Experimental Physics and
Industrial Control System

"Kasemir, Kay" <[email protected]> · Tue, 3 Mar 2015 17:04:48 +0000

Hi:

For what it's worth, I can confirm that it's not an MEDM issue. CSS gives the same behavior: The long string (char waveform) served by pcaspy can be read and written just fine once it's been initialized, for example via

   caput -S ThatPV "Hello"

from the command line, or via

   self.setParam('ThatPV', [ 72, 101, 108, 108, 111 ])

inside my pcaspy-based CA server.

Until the PV is initialized, I get the same 

  No conversion between src & dest types no conversion between event app type=16 and DBR type=18 Element count=60

error.

Thanks,
Kay

On Mar 3, 2015, at 10:32 AM, Andrew Johnson <[email protected]>
 wrote:

> Hi Jamie,
> 
> On 03/02/2015 04:09 PM, Jameson Graef Rollins wrote:
>> Hi, folks.  We have applications using pcaspy (v0.4.1) to serve up 
>> character array records for long strings.  Whenever we use MEDM to 
>> connect to one of these character array channels, we see the
>> following error messages:
>> 
>> On the server side:
>> 
>> filename="../../../../src/cas/generic/casStrmClient.cc" line
>> number=895 No conversion between src & dest types no conversion
>> between event app type=16 and DBR type=18 Element count=60
> 
> DBR type=18 is DBR_TIME_CHAR, which is the data type MEDM will request
> for a channel with native type DBF_CHAR. The cas app type number 16 is
> gddAppType_value.
> 
>> On the MEDM side:
>> 
>> medmUpdateChannelCb: Bad status [400] for T1:TEST-SYS_STR: No
>> reasonable data conversion between client and server types
> 
> That message string is the error string for ECA_NOCONVERT, which
> matches the server's error.
> 
>> Using caget I don't see any particular issue with the channel
>> (other than maybe the "Status: UDF" and "Severity: INVALID"):
>> 
>> servo:~/ligo/src/guardian [master*] 0$ caget -d DBR_CTRL_CHAR
>> T1:TEST-SYS_STR T1:TEST-SYS_STR Native data type: DBF_CHAR Request
>> type:     DBR_CTRL_CHAR Element count:    60 Value:            0 0
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Status:
>> UDF Severity:         INVALID Units: Lo disp limit:           0 Hi
>> disp limit:           0 Lo alarm limit:          0 Lo warn limit:
>> 0 Hi warn limit:           0 Hi alarm limit:          0 Lo ctrl
>> limit:           0 Hi ctrl limit:           0 servo:~ 0$
> 
> The server error message comes from the cas code that handles monitors
> rather than gets, so you won't be able to replicate it exactly using
> caget, and camonitor doesn't let you select the data type to request.
> 
>> After the channel has been updated at all, even if it's been set
>> back to an empty array, new screens do not produce this error.  So
>> it appears there's something peculiar about how pcaspy is
>> initializing char arrays that is problematic.  If I configure the
>> server to set the alarm/severity so that they both show up as
>> NO_ALARM the problem persists, so I don't think it has to do with
>> that.
> 
> Very unlikely to be anything related to the alarm status/severity,
> those are just data fields to the CA client and server code and are
> not used for error reporting or handling.
> 
>> Anyone know what might be causing this, and how to avoid it?  Is
>> there a way to get a full dump of the channel record that might
>> help illuminate what it is that's different between the cases when
>> it the error is throw and when it's not?
> 
> I have no experience with pcaspy and not much with writing cas tools
> so I can't really help there, but these errors appear to be consistent
> with MEDM making requests that the server can't seem to handle.
> Whether this is a pcaspy or a cas problem I'm not sure though, any cas
> users out there want to chime in? Does anyone else implement long
> strings in a cas tool?
> 
> - Andrew
> -- 
> Doctorow's Law: Anytime someone puts a lock on something you own,
>    against your wishes, and doesn't give you the key, they're
>    not doing it for your benefit.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System