Re: EPICS device disconnects and reconnects

[Kate Feng <feng@bnl.gov>]

Hi Mark,

    Would it be easier to use a file name to specify the records that 
needed to be reset/restore
at reconnection, instead of using 'PINI' ?  The filename could be 
specified via prosilicaConfig().
For example:
prosilicaConfig(("$(PORT)", 192.168.0.20, 50, 
0,"prosilicaConn.rec","P=13PS1:,R=cam1:")

    The advantage is that the file could be modified even after IOC is 
initialized.
The file would be similar to the format of auto_settings.sav file, but 
one could specify the mode of
reconnection at the beginning of the file. The option of mode could be: 
: 0) the specified default value,
1) the value from save/restore, if available, otherwise take the 
default, 2) read from current camera
setting.

For example: prosilicaConn.rec

Mode 0
$(P):$(R):Gain 0
$(P):$(R):Acquire 0
$(P):$(R):FrameType 0
$(P):$(R):ImageMode 2
$(P):$(R):TriggerMode 0

.......

     Just some thoughts.

    Cheers,
    Kate


On 06/18/2013 04:16 PM, Mark Rivers wrote:
> Hi Lewis,
>
> > > What are the intended meanings of the other values of PINI and where is that documented?
> > Hi, Mark.
> >
> > The one place I know of is in the release notes:
> >
> >   http://www.aps.anl.gov/epics/base/R3-14/12-docs/RELEASE_NOTES.html
> >
> > Search for "initHookAtIocRun".
> Thanks, I knew I had seen it somewhere.  The Application Developer's Guide needs to be updated.
>
> It seems to me that records with YES, RUN, and RUNNING would process (i.e. send values to the device) on reconnect events under my proposal #3.
>
> Records with PAUSE, PAUSED could indicate records that should process on disconnect events?
>
> But this is really overloading the meanings of these values and the PINI field, and a new record field or fields would be better.
>
> Mark