Wireless motion detectors not triggering programs after migration to eisy.

[Oakland Jeff]

I can write updates to my wireless motion sensors, but when they are tripped, none of my auto-off programs are triggered like they used to trigger.

Even my RemoteLinc 2 won't trigger the scene it's part of... pic attached of scene test results.

It also seems relevant that in my main device overview screen, the status of my lights is not accurately reflected. If I do a manual query, they will update correctly, but I used to see real-time changes before I migrated.

Any ideas?

Edited August 7, 2023 by Oakland Jeff
incomplete description of issue

[Techman]

What controller are you using. What's its firmware version?

What did you migrate from and to?

You can also try saving the associated programs and do a reboot of the controller.

 

[Oakland Jeff]

Thanks Techman,

I migrated from ISY994i. Attached are my current eisy versions.

I want to make sure I understand what you are suggesting in the last sentence... Select My Programs from the tree, select Export from the context menu and save to a file. Then reboot the eisy and import all those programs again?

[Techman]

It's possible that during migration individual programs can get corrupted. 

Open each individual program associated with your devices and make sure that the program data is correct, then save the program.

If everything looks OK then reboot the Polisy/EISY. Keep in mind that after a reboot the device status window may not display data until the device changes state.

 

[Oakland Jeff]

OK, I saved all of my programs but got little dialogs each time that nothing had changed, so I'm not sure if anything was actually accomplished or not. Anyway, I rebooted and notice that 2 of my 97 programs say "Not Loaded" in the activity column and Out of Memory in the status column.

[Techman]

I'm not sure what would be causing those errors. Try updating packages, and then a reboot. If you then still have those errors you should probably submit a support ticket.

Also take a look at those two programs and make sure that there's no errors in the syntax.

[tazman]

1 hour ago, Oakland Jeff said:

OK, I saved all of my programs but got little dialogs each time that nothing had changed, so I'm not sure if anything was actually accomplished or not. Anyway, I rebooted and notice that 2 of my 97 programs say "Not Loaded" in the activity column and Out of Memory in the status column.

Does your firmware and UI match in Help About? It sounds like you need to clear your Java cache. 

[Oakland Jeff]

Techman, thanks, great insight! Those two programs both referenced a 2-wire dimmer that hadn't changed in forever, but for whatever reason its record was corrupted during the migration so the lines to beep it, turn it on, and turn it off were replaced by comments that said that node ###### was missing or something like that and that the desired action could not be taken. I restored those lines and the program now works, but none of the sensors are triggering the programs. I'm just bench testing by executing the Thens & Elses by hand.

 

 

Edited August 7, 2023 by Oakland Jeff

[Oakland Jeff]

So most of my motion sensors are the 2420Ms, and they fast flash when they are having comm problems. They are ALL fast flashing when they detect motion. When I watch the event viewer and I manually trigger the sensor by tapping the set button, the event viewer just sits there, silent, blank. To compare, I got my remoteLinc working, and I see 15 lines of events instantly when I tap a button on it. So while motion is not triggering events, the AC successfully writes updates when I do so on the Motions when they are in linking mode, according to the event viewer. So what's happening here? Why doesn't PLM/eisy see motion triggers, but they can communicate with the motion otherwise?

[Techman]

To put the MS into the linking mode you need to hold down the set button for 5 seconds. When it's in the linking mode its red led will steady blink. The MS will stay in the linking mode for 4 minutes or until you tap the set button twice. Once a MS is in the linking mode do a restore device. 

Note: The MS won't detect motion when in the linking mode.

 

[Oakland Jeff]

Hmmmmm... Identical, but still not triggering "control is switched on" in the IF of my programs like they used to before migrating...

[Techman]

The link tables look good, and your communication between the MS and the eisy is good being that you're able to read the link tables.

Try deleting and recreating your program.

[Oakland Jeff]

Oof, OK thanks, I'll give it a try next weekend... THIS weekend's project already spilled well into Monday and today. LOL! I'll reply with results later. Thanks again!

[Oakland Jeff]

So I'm not having any luck getting controls to trigger programs. Here's a dead simple one that's not working. It was created fresh after I went through each device and scene, restoring every device (and appreciating the smaller resulting link tables) and testing scenes, and editing them if necessary (by removing devices and replacing them if the Scene Test indicated they "failed"). My link tables should be pretty clean. After all that, all my migrated programs were not working, so I tried to create a new simple test one... seen below. It doesn't get triggered. I'm pretty confident coms is not the culprit, as I can perform any operation I want on that switch from the AC.

Edited August 14, 2023 by Oakland Jeff

[IndyMike]

On 8/7/2023 at 2:47 PM, Oakland Jeff said:

I can write updates to my wireless motion sensors, but when they are tripped, none of my auto-off programs are triggered like they used to trigger.

Even my RemoteLinc 2 won't trigger the scene it's part of... pic attached of scene test results.

It also seems relevant that in my main device overview screen, the status of my lights is not accurately reflected. If I do a manual query, they will update correctly, but I used to see real-time changes before I migrated.

Any ideas?

Hi Jeff,

Your event viewer above has a "Link for responder... not found" entry that I have never seen before.  Basically, you executed a Scene (00.00.28) and requested status (something the ISY normally doesn't do for scenes).  The Eisy did not expect to see the responder @1C.47.5E.

Something appears to be out of synch between your Eisy/PLM/devices.  You have a device that believes it's part of scene 00.00.28 but the Eisy doesn't know about it. I don't have the Eisy, but would guess that something went wrong during the migration.

The fact that you do not see Control triggers in the event viewer is also indicative of missing links in the PLM/Eisy.

I can't help much past that.  Unless someone else has experienced this, it may be time to open a ticket...

[IndyMike]

On 8/8/2023 at 1:08 AM, Oakland Jeff said:

Hmmmmm... Identical, but still not triggering "control is switched on" in the IF of my programs like they used to before migrating...

Ok, so I looked a little more.  I'm thinking something is hosed with your link tables. 

Normally the first entries are links back to your PLM.  The link table below is for my "all-on" detector.  It's only linked to my PLM @53.BC.3A. 

Your device appears to be linked to address 00.00.00 - Not Good.

If this were just 1 device, I would suggest deleting/reset/relink for this device.  If your entire system has this issue (you can look at the tables for a number of devices) I would again suggest opening a ticket.  

 

 

[Oakland Jeff]

IndyMike, thanks so much for sticking with this one. So, not all my devices have that null link where my PLM should be. For example, the switch that I used for my test... its link table seems fine in this regard. Attached are both the link table and the program. When I tap the physical switch on, the program doesn't get triggered.

Given this info, any other ideas before I submit a ticket?

[IndyMike]

Hey Jeff,

The most recent table is "better".  The first link @address 0FF8 shows that your device is a responder(A2) to group 0 device 70.1A.5A - your PLM I presume.

What I don't see is a corresponding link where your device is listed as a Controller to the PLM.  This would be

 0XXX: E2 XX 70.1A.5A.

What you have in the table is a HALF link.  The PLM can control your device, but the device can't issue control events back.

Again, you can probably rectify this by deleting/resetting/relinking your device.  If you have many devices that fall into this category, a ticket might still be a wise course of action. 

[Oakland Jeff]

OK, ticket it is. After a quick spot check, I don't think any of my devices have that E2 link to control the PLM. Thank you so much!!!

[Oakland Jeff]

I did one final verification. I deleted the switch, reset and relinked it. As you predicted, it now has a well-formed E2 link and it is now triggering programs as desired.

I'm hoping that the folks at UD can work some magic on a backup file or something so I don't have to manually delete and re-link and reconfigure everything.

Anyway, here is my happy-place new link table for the switch I was testing with.

[IndyMike]

Nice -glad it worked.

The UD folks are good at coming up with magic.  That's why we all here.