Raw banks from multiple events included in a single event
When sprucing Turbo data (lhcb-dpa/prod-requests#37 (closed)) an event was found which is corruption and results in the MDF reading getting stuck in an infinite loop (I'll open an issue for LHCb soon).
It looks similar to the issue that was seen in the Lead23 data (#697) where a single event contains multiple ODIN banks with different event numbers. I think the only reasonable place this corruption could originate is in HLT2.
See here for an annotated hex dump of the problematic event, including the contents of the decoded ODIN banks:
https://cburr.web.cern.ch/Moore-697/289233_00120016_0000.evt-404335_event-1.html
Here is an MDF file contianing only the bad event:
Edited by Chris Burr