40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  PSL  Not logged in ELOG logo
Message ID: 760     Entry time: Fri Dec 9 11:49:17 2011
Author: Frank 
Type: Summary 
Category: DAQ 
Subject: daqd problems (which turn out to be not critical) 

Talked to Alex about the daqd and it's strange behavior on our machine.

  1. The first thing is that the daqd did not show up as a process even it was running. According to Alex this happens because we only write about 100 EPICS channels so they cpu load is so small that it does not show up all the time so you can easily miss it. The safest way to find out if it's running is to try to telnet into it on port 8087 and check the status.
  2. We have the following error message in our log file (since a long time ago):
    A call to "assert (pE == &entry)" failed in ../../cds/project/daq/fb2/exServer.h line 346
    EPICS Release EPICS R3.14.10- $R3-14-10-RC2$ $2008/10/10 15:01:51$.
    E-mail this message to the author or to tech-talk@aps.anl.gov
    Calling epicsThreadSuspendSelf()

    This happens because it's trying to start an epics ioc using a wrong version of the epics library. We tried different version but couldn't fix it. Looks like the easiest solution would be to re-compile the daqd.
  3. Error message: failed to rename file; errno 2
    This means it can't rename the current frame file it is writing to which can have several reasons. Most often it does not have the rights to do so and a second option is that if the daqd is already running but invisible and you try to start (another) one there is no exception catching for that case and those two processes create a conflict because they try to create/write to the same file.
  4. If you want to grab data using the dataviewer you get the following message even if data exists
    read(); errno=9
    read(); errno=9
    T0= some time and length of data
    No data output

    Unknown problem. Even the data exists and can be grabbed using different command line tools to directly access the frames the dataviewer does not get data. A second weird behavior is that if you request let's say 60 seconds it request 120 seconds. Should not be like that but that's a dataviewer problem.
ELOG V3.1.3-