DDF hot reload doesn't re-evaluate match conditions

[ebaauw]

Does the issue really belong here?

• I definitively want to report a bug within deCONZ or its REST-API

Is there already an existing issue for this?

• I have searched the existing issues and there is none for the bug at hand

Describe the bug

Got some new LTA009 Hue lights with ancient firmware, not yet exposing the Hue Effects (0xFC03) cluster. After firmware upgrade, and reload of the Simple Descriptor, the GUI would show the cluster. However, Hot Reload doesn't link the DDF to the device. The DDF would only be matched after deCONZ restart. Note that the DDF has a filter on the existence of the cluster: "matchexpr": "R.hasCluster(0x0b,0xfc03)"

Steps to reproduce the behavior

1. Pair LTA009 with old firmware and check that it doesn't expose the Hue Effects cluster;
2. Update the firmware through the OTAU plugin, re-read the Simple Descriptor and notice that the light now exposes Hue Effects;
3. Try Hot Reload (on another LTA009 that was already exposed through the DDF), and notice the new LTA009 remains unmatched;
4. Try to select the DDF manually on the new LTA009 and Hot Reload, and notice the new LTA009 remains unmatched;
5. Restart deCONZ and notice the new LTA009 is matched to the DDF.

Expected behavior

Not sure, but I would expect the DDF to be (re-) assigned after updating the device firmware.

Screenshots

No response

Environment

• Host system: Raspberry Pi
• Running method: Raspbian
• Firmware version: 26780700
• deCONZ version (not Home assistant Addon version!): 2.28.1
• Device: RaspBee II

deCONZ Logs

No response

Additional context

No response

[SwoopX]

Interestingly, I had a rather similar experience yesterday while trying to add support for the frient pet aware motion sensor. DDF was there and correctly using name and ID but reading those attributes did not trigger the DDF association mechanism 🤔 Had to delete it in deconz and reset'n're-pair to make it work.

[bluemoehre]

I ran into similar problems - not only using hot reload, but also after restarting deCONZ I had flaky results:

• deleted devices have still been known by name after re-adding.
• also old DDF was matched even if device had new firmware and matchexp should not match anymore.
• additionally I have 3x identical devices: one was not matched by DDF while all others were matched.

More info is here (in case it helps):
#6251

[github-actions]

As there has not been any response in 21 days, this issue has been automatically marked as stale. At OP: Please either close this issue or keep it active It will be closed in 7 days if no further activity occurs.

[ebaauw]

Bump

[github-actions]

As there has not been any response in 21 days, this issue has been automatically marked as stale. At OP: Please either close this issue or keep it active It will be closed in 7 days if no further activity occurs.