Experimental Physics and
Industrial Control System

Timo Korhonen via Core-talk <core-talk at aps.anl.gov> · Tue, 8 Mar 2022 17:04:00 +0000

Thanks Andrew and Michael for pointing that out.

I missed the severity setting issue, sorry for causing confusion. As a lame excuse, at least I asked before filing a useless bug report in EPICS base.

So the only issue is in the VTypes, then. Or almost only, we need to check our databases that the severities are properly set. I think they are; I just missed that point when I was looking at the issue yesterday. 

Timo

On 2022-03-08, 17:11, "Michael Davidsaver" <mdavidsaver at gmail.com> wrote:

    On 3/8/22 00:39, Timo Korhonen via Core-talk wrote:
    > Hi all,
    > 
    > We noticed recently that if for instance the lower alarm limits are not explicitly defined in the EPICS database, CS-Studio reports all limits to be NaN. See here https://github.com/ControlSystemStudio/cs-studio/issues/2716 <https://github.com/ControlSystemStudio/cs-studio/issues/2716> for a description and a following discussion.
    > 
    > Kay pointed out that even when the limits are explicitly defined as 0, the IOC reports them as NaN. Now this starts to smell like a bug. But before I submit a bug report, I wanted to ask if there is a reason behind this behavior? Maybe it is a feature and not a bug?

    The code in question (for aiRecord) seems reasonably clear.

    Defining a level without a severity (implicitly NO_ALARM) isn't meaningful.

    >         pad->upper_alarm_limit = prec->hhsv ? prec->hihi : epicsNAN;
    >         pad->upper_warning_limit = prec->hsv ? prec->high : epicsNAN;
    >         pad->lower_warning_limit = prec->lsv ? prec->low : epicsNAN;
    >         pad->lower_alarm_limit = prec->llsv ? prec->lolo : epicsNAN;

    https://github.com/epics-base/epics-base/blob/91941af992f6c32ef4f4ef55a2952f581aeee36a/modules/database/src/std/rec/aiRecord.c#L295-L298

    > There is another interesting case, described by Kay in the above issue report, when only one HIGH and not HIHI is defined. How should this case be interpreted? The “Process Database Concepts” documentation ( https://docs.epics-controls.org/en/latest/guides/EPICS_Process_Database_Concepts.html#alarm-conditions-configured-in-the-database <https://docs.epics-controls.org/en/latest/guides/EPICS_Process_Database_Concepts.html#alarm-conditions-configured-in-the-database> ) indicates that the alarm limits express a range, however not very explicitly. But this is not consistent with this case.
    > 
    > Any thoughts?
    > 
    > Timo
    > 

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System