From: Jeff Landgraf (jml_at_bnl.gov)
Date: Fri Mar 14 2003 - 10:03:08 EST
Problem was after some number of events TOF TCD goes 100% busy. Can be
fixed by stoping run and restarting but reoccurs after ~<5000 events.
This happened perhaps 5 times, rebooting trigger, trying l2 writes off
etc, before I had them remove it.
-jeff
On Thu, 13 Mar 2003, Frank Geurts wrote:
> Dear all,
> i am puzzled -again- ... checked all systems (pVPD-TOFp-TOFr HV,
> TOFp-TOFr NPS, TOFr gas all of it) but nothing shows up bad or suspect.
> In fact, everything looks very stable. The det.log file doesn't show any
> problems and the shiftcrew repeatedly affirms that TOF never went RED on
> the RTS display. The only 'suspect' was a rogue xterm process on tofpdaq
> that had grown to ~20Mb in size but didn't seem to take any CPU
> resources (all tail's and xterms have been killed now).
>
> Maybe Jeff can comment on why exactly he thought TOF had to be removed
> from runcontrol ... apparently it did seem to "solve" some problem down
> stream. (I cc'd Jeff and Tonko)
>
> I first suspected the UPC trigger to dramatically lower our local (pVPD)
> trigger efficiency (and maybe revealing an unknow issue there) but Tapan
> told me that UPCcombined has a significant dAuMinbias contribution in it...
>
> I told Tapan to leave the TOF out for now but I do hope Jeff can follow
> up on this first thing tomorrow morning explaining what is going on.
>
>
> -frank
>
>
>
>
> Frank Geurts wrote:
> > Hi,
> > i monitored the log files ... absolutely nothing indicates a problem
> > with TOF in fact I doubt that TOF went red at all in the RTS display.
> > Below the TOF relevant lines in the det.log ... no CRITICALs or
> > OPERATORs! I'll check the sanity of the other systems and contact SCR
> > later.
> >
> >
> > [tof02 22:16:59] (localmon): WARNING: localmon.cxx [line 171]:
> > localmon socket close, reconnecting ...
> > [tof02 22:17:14] (localmon): WARNING: localmon.cxx [line 221]:
> > connected with local mon server!
> > [tof02 22:48:18] (runcontrol): WARNING: runcontrol.cxx [line 457]:
> > Received a RTS_SEND_CONFIG command!!
> > [tof02 22:48:19] (runcontrol): WARNING: runcontrol.cxx [line 495]:
> > Physics run!
> > [tof02 22:48:31] (runcontrol): WARNING: runcontrol.cxx [line 396]:
> > received RTS_RUN_START, going into RC_RUNNING state!!
> > [tof02 22:48:31] (senddata): WARNING: senddata.cxx [line 413]:
> > received RTS_RUN_START command.
> > [tof02 22:48:31] (getdata): WARNING: getdata.cxx [line 786]: received
> > RTS_RUN_START command
> > [tof02 22:49:19] (runcontrol): WARNING: runcontrol.cxx [line 417]:
> > Received RTS_RUN_STOP, going into RC_READY state!!
> > [tof02 22:49:20] (senddata): WARNING: senddata.cxx [line 729]:
> > Received signal USR1, RTS_RUN_STOP
> > [tof02 22:49:20] (senddata): WARNING: senddata.cxx [line 429]:
> > received RTS_RUN_STOP command
> > [tof02 22:49:20] (getdata): WARNING: getdata.cxx [line 1587]:
> > received RTS_RUN_STOP command
> > [tof02 22:49:21] (runcontrol): WARNING: runcontrol.cxx [line 511]:
> > Received a RTS_FLUSH_TOKENS command!!
> >
> >
> >
> >
> >
> >
> >
> >
> > Tapan Nayak wrote:
> >
> >> Dear Frank and Bill,
> >>
> >> After we got a good store today we started running. But the run
> >> wouldn't start for some reason. Jeff Landgraff diagnosed the problem
> >> to be relared to TOF. We took TOF out of DAQ and set the L2DataWrite
> >> to be
> >> Write_NONE and then we could run normally.
> >>
> >> Just now after talking to you: after run 4072055 we put the TOF back
> >> into the DAQ and left the L2DataWrite to be Write_NONE. But the run
> >> didn't start. We took TOF out. Now we are back to running again.
> >>
> >> So I don't understand the problem: it could be something do with TOF
> >> as L2DataWrite is still Write_NONE. Could you please check that.
> >>
> >> Thanks a lot
> >>
> >>
> >> Tapan Nayak
> >> (11pm, March 13)
> >>
> >
>
>
This archive was generated by hypermail 2.1.4 : Thu Jul 24 2003 - 00:39:37 EDT