Re: Adding additional fields to be archived in "policies.py"

[Timo Korhonen via Tech-talk <tech-talk at aps.anl.gov>]

> when using caget on the commandline. However, when looking into the
> archiver data, .STAT changes from 0 to 1 while .SEVR stays at 0 all the
> time. I would have expected 10 for .STAT and 3 for .SEVR as seen on the
> commandline with caget.

That would have been my expectation as well. The behavior you see does not make sense to me.
Caget/camonitor, as well as pvmonitor behave as I expect.
I have not tried that with AA myself, maybe somebody with more insight could chime in?

But if that does not work, then I guess the only option left would be to add one (or two, if you want separate severity and status) more PVs.

Timo

On 2024-09-09, 12:09, "Gerrit Kühn" <gerrit.kuehn at aei.mpg.de <mailto:gerrit.kuehn at aei.mpg.de>> wrote:


Am Fri, 6 Sep 2024 10:34:52 +0000
schrieb Timo Korhonen <Timo.Korhonen at ess.eu <mailto:Timo.Korhonen at ess.eu>>:


> > Maybe the plugin doesn't support this? In that case would could
> > probably fall back on creating an extra channel per device that
> > reflects the error status. Or would the first idea we had (simply
> > adding PVNAME.STAT explicitely to the archiver) work as well? 


> In the meantime, until the requested feature is available in the plugin,
> you can of course use your first idea. That should work, albeit with the
> caveats that I mentioned earlier (non-atomicity). Adding an extra
> channel (=record) should not be necessary.


We tried this meanwhile and added .STAT and .SEVR excplicitely to the
archiver. However, I'm not quite sure what to make of the result:
When unplugging the device, we get
.STAT TIMEOUT
.STAT 10
.SEVR INVALID
.SEVR 3


when using caget on the commandline. However, when looking into the
archiver data, .STAT changes from 0 to 1 while .SEVR stays at 0 all the
time. I would have expected 10 for .STAT and 3 for .SEVR as seen on the
commandline with caget.




cu
Gerrit