Missing BLACS tabs

[Roger]

Hi,

For some reason, it seems like BLACS isn't loading all devices specified in the connection table. In my case, BLACS loads the PulseBlaster and the two Novatech 409B but isn't loading NI PXI/PXIe devices nor a Spinnaker camera:

This seems to be a recent issue so I'm hoping someone has ideas on what to check (or what I may have missed).

Thanks,

Roger Ding

[Johannes Schabbauer]

Hi Roger,

My fist guess for this would be: Try to delete the BLACS hdf5 file in the "app_saved_config" folder. When you start BLACS afterwards this file is created automatically and you have have to select the globals file in BLACS again (and maybe other changes if you did some changes to the default settings).

Best wishes,

Johannes

[Roger]

Hi Johannes,

Thanks for the suggestion. I delete the .hdf5 files but unfortunately that didn't seem to solve the issue.

The compiled connection_table.h5 includes the missing devices that aren't showing up in the BLACS tabs (e.g., PXI-6733, PXIe-6361) but it seems like BLACS isn't pulling in the devices. I may try rolling-back some of the packages and/or create a new conda virtual environment (again...).

Roger Ding

[Ian B. Spielman]

Roger,

I am seeing, as marked, three devices starting up and three blacs tabs, so I conclude that there are no problems at that level. Make 100% sure that the .py file that you are compiling as the connection table has the desired entries active. If yes, then open up the h5 file that the connection table compiles into and see what made it there.

— Ian

NIST Fellow
Group Leader, NIST laser cooling and trapping group
Fellow, Joint Quantum Institute

National Institute of Standards and Technology and the University of Maryland

----- WEB -----
http://ultracold.jqi.umd.edu

----- EMAIL ——
spie...@jqi.umd.edu
ian.sp...@nist.gov (NIST administration only)

----- ZOOM -----
https://umd.zoom.us/j/7984811536

----- MAIL -----
UMD:
2207 Atlantic Bldg.
College Park, MD 20742

NIST:
100 Bureau Drive, Stop 8424
Gaithersburg, MD 20899-8424 USA

----- OFFICE -----
UMD: Physical Sciences Complex, Room 2153

> On Mar 18, 2026, at 22:14, Roger <rog...@gmail.com> wrote:
>
> Hi Johannes,
>
> Thanks for the suggestion. I delete the .hdf5 files but unfortunately that didn't seem to solve the issue.
>
> The compiled connection_table.h5 includes the missing devices that aren't showing up in the BLACS tabs (e.g., PXI-6733, PXIe-6361) but it seems like BLACS isn't pulling in the devices. I may try rolling-back some of the packages and/or create a new conda virtual environment (again...).
>
> Roger Ding
>
> On Wednesday, March 18, 2026 at 2:22:31 AM UTC-6 johannes....@tuwien.ac.at wrote:
> Hi Roger,
>
> My fist guess for this would be: Try to delete the BLACS hdf5 file in the "app_saved_config" folder. When you start BLACS afterwards this file is created automatically and you have have to select the globals file in BLACS again (and maybe other changes if you did some changes to the default settings).
>
> Best wishes,
> Johannes
>
> Roger schrieb am Mittwoch, 18. März 2026 um 03:54:00 UTC+1:
> Hi,
>
> For some reason, it seems like BLACS isn't loading all devices specified in the connection table. In my case, BLACS loads the PulseBlaster and the two Novatech 409B but isn't loading NI PXI/PXIe devices nor a Spinnaker camera:
>
>
>

> This seems to be a recent issue so I'm hoping someone has ideas on what to check (or what I may have missed).
>
> Thanks,
> Roger Ding
>

> --
> You received this message because you are subscribed to the Google Groups "the labscript suite" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to labscriptsuit...@googlegroups.com.
> To view this discussion, visit https://groups.google.com/d/msgid/labscriptsuite/d42b6b63-52bc-45a4-b380-11794aef1696n%40googlegroups.com.

[Roger]

Hi Ian,

I should have taken a more informative screenshot. I was trying to show that BLACS wasn't instantiating NI devices defined in the connection table.

I attached an example connection_table.py and the compiled connection_table.h5. The screenshots below shows PXI-6733 and PXIe-6361 appearing in connection_table.h5:

but not being pulled into BLACS:

After some more investigating, it seems like using labscript==3.4.1 was the issue. BLACS correctly instantiated the NI devices after reverting to labscript==3.4.0.

Roger Ding

[Chris Billington]

Hi Roger,

Looks like a regression introduced by this recent change:

https://github.com/labscript-suite/labscript/pull/119

This change should fix it, and hopefully there'll be a release with the fix soon:

https://github.com/labscript-suite/labscript/pull/122

Cheers,

Chris

To view this discussion, visit https://groups.google.com/d/msgid/labscriptsuite/f8d2ccc1-60d5-4471-8bd4-d700d4d4be96n%40googlegroups.com.

[dihm....@gmail.com]

Labscript==3.4.2 is released with the fix. Let us know if that sorts everything.

-David

[Roger]

labscript==3.4.2 seems to have fixed the issue.

Roger Ding