Experimental Physics and
Industrial Control System

Michael Davidsaver <[email protected]> · Tue, 24 Mar 2015 21:55:20 -0400

On 03/24/2015 08:30 PM, Alireza Panna wrote:
> Hello tech-talkers,
> 
> Currently I am using procServ and devIocStats to successfully
> restart/reboot linux and windows based iocs. I did have a couple of
> questions though
> 1. What is the limit on the log file that procServ generates

2GB on 32-bit Linux.  On a 64-bit system it will probably fill up your
disk before it hits any API limit.

Of course OS and system policies may limit this further (cf. "ulimit -f").

> and does
> the ioc stop running through procServ if this limit gets exceeded?

It keeps running (determined experimentally).

> 2. On windows when I run an epics based ioc using procServ, a blank
> terminal opens up seperately for procServ.exe. Is there a way to not
> display this. I am sure there is a way and I haven't searched hard enough!

Probably not without code changes, although someone with more MS
experience should confirm.

> 3. For pcaspy based iocs, it is fairly easy to exit the ioc by just
> adding another PV to exit the python server, however I haven't found an
> elegant way to restart the IOC  automatically similar to what procServ
> does. Any suggestions?

Run your python app in procServ?  It works for any process and is a
handy way to daemon-ize programs which weren't designed with this in mind.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System