disks filled, stopping DAQ

NEXT
Make New Entry, Make Followup Entry

User name R. Michaels

Log entry time 07:43:41 on October31,2008

Entry number 245712

Followups:

keyword=disks filled, stopping DAQ

The adaql2 disks filled to ~100% (and I got paged by my scripts,
so at least that works).

Two reasons:

#1. Someone apparently created CODA configs which do not filesplit
thus we have some HUGE files, like one that is 42 Gbytes !!
The huge files don't fit into MSS and hang around forever
because they can't be deleted. (I need to get better at
sensing this problem.)

(Just now I fixed 3 CODA configs including RICHDMA which had no
SPLITMB parameter set.)

#2. Presently with TWOARM_FB the event size is 5.6 kBytes. The
breakdown by crate is shown below. One may consider looking
into ROC9 and ROC10, what is making all the hits there ?

TS11 and TS12 = 0.08 kBytes
ROC3 = 0.35 kB
ROC4 = 0.33 kB
ROC5 = 0.69 kB
ROC8 = 0.55 kB
ROC9 = 2.8 kB !!! <<<<<
ROC10 = 1.3 kB ! <<<<

I will move some files around (my policy is never to delete).
I will need to split those enormous files up by hand.
Please take care about SPLITMB.