EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  <20232024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  <20232024 
<== Date ==> <== Thread ==>

Subject: Re: Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True)
From: "Marco A. Barra Montevechi Filho via Tech-talk" <tech-talk at aps.anl.gov>
To: Torsten Bögershausen <torsten.bogershausen at ess.eu>, "tech-talk at aps.anl.gov" <tech-talk at aps.anl.gov>
Cc: SWC <swc at lnls.br>
Date: Tue, 17 Jan 2023 16:52:05 +0000
Hi, Torsten, thanks for the answer.

> For your case, writing to CNEN and wait for the controller to have the amplifier switched on: Is this what you want ?

Kinda. I just expect the function to return in one of two cases:
  • It sees that CNEN value has been set in the database. Not necessarily needs confirmation from driver.
  • Timeout occurs.
But what im getting seems to indicate that some other thing is happening:
Investigating the IOC code, the only time when the "SVO" command is used is in:

#############################################
asynStatus PIGCSController::setServo(PIasynAxis* pAxis, int servoState)
{
    char cmd[100];
    std::cout << "TESTE" << std::endl;
    sprintf(cmd, "SVO %s %d", pAxis->m_szAxisName, servoState);
    asynStatus status = m_pInterface->sendOnly(cmd);
    if (status != asynSuccess)
    {
        return status;
    }
    int err = getGCSError();
    if (COM_NO_ERROR == err)
    {
        pAxis->m_bServoControl = (servoState == 1);
        if (pAxis->m_bProblem && pAxis->m_bServoControl)
        {
            pAxis->m_bProblem = false;
        }
        return asynSuccess;
    }
    asynPrint(m_pInterface->m_pCurrentLogSink, ASYN_TRACE_ERROR|ASYN_TRACE_FLOW,
              "Could not set servo state!\n");
    return asynError;

}

#############################################
which seems to mean that CNEN field is not being polled. A way i got to confirm this was:

start IOC. Check CNEN value. CNEN value is 1. 
Change CNEN value to 0. Check CNEN value. CNEN value is 0.
Restart IOC. Check CNEN value. CNEN value is still 1.

If CNEN is not being polled, then there is no agent modifying its value, not even a driver response.
If there is no driver response changing the value at database and i just did:

Motor.put("CNEN", 0,wait=True) and my function returned in much less then the default timeout, why is CNEN value!=0 when i check it immediately after?

Best regards,

Marco


From: Torsten Bögershausen <torsten.bogershausen at ess.eu>
Sent: 17 January 2023 04:22
To: Marco A. Barra Montevechi Filho <marco.filho at lnls.br>; tech-talk at aps.anl.gov <tech-talk at aps.anl.gov>
Cc: SWC <swc at lnls.br>
Subject: Re: Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True)
 
Hej all.

I think that this behaviour is - what it is.
In theory, the motorRecord could behave as you (and me) whish,
in proactise there are too many applications depending on this frequent
callbacks.

So this commit:

<https://github.com/epics-modules/motor/commit/c970afbfe71521d1587c51b0c7f3a634b03f5833>

Had to be reverted later:

<https://github.com/epics-modules/motor/commit/0ef39053aac72d8718ef99d08164da1be0825370>


For your case, writing to CNEN and wait for the controller to have
the amplifier switched on: Is this what you want ?

On our systems this power-on can take 0.6 seconds (typically)
up to 6 seconds (on a heavy drive, doing a self test after coldstart).
And if the safety relay is off after a power-fail,
it will never turn on.

So yes, in principle we could change the motorRecord to do what we want:
Wait for the driver to report "amplifier on".
In practise we would need a time out.
And a way to configure it.

I am not sure if it ith worth the effort (to change the motorRecord)
or if a simple poller with a timout in python is easier to implement.

HTH
BR
/Torsten





On 1/16/23 22:26, Marco A. Barra Montevechi Filho via Tech-talk wrote:
> Good evening all.
>
> Im working with the same IOC i was using here:
> https://epics.anl.gov/tech-talk/2023/msg00020.php
> <https://epics.anl.gov/tech-talk/2023/msg00020.php>. Now i loaded some
> motor records in it and am trying to control the IOC via pyepics.
> Something weird is happening.
>
> i made the following python script:
>
> import epics
> a = epics.Motor("MGN:B:E873A:DIAG:X:m1")
> a.put("CNEN", 1,wait=True)
> assert a.get("CNEN")==1
>
> And i frequently get assertion errors. What is happening? Shouldn't
> wait=True prevent this? How do i debug this?
> Even more weird: i tried debugging this with callbacks:
>
> import epics, time
>
> a = epics.Motor(
> "MGN:B:E873A:DIAG:X:m1")
>
> defprintvalue(**kwargs):
> print(kwargs["value"])
>
> b = epics.PV("MGN:B:E873A:DIAG:X:m1.CNEN")
> b.add_callback(printvalue)
>
> c = 1
>
> whileTrue:
>      c=1-c
> print("#####")
>      a.put("CNEN",c)
>      time.sleep(0.1)
> And i get prints like:
>
> #####
> 1
> #####
> 0
> #####
> 1
> 0
> #####
>
> Why do i sometimes get two callbacks and sometimes get only one? Thanks
> in advance for any help.
>
> I append the IOC initialization log and the substitutions file it loads:
>
> *INIT LOG:*
>
> #!/usr/local/epics/apps/pigcs2IOC/bin/linux-x86_64/pigcs2
> < /usr/local/epics/apps/pigcs2IOC/iocBoot/iocPIGCS2/envPaths
> epicsEnvSet("IOC","iocPIGCS2")
> epicsEnvSet("TOP","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/modules/motorPIGCS2-R1-1/iocs/pigcs2IOC")
> epicsEnvSet("SUPPORT","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support")
> epicsEnvSet("ASYN","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/asyn-R4-36")
> epicsEnvSet("STREAM","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/StreamDevice-2-8-9")
> epicsEnvSet("CALC","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/calc-R3-7-3")
> epicsEnvSet("RECCASTER","/usr/local/epics-nfs/apps/recsync/1.4_epics_3.15/client")
> epicsEnvSet("SNCSEQ","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/seq-2-2-6")
> epicsEnvSet("BUSY","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/busy-R1-7-2")
> epicsEnvSet("IPAC","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/ipac-2-15")
> epicsEnvSet("MOTOR","/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1")
> epicsEnvSet("EPICS_BASE","/usr/local/epics-nfs/base/R3.15.6")
> cd
> "/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/modules/motorPIGCS2-R1-1/iocs/pigcs2IOC"
> ## Set up environment
> epicsEnvSet("DEV","MGN:B:E873A:DIAG:X:")
> epicsEnvSet("BL", "MGN:B:E873A:DIAG:X")
> epicsEnvSet("IOCNAME", "MGN-B-E873A-DIAG-X")
> epicsEnvSet("EPICS_CA_MAX_ARRAY_BYTES",15728645)
> # protocol file paths
> epicsEnvSet ("STREAM_PROTOCOL_PATH",
> ".:/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/db")
> ## Register all support components
> dbLoadDatabase "dbd/pigcs2.dbd"
> pigcs2_registerRecordDeviceDriver pdbbase
> cd
> "/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/modules/motorPIGCS2-R1-1/iocs/pigcs2IOC/iocBoot/iocPIGCS2"
> ## motorUtil (allstop & alldone)
> dbLoadRecords("/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/db/motorUtil.db", "P=MGN:B:E873A:DIAG:X:")
> ##
> # < PI_GCS2.cmd
> # PI GCS2 support
> dbLoadTemplate("/usr/local/epics/apps/config/PIGCS2/e873aX.substitutions")
> drvAsynIPPortConfigure("E873_ETH","CDE5B3D.abtlus.org.br:50000",0,0,0)
> # Turn on asyn trace
> # asynSetTraceMask("E873_ETH",0,3)
> # asynSetTraceIOMask("E873_ETH",0,1)
> # PI_GCS2_CreateController(portName, asynPort, numAxes, priority,
> stackSize, movingPollingRate, idlePollingRate)
> # # this is the final setup
> # PI_GCS2_CreateController("E873", "E873_ETH", 1, 0, 0, 100, 1000)
> # speedup debug
> PI_GCS2_CreateController("E873", "E873_ETH", 1, 0, 0, 50, 1000)
> 2023/01/16 18:23:03.949 read from E873_ETH: (c)2016-2020 Physik
> Instrumente (PI) GmbH & Co. KG, E-873.1AT, 120040762, 03.033
> PIasynAxis::PIasynAxis() 0: 1
> ----------------- axis num: 0 createCLParams() m_szAxisName:
> 1-------------------
> # Turn off asyn trace
> # asynSetTraceMask("E873_ETH",0,1)
> # asynSetTraceIOMask("E873_ETH",0,0)
> # Set end-of-string terminators (port, addr, terminator)
> asynOctetSetInputEos("E873_ETH", 0, "\n")
> asynOctetSetOutputEos("E873_ETH", 0, "\n")
> dbLoadRecords("/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/motor-R7-1/modules/motorPIGCS2-R1-1/iocs/pigcs2IOC/db/PI_Support.db","P=MGN:B:E873A:DIAG:X:,R=m1:,PORT=E873,ADDR=0,TIMEOUT=1")
> # asyn record for troubleshooting
> dbLoadRecords("/usr/local/epics-nfs/modules/R3.15.6/synApps/R6.1/support/asyn-R4-36/db/asynRecord.db","P=MGN:B:E873A:DIAG:X:,R=asynEth,PORT=E873_ETH,ADDR=0,OMAX=256,IMAX=524288")
> # RECSYNC
> dbLoadRecords("/usr/local/epics-nfs/apps/recsync/1.4_epics_3.15/client/db/reccaster.db", "P=MGN:B:E873A:DIAG:X:REC:")
> # protocol
> #dbLoadRecords("$(MOTOR)/db/e873.db", "P=$(DEV),PORT=E873_ETH")
> iocInit
> Starting iocInit
> ############################################################################
> ## EPICS R3.15.6
> ## EPICS Base built Sep 27 2021
> ############################################################################
>
>
> *e873aX.substitutions FILE:*
>
> file "$(MOTOR)/motorApp/Db/asyn_motor.db"
> {
> pattern
> {P,              M,     DTYP,       PORT, ADDR,      DESC,  EGU,  DIR,
>   VELO,  VBAS,  ACCL,  BDST,  BVEL,  BACC,  MRES,  PREC,  DHLM,  DLLM,
>   INIT}
> {MGN:B:E873A:DIAG:X:,  "m1",  "asynMotor",  "E873",    0,  "Motor 1",
> mm,  Pos,    1,    0,    .1,    0,      0,    0,   0.000001,  7,    
> 5.0,   -5.0,  ""}
> }
>
> Aviso Legal: Esta mensagem e seus anexos podem conter informações
> confidenciais e/ou de uso restrito. Observe atentamente seu conteúdo e
> considere eventual consulta ao remetente antes de copiá-la, divulgá-la
> ou distribuí-la. Se você recebeu esta mensagem por engano, por favor
> avise o remetente e apague-a imediatamente.
>
> Disclaimer: This email and its attachments may contain confidential
> and/or privileged information. Observe its content carefully and
> consider possible querying to the sender before copying, disclosing or
> distributing it. If you have received this email by mistake, please
> notify the sender and delete it immediately.
>

References:
Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True) Marco A. Barra Montevechi Filho via Tech-talk
Re: Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True) Torsten Bögershausen via Tech-talk

Navigate by Date:
Prev: Re: Weird behaviour in wait=True when using epics.Motor.get(something, something, wait=True) Matt Newville via Tech-talk
Next: Re: Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True) Marco A. Barra Montevechi Filho via Tech-talk
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  <20232024 
Navigate by Thread:
Prev: Re: Weird behaviour in wait=True when using epics.Motor.get(something,something,wait=True) Torsten Bögershausen via Tech-talk
Next: Re: Weird behaviour in wait=True when using epics.Motor.get(something, something, wait=True) Matt Newville via Tech-talk
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  <20232024 
ANJ, 17 Jan 2023 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·