IndyMike

@kvolger , glad that you've thought through a few of the failure scenarios. It sounds like you are comfortable with the system risks. I hope things work well for you.

Your description of the pumping system concerns me. It sounds like you are transitioning from a dedicated system with hardwired sensors/safeties to a distributed system with communicating sensors/safeties. This doesn't sound like something that I would trust to home automation devices. At least not without backup sensors and over-rides (yes I've been burned). The 2450 IOLinc may seem like a handy solution, but it is powerline only communication (prone to communication errors) and will respond to Insteon All-on errors. I am not sure that the All-on matters if you are using it in a "sensor only" configuration. The micro ON/OFF switches are more recent technology and MAY be immune to All-On/OFF commands. Please do test this. Beyond the above, you need to think through failure modes: What happens if an micro on/off doesn't receive communication and doesn't turn on/off? Same for the 2450 - it doesn't transmit the contact open/close. What happens if the system is running and power fails? What happens when power is restored (many modules will default to the last power state). Sorry to be Debbie Downer here, but I'm concerned the "criticality" of this system. What happens when something doesn't turn on/off as planned? Is it a minor inconvenience, a moderate cost, or a major explosion? If the answer is an inconvenience, then by all means proceed. If other, then not as much.

I'm glad I'm not the only one. It's really bad when things suddenly break for some unknown/unfathomable reason. It's worse when I realize that I AM THE Reason. Like @dbwarner5, I use the find/replace feature often. I also disable programs and folders to prevent things from running unintentionally. Stuff still happens. More and more I am finding that my Home Assistant "test" programs are being re-enabled (me again) and running when they shouldn't. I have to say that UD did things right with their program/folders/variable organization setup. I wish HA would adopt something similar.

I just tried this on my ISY994 and it works as expected (I have not used the "Init to" previously). Since you have other programs working properly, I'm going to guess that your program didn't save properly for some reason. When this has happened in the past I've found it easiest to simply delete the program and start over. Do NOT copy the program. The visual representation will look correct, but the XML may be incorrect.

When in doubt, re-boot?

@pinlawr , thanks for the update. Please keep us posted as things go forward.

@Guy Lavoie , You are absolutely correct. I did not think that one through.

@Guy Lavoie , Rather than placing your motion sensors in linking mode, you can use a program to write updates to your motion sensors. The program watches for the device to "wake up", then writes the updates (thank you UD). Saves a lot of angst. I would think this would work for other battery devices as well (I have not tried it). In the past I have put all my motion sensor "Program" routines in a folder and disabled the folder. I enable the folder only when I want to write updates. BSMT Stair Sensor Program - [ID 0036][Parent 0066][Not Enabled] If 'Motion/RF / BSMT Stair.1 Motion' is switched On Or 'Motion/RF / BSMT Stair.1 Motion' is switched Off Then Set 'Motion/RF / BSMT Stair.1 Motion' Write Changes Else - No Actions - (To add one, press 'Action')

@hspindel , The instructions for the SH!)A are located here: http://www.authinx.com/manuals/X10/SH10A.pdf The indicate that you can use the device as an annunciator by pressing Unit Code-On followed by Bright or Dim on an X10 controller. I read that as the following: If - No Conditions - (To add one, press 'Schedule' or 'Condition') Then Send X10 'H6/On (3)' Wait 1 second Send X10 'H/Bright (7)' Else - No Actions - (To add one, press 'Action')Not sure if the wait is necessary. Since you are trying to mimic and button controller some sort of delay may be required.

@hspindel , Let me start by saying that I'm using the ISY994. It does allow me to send the "Bright" with only a House code as shown below. The result is different and allows a delay between the two commands. I can't say for certain that this option is still available on the EISY, but UD normally keeps items like this consistent across their platforms. I also cannot say that it will help with your problem. Which specific X10 siren are you using? Wed 04/08/2026 07:58:45 AM : [ Time] 07:58:47 4(0) Wed 04/08/2026 07:58:45 AM : [ X10] H6 Wed 04/08/2026 07:58:45 AM : [X10-RSP ] 02 63 D9 00 06 Wed 04/08/2026 07:58:55 AM : [ X10] H6/Bright (7) Wed 04/08/2026 07:58:55 AM : [X10-RSP ] 02 63 D5 80 06

@hspindel , You can add a delay by executing the House Code and Dim commands separately. By sending only the "H6" command you can insert a delay and then then the "H/Bright" command as shown below. I have no clue if this is related to your issue. Difficult to troubleshoot without a Siren to test. X10 Dim - [ID 0019][Parent 0003][Not Enabled] If - No Conditions - (To add one, press 'Schedule' or 'Condition') Then Send X10 'H6/Bright (7)' (Normal Bright Method) Wait 5 seconds Send X10 'H6' (Housecode address) Wait 2 seconds Send X10 'H/Bright (7)' (Bright command) Tue 04/07/2026 09:44:22 AM : [ Time] 09:44:51 3(0) Tue 04/07/2026 09:44:22 AM : [ X10] H6 Tue 04/07/2026 09:44:22 AM : [X10-RSP ] 02 63 D9 00 06 Tue 04/07/2026 09:44:22 AM : [ X10] H6/Bright (7) Tue 04/07/2026 09:44:23 AM : [X10-RSP ] 02 63 D5 80 06 Tue 04/07/2026 09:44:27 AM : [ X10] H6 Tue 04/07/2026 09:44:27 AM : [X10-RSP ] 02 63 D9 00 06 Tue 04/07/2026 09:44:29 AM : [ X10] H6/Bright (7) Tue 04/07/2026 09:44:29 AM : [X10-RSP ] 02 63 D5 80 06

@hspindel , The EISY and plm appear to be doing their jobs. The Top 4 entries in the viewer (green) are the EISY sending the X10 commands to the PLM. The Bottom 4 entries (blue) are the PLM echoing the commands back to the EISY. Presumably the PLM put the commands on the powerline. As @Brian H and @oberkc have indicated, try moving the Siren closer to the PLM to rule out signal absorption. Not sure if you need a delay between the bright and dim commands.

@NewB , Suffice it to say that things have gotten a bit hosed (technical term). The PLM links table you posted is extremely short (21 entries). The PLM is listed as the controller for all the devices in the ID column. There are no scenes listed. It is possible that the PLM read process was interrupted. Try this a couple of times to see if you get the same result. If so, proceed with the "recovery below". The device table you posted indicates that the "Dining Vanity Mirror" is a Responder to the devices listed in the ID Column. It can't control anything. Recovery: Please verify that your PLM address is 71.18.59. Since your PLM link table is incorrect (after multiple PLM restores), you'll need to perform an ISY restore. LOOK at your backup file sizes. If you have a recent backup that is significantly different in size (from previous), it's suspect. Choose a recent backup that looks consistent. If you get socket errors during the restore (or a backup), do not trust it. Logout of the admin console, clear the cache, log back in and restart the process. When the restore ISY complete correctly, it will reboot the ISY. Log back in, turn off "writes to battery devices", and perform a "Restore Modem". This should re-write the PLM records. After the above completes, re-check your PLM records. If it's still very low (<100) post back and we'll rethink things. It's possible that that the ISY will not re-write the device link tables in step 3 (if it didn't detect a PLM change). The Only way to know for sure is to perform "link table read/compare's" on devices. You can address these individually by running a "device restore", but if there are many, it's may be easier to use the "restore devices" from the file menu.

The ISY is still sending a group 12 Off command that may be affecting things. A program is being triggered because of your "scene test". Likely not the cause of everything failing, but it's definitely complicating things. Can you open the event viewer on level 3 and perform "Link Table Read" on "Dining Vanity Mirror (4F 3A F7 1)"? Copy the event viewer to the clipboard and paste to the forum. Please also post the contents of the Link table. I'm curious if you are getting Nack responses. The following are examples of the event viewer and link tables for one of my simpler devices. I've edited the event viewer to show only the 1st two link table reads. Sat 04/04/2026 09:48:19 AM : [INST-TX-I2CS] 02 62 54 A1 F5 1F 2F 00 00 00 0F FF 01 00 00 00 00 00 00 00 00 C2 Sat 04/04/2026 09:48:19 AM : [INST-ACK ] 02 62 54.A1.F5 1F 2F 00 00 00 0F FF 01 00 00 00 00 00 00 00 00 C2 06 (00) Sat 04/04/2026 09:48:20 AM : [INST-SRX ] 02 50 54.A1.F5 53.BC.3A 2F 2F 00 (00) Sat 04/04/2026 09:48:20 AM : [Std-Direct Ack] 54.A1.F5-->ISY/PLM Group=0, Max Hops=3, Hops Left=3 Sat 04/04/2026 09:48:20 AM : [INST-ERX ] 02 51 54 A1 F5 53 BC 3A 15 2F 00 00 01 0F FF 00 A2 00 53 BC 3A FF 1F 01 B8 Sat 04/04/2026 09:48:20 AM : [Ext-Direct ] 54.A1.F5-->ISY/PLM Group=0, Max Hops=1, Hops Left=1 Sat 04/04/2026 09:48:20 AM : [INST-TX-I2CS] 02 62 54 A1 F5 1F 2F 00 00 00 0F F7 01 00 00 00 00 00 00 00 00 CA Sat 04/04/2026 09:48:20 AM : [INST-ACK ] 02 62 54.A1.F5 1F 2F 00 00 00 0F F7 01 00 00 00 00 00 00 00 00 CA 06 (00) Sat 04/04/2026 09:48:21 AM : [INST-SRX ] 02 50 54.A1.F5 53.BC.3A 2F 2F 00 (00) Sat 04/04/2026 09:48:21 AM : [Std-Direct Ack] 54.A1.F5-->ISY/PLM Group=0, Max Hops=3, Hops Left=3 Sat 04/04/2026 09:48:21 AM : [INST-ERX ] 02 51 54 A1 F5 53 BC 3A 15 2F 00 00 01 0F F7 00 A2 06 29 53 3C FF 1F 01 4B Sat 04/04/2026 09:48:21 AM : [Ext-Direct ] 54.A1.F5-->ISY/PLM Group=0, Max Hops=1, Hops Left=1

Hey @NewB , The "scene test" is a nice tool, but is has some limitations. If there is activity on the powerline (motion sensors, devices activating) or Programs executing on the ISY, the PLM will exit the test. The red communication above looks like an ISY program ran and turned off group 13. It's generally best to disable programs to get consistent results. The communication above shows the ISY trying to write link records to the device @11.96.2A. The write operation times out because the device is not responding. This device should be tagged with the green "1011" in the admin console. The device isn't part of the scene that you tested, but if you have other devices with pending writes it indicates they do not have all of their links (scenes). Try resolving the pending writes to see if things resolve.