![]() |
![]() ![]()
Experimental Physics and
| ||||||||||||||||
|
Hi Guobao,
On Tue, May 15, 2018 at 10:36 AM, Shen, Guobao <[email protected]> wrote:
Hm, maybe. Perhaps the Jupyter kernel has different environmental variables set for how it finds the CA shared library, and how it searches for PVs? Getting that far suggest that a) the CA shared library was found and initialized, b) that the channel was found and connected, c) that a `ca_array_get_callback()` was called and did not return a SEVCHK error, but that the program gets stuck waiting for the get callback to actually run. The 5 second timeout for get() on a connected channel seems plenty long enough. We might need a little bit more information to go on. First, can you verify that doing import epics print(epics.ca.find_libca()) gives the same libca as when running python from the Terminal? Second, if you do import epics chan = epics.ca.create_channel('test') conn = epics.ca.connect_channel(chan) print("Connected ", chan, conn== epics.dbr.CS_CONN) val = epics.ca.get(chan) do you really see that the channel is connected? Do you get the same results from running python in the Terminal?
I'm not sure. It may be that cothread does not always do `ca_array_get_callback()` and then expect that the callback will be called. I don't know why that would cause a problem from Jupyter for you. --Matt Newville
| ||||||||||||||||
ANJ, 15 May 2018 |
![]() · Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing · |