Experimental Physics and
Industrial Control System

"LEI Ge" <[email protected]> · Fri, 23 May 2014 07:21:06 +0800 (GMT+08:00)

Hi, all

Yesterday my colleague located a mal-functional hardware, and by this clue I checked the database logics in the IOC related to that hardware and found the PV which wrote 0 to our gun-trigger-enable PV.

Many thanks to all the help from tech-talk. We will try the software you provided soon, then if we have some other troubles in the future, it will be easy to locate them. :)

Have a nice day,

Ge

> -----原始邮件-----
> 发件人: "Ralph Lange" <[email protected]>
> 发送时间: 2014年5月21日 星期三
> 收件人: "EPICS Tech-Talk" <[email protected]>
> 抄送: 
> 主题: Re: how to know which PVs write to a PV
> 
> On 21/05/2014 10:54, LEI Ge wrote:
> > Hi, all,
> >
> > This week we met a problem, a PV's value should be 1 (to enable our e-gun trigger), but sometimes it becomes 0 unexpectedly. I checked the software logic I can remember to this pv, and have not found the reason. Maybe there is some logic I forget or I don't know, to write to this pv. So is there a way to check in our control system, which PVs write to this pv? Or, this PV is written from which IP?
> >
> > Thanks a lot ,
> >
> > Ge
> 
> Hello Ge,
> 
> There is a module to log all Channel Access Puts [1], which logs all
> Channel Access put operations to an iocLogServer, and should tell you if
> the PV has been changed from the outside.
> 
> DB writes from the same IOC are harder to log. You can always try to set
> the TPRO field of the target record. This will show you when it is being
> processed, and by which thread, but not exactly which other record is
> writing. (You should be able to find this by searching the IOC's
> database, though.)
> 
> Writes from records on other IOCs are done through Channel Access, and
> logged by CAPutLog.
> 
> Good luck,
> ~Ralph
> 
> [1] http://www-csr.bessy.de/control/SoftDist/caPutLog/

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System