Consolidate RecoMon
There are a few things that come to mind about RecoMon this year. I am putting the issue here, although corresponding modifications may affect also Rec and Moore projects.
- there is an outstanding MR on some new monitoring tasks that we used during PbPb and need to be included into the general monitoring Moore!4027 (merged) Rec!4124 (merged)
- There is a crash recently with getting the conditions in RecoPhysMon only (RecoMon works), this needs urgent fixing Moore#907 -> now moved to #96, fixed by !579 (merged)
-
we may want to have correlations between nPVs and n(VELO/Long)Tracks and detector occupancy, possibly even distance between PVs correlated to nTracks-> this is a nice study, but no monitoring value - 28.03.2025 update: Rec!4309 (merged) [@pkopciew working on it] @collinsp was asking to duplicate the detector correlation plots - making a coarse histogram with a large range (mainly for PbPb, but also for monitoring splash events, which appear as well in pp) and a fine one with a smaller range (mainly for pp)
-
[Titus] RecoMon is slow, which especially harms in PbPb (also in context of the correlation plots, but not only), in particular the Detector Correlation Monitors may be outsourced to a different task that has less overhead (for detector correlation plots no reconstruction is needed, only decoding)
- !574 (merged)
- need to setup ecs script - run from RecoMon stack?
- include in run control
- some of the histograms may be too large Rec#597
- We should possibly run RecoSMOGMon during nominal pp in 2025 as well
- in terms of slowness, removing the upc from the physics RoutingBit in pbpb was in the end not improving things.
-
run RecoMon from released stack
/group/hlt/monitoring/recomon{A,B}- need to update to 2025 conditions !575 (merged)
- need to update the corresponding stack
- update page documentation
- update past references
- Explore adding custom features (JS embeddings) for Monet bokeh plots, such as colorbar (Z-axis range) slider
- Use StaticHistograms and StaticProfileHistograms throughout Rec#602 (negligible impact expected, so low priority)