-1-0 version of the flight software (error events are now compressed).
- Lower limits on digi trending Mean_CompressionLevel changed from 7.0, 7.8 to 8.0, 8.0 to accommodate the B2-1-0 version of the flight software. Now all the events are compressed and everything whose compression level is different from 8 should be regarded as unexpected.
dataMonitoring/FastMonCfg: FastMonCfg-02-00-01
- Added a new xml file with the definition of the SAA uploaded on the spacecraft (GDQMQ-311).
dataMonitoring/Common: Common-06-01-01
- Some more improment in the calculation of the badness in pAlarmLimits.py. In the completely degenerate case in which all the four limits coincide the badness is no longer set to ERROR_BADNESS + DELTA_BADNESS, as it used to be. Rather, an additional constant, proportional to the distance between the (best) value and the center of the interval (possibly weighted with the error) is added in such a way the sorting (badness-wise) between the data points is preserved (i.e. the they don't have all the same badness, but rather worts is labeled as worst).
- Bug fix in how the alarm limits are handled. The issue was the case in which all the four limits were identical and the actual value was outside the limits themselves. A similar issue was documented and fixed in GDQMQ-272, but I missed one of the cases in the logic (now the code looks perfectly simmetric).
dataMonitoring/FastMon: FastMon-05-01-00
- Test program added to read a m7 file and make plots of the orbit and dstance to the SAA (GDQMQ-311).
- Some more work for the calculation of the distance to the SAA. Still pretty rough, but first reasonable implementation.
- New small module pTETEUtils.py added to the repository to handle the transformation between the J2000 and the True Equator True Equinox (TETE) systems.
svac/Monitor: Monitor-01-03-03
- Added new object to retrieve quantity FastMon_Spacecraft_Distance_To_SAA (GDQMQ-311).
- Added new object to retrieve quantity FT1EventClass (GDQMQ-308).
v1r78
- Use new style tags for monitoring packages.
- FastMon called with the -s option (SAA config).
- Running the ScienceTools from afs.
- Cleaning up the JO for digi/recon.
GlastRelease: GlastRelease-15r47p12gr06
- There are a few changes in this GR wrt the version in use in L1, GR v15r47p12 (mostly implemented to get ready for RHEL4):
- enums v3r0p1: set TEMBUG flag to 0x100000 so it is distinct from PHASEERROR
- Trigger v6r4p1gr2: update test JO; modify handling of TriggerAlg.mask to use Gaudi's StringProperty to avoid any questions from conversion from int to unsigned int
- TkrDigi v2r7p3gr1: Mods to unit test JO
- RootIo v21r12p0gr05: Try making RootConvert public, but no_auto_imports for OnboardFilterTds; adjust req file to make RootConvert private; Patch to allow use of EvtMax==0 to read in a complete input ROOT file
- OnboardFilter v4r14p0gr1: Update to unit test JO
- LdfConverter v4r3p4gr1: mod to use EvtMax==0 to denote reading in a complete input file, to avoid issues with setting EvtMax > MAXINT
- IExternal/Geant4Runtime v3r801p7: Rebuild of g4 to set G4_NO_VERBOSE and G4_NO_STORE_TRAJECTORY, and now includes the tables with the windows distribution
- IExternal/Geant4 v3r801p6: Increment tag so installer recognizes we have a new version of G4
- HepRepSvc v0r27p2gr1: replace abs with fabs in src/FilterFillter.cxx
- HepRepCorba v2r0p0gr1: Pick up some fixes from the HEAD for a new version of L1proc
- Gleam v7r2p3gr1: Patch JO for TriggerAlg.mask setting via string and remove mask setting within ldf2digi.txt JO, as it is unnecessary due to use of ConfigSvc
- GlastSvc v9r26p1gr1: apply updates for random service to apply seed once per run the default
- CalXtalResponse v0r21p8gr1: removed use of obf, no necessary now that RootIo is hiding use of RootConvert
- CalRecon v6r7p0gr1: Modified src/CalEventEnergyAlg.cxx to remove tools Philippe Bruel reminds us are no longer used
- AnalysisNtuple v2r51p0gr03: Remove obsolete CAL merit variables as requested by Philippe Bruel; Fix init of FT1EventClass; Fix for FT1EventClass init
- There are a few changes in the system tests wrt GR v15r47p12. These changes arose from a change in seed handling in GlastSvc to switch from an event to run seed for MC generation.
dataMonitoring/AlarmsCfg: AlarmsCfg-05-20-02
- reactivate the alarm on the FSW compression leveldataMonitoring/DigiReconCalMeritCfg: DigiReconCalMeritCfg-01-04-03
- Added FastMon trending quantity FastMon_Spacecraft_Distance_To_SAA (GDQMQ-311).
svac/TestReport: TestReport-10-01-00
(note: this version was approved for L1Proc 1.77 but accidentally not implemented)
- error list is now pointing to the evtGemId.
- checking the FSW compression level (GDQMQ-320).
- checking that periodic evts never have the TEM bug (GDQMQ-319).
Complete set of tags for L1Proc 1.75
Code Versions
GlastRelease (sim/recon):
...
GlastRelease-v15r47p12gr06*
ScienceTools (Level 2) : v9r15p3gl2
...
svac/L1Pipeline: L1Pipeline v1r77v1r78*
calibTkrUtil v2r7p3
calibGenTKR v4r5
dataMonitoring/AlarmsCfg: AlarmsCfg-05-1920-0501*
dataMonitoring/FastMonCfg: FastMonCfg-02-00-01*
dataMonitoring/DigiReconCalMeritCfg: DigiReconCalMeritCfg-01-04-03*
dataMonitoring/Common: Common-06-01-01*
dataMonitoring/FastMon: FastMon-05-01-00*
datMonitoring/IGRF: v1r0p1
svac/Monitor: Monitor-01-03-03*
svac/EngineeringModelRoot: v4r4
svac/TestReport: TestReport-10-01-00*
...
evtClassDefs v0r14
GPLtools: fileOps3*