Strange problem -- Looking for ideas

[IndyMike]

Jack,

 

You've noted several times that the device has issues at the Sunrise + 10 minute point, but functions normally otherwise.  We are looking for a noise source / signal absorber that is RELIABLY active at that time.  Assuming that you aren't actively turning on another device programmatically, the only other device I can think of would be a photocell.  This could be outdoor or indoor, but normally the indoor night lights don't pull enough power to be a problem.

 

I have had issues with photocells generating noise in the past (post lamp).  They can generate noise due to poor design, or aging of the sensor itself.  I finally removed the sensor from my post light and used an Insteon switch to activate programmatically.

 

Other possible offenders could be timed kitchen devices (coffee pot), a setback thermostat warming the house around that time, etc.

Edited December 20, 2017 by IndyMike

[jrainey]

An update since the last tests.

 

As I mentioned, I had changed the execution time for the standalone sink light program to 01:00.  This worked perfectly.  Since I was still awake, I changed the execution time to 07:00.  The light was off when I got up this morning.

 

Before going to bed, I browsed around the forums and found a note of advice to someone about upgrading firmware .. the firmware and UI versions must be identical.  When I checked mine I found that I had firmware V4.6.2 and UI V4.5.2!  This has since been rectified.

 

Feeling a little stupid right now.

 

The program has been returned to it's former version and we will see if this is solved tomorrow morning.

 

Will provide an update tomorrow.

Edited December 20, 2017 by jrainey

[jrainey]

Thought I had identified and resolved the problem.  Nope.  Actually it is worse now.  Multiple devices are not responding to program initiated scene changes, when that program is run on a time schedule.  The devices show incorrect status in console, and when queried will show the correct status.  When the programs are run manually, everything works fine.

 

 

At this point, I am not sure what next steps to take.

 

Any suggestions would be eagerly welcomed.

[stusviews]

Now that both the UI and firmware are identical, start troubleshooting anew. You already have some tools at hand, for example, creating an entirely new program. Do not use copy. Start with the original condition and one Then statement. Each day, add another one or two statements to be executed, no more. There are (usually) no time savers to troubleshooting.

[oberkc]

I have had some issues with other devices (roku, TV) lately that were solved by the most simple of steps: reboot.  Now that you have compatible UI and firmware, have you power cycled the ISY?

[ScottAvery]

Thanks Paul,  good to hear from you.

 

Had not tried a factory reset.  Waited until my wife finished the dishes, then removed from scene, did factory reset, and added back into scene.

 

Now the device is controllable directly, but  not with the scene.  With the scene, it shows the light turning off and on in console, but the light itself is not turning off and on.  Will keep testing,but it looks like I will be changing the switch.

 

BTW, how is the Hydreon doing?

 

Jack 

 

Going back to this previous post, It looks like you factory reset then added to a scene without restoring the PLM link.  The ISY will not have status for the device in this case without querying.  I would delete from ISY, reset, and re-add to ISY and scene.

[jrainey]

Again, thanks for the feedback.

 

I will be away over Christmas, so will not be doing any updates for a while.

 

In the mean time, here is what has happened since my last update.

 

1) The worsening of the situation appears to have been the result of a Restore Devices and/or Restore Modem (PLM) procedure that did not complete cleanly.  I upgraded my ISY to PRO, gathered in my outdoor motion sensors (had to fight my way through several feet of snow) and redid the procedures, all the time watching and ensuring that each and every device was restored properly.  The next morning everything had executed normally with the exception of that pesky kitchen sink light.

 

2) Created a support ticket with UDI.  Michel got back to me and suggested that I move the PLM to another outlet.  Unfortunately easier said than done, as the PLM/ISY combo is mounted behind an access panel in my basement ceiling, connected to a dedicated outlet.  setting up the necessary network connection and plugging the PLM into a convenient outlet will take some effort.  Will attempt that after the holidays if the situation is still not resolved.

 

3) I added and additional program to turn off the kitchen light 20 or 30 minutes after the sunrise program.  The next day the light was still on.  (aaarghh)

 

4) Did nothing, and the next day, the light was off in the morning as it should have been. (yay!!! ... i think).  Was totally puzzled by this until my wife informed me that the previous evening she had noticed that the light was on, but showing as off in mobilinc.  She exercised the light several times through mobilinc several times and then left things alone.  Now instead of being totally puzzled I am only completely puzzled.  She claims she didn't want to disturb me as I was re-watching The Force Awakens in preparation for going to The Last Jedi with my son-in-law and the grandsons.   

 

Again, thanks for the suggestions. 

 

stusviews .. if the problem re-emerges I will absolutely go back to square one and start troubleshooting.

 

oberkc .... I have soft-reset the ISY many times during this struggle, but so far I have not power cycled the system .. will do that in the new year.

 

ScottAvery .. I actually did delete the device before the factory reset and added it back in after.  It was also deleted and re-added when I changed it out for a new device.  In both cases the behavior continued.  Thanks for that note though.

 

 

Merry Christmas everyone, and if there are any amazing spoilers for The Last Jedi - I don't want to hear about them.

 

Thanks again,

 

Jack  

 

 

[stusviews]

1) I found that sometimes things (e.g., scenes, programs) that should have worked but didn't, did a day later,

 

2) Do you mean that your "dedicated outlet" has absolutely nothing else on that circuit breaker?

 

3) & 4) See response 1)

[oberkc]

Keep in mind that "moving" the PLM might be nothing more than keeping the PLM in the same general location, but plugged into a long extension cord from another outlet.

[jrainey]

Merry Christmas guys.

 

stusviews - yes a dedicated breaker - absolutely nothing on it except the PLM and the ISY power supply.  Installed for that purpose when the basement was remodelled.

 

oberkc - great suggestion - did not think of that at all!

 

btw .. the light was off this morning.

 

thanks

 

Jack

[jrainey]

Hi Guys (and Gals if applicable)

 

Back home from my holiday travels,  and still struggling with system problems.

 

While away, the system was erratic in a number of ways.

 

Sometimes the Christmas lights didn't turn on .. as observed by my next door neighbour.  Was able to turn them on with a struggle through Mobilinc.

 

Status of devices in Mobilinc were (at random) sometimes correct, sometimes incorrect and frequently unknown.

 

All in all, while I was away the system that was first set up in 2010 and has been running steadily ever since was a bit of a headache while I was away.

 

When I returned, I did a hard reset, and things seemed to straighten out  .... for a while.  In the last few days issues re-appeared.  Certain lights did not turn on, others did not turn off.   One of my simple bathroom fan programs stopped working.

 

Hard reset again about a half hour ago.   Have yet to try the PLM move item. but will do that tomorrow.

 

In the meantime, my frustration level has reached the point that I am researching Homeseer controllers.

 

I also noticed that my problems bear some resemblance to what smokegrub is describing in 'The PLM Saga continues' topic

 

Thoughts anyone?

 

Thanks

 

Jack 

[oberkc]

Still thinking comms problem and still waiting on results from earlier suggestions and thoughts.  Something is interfering with insteon communications.  Unfortunately, it is often not easy to identify the culprits.  Moving the PLM can sometimes yield a clue (and still waiting for you to do this, I think).  When the easy stuff is tried without success, the only way I know to identify a problem is to unplug suspect gadgets (TV, theaters, UPS, surge suppressors, LV lighting, power supplies and chargers, old fluorescent fixtures, etc....) and see if this solves the problem. 

 

Yes, it is tedious and PIA.  

 

IN case it has not been asked...you have confirmed communication between the legs of your electrical system, correct?

[jrainey]

Thanks oberkc.

 

One of my first purchases (and installations) with this system was a passive phase coupler, along with a couple of RF range extenders.  Up until now I haven't had any comm problems, and I have many devices installed on both legs of the electrical system.  If there is a specific procedure to confirm communication between the legs other than this, I am unaware of it, and have not performed it.

 

Issues with the system increased over the past several days, to the point where few of the programs worked, and few devices would operate remotely.  

 

Today I finally pulled out an extension cord and prepared to 'move' the PLM.  The first observation I made was the the PLM was loose in the outlet, and I immediately thought that this could be the root of the problem.  I proceeded anyway to plug the PLM into the extension cord, reset the system, and watched through the admin console as the system restored itself.  There were many 'cannot communicate with device' messages, and only a few devices came up good.  After several tires, I am now at the point that the only devices that are visible to the ISY are the few z-wave devices that I have.

 

I have concluded that my PLM has failed, although I am puzzled that it lasted seven years, and seemed to fail gradually over a number of weeks, starting with minor, but consistent comm issues that occurred only when programs were run, progressing to a total failure.  I was under the impression that these devices tended to fail catastrophically after a couple of years due to low quality capacitors used in the manufacture.

 

I have ordered a replacement for the PLM, which should arrive in four or five days.  We'll see if the issues continue with the new device.

 

Again, thanks to all for your help with this problem.

 

Jack

 

[oberkc]

I have concluded that my PLM has failed, although I am puzzled that it lasted seven years, and seemed to fail gradually over a number of weeks, starting with minor, but consistent comm issues that occurred only when programs were run, progressing to a total failure. ​

 

I suspect that this is consistent with other's experience.  Certainly it is with mine.  

 

I had believed, up to the point of my system "failures" that devices such as the PLM either worked or didn't...failure was all or nothing.  After seeing a new PLM greatly improve reliability of my comms and system, I had to adjust my beliefs.

[stusviews]

PLMs, like many other devices, seem to show signs of age. This is consistent with the common repair, replacing capacitors. Capacitors age, especially those involved with power.

[larryllix]

Thanks oberkc.

 

One of my first purchases (and installations) with this system was a passive phase coupler, along with a couple of RF range extenders.  Up until now I haven't had any comm problems, and I have many devices installed on both legs of the electrical system.  If there is a specific procedure to confirm communication between the legs other than this, I am unaware of it, and have not performed it.

 

Issues with the system increased over the past several days, to the point where few of the programs worked, and few devices would operate remotely.  

 

Today I finally pulled out an extension cord and prepared to 'move' the PLM.  The first observation I made was the the PLM was loose in the outlet, and I immediately thought that this could be the root of the problem.  I proceeded anyway to plug the PLM into the extension cord, reset the system, and watched through the admin console as the system restored itself.  There were many 'cannot communicate with device' messages, and only a few devices came up good.  After several tires, I am now at the point that the only devices that are visible to the ISY are the few z-wave devices that I have.

 

I have concluded that my PLM has failed, although I am puzzled that it lasted seven years, and seemed to fail gradually over a number of weeks, starting with minor, but consistent comm issues that occurred only when programs were run, progressing to a total failure.  I was under the impression that these devices tended to fail catastrophically after a couple of years due to low quality capacitors used in the manufacture.

 

I have ordered a replacement for the PLM, which should arrive in four or five days.  We'll see if the issues continue with the new device.

 

Again, thanks to all for your help with this problem.

 

Jack

Have you just tried to insert a Wait 2 seconds between each scene in the list? You may have created an Insteon data storm with some new device, or some change in system, or a failing PLM etc.. It would be worth a try. Insteon is the weak link in the ISY system most of the time.

 

Also, I should mention the delayed write to Battery operated devices can cause big problems. I have MSes that cannot handle it, and everytime an MS saw motion, ISY would operate my delayed update programs to write to devices, that couldn't handle it. They would never ACK.  It seemed like my ISY write caches were laden full, and bogging my whole Insteon system down with every motion sensed. About ten minutes of clearing these caches with my MSes, deleting the update programs, and turning that nonsense off worked wonders. Now I can put an MS into linking mode, write an option change and it finishes in 30 seconds instead of ten minutes. Now my constant Busy Box in admin console actually goes away.

Edited January 10, 2018 by larryllix

[jrainey]

Thanks for the suggestions larryllix.

 

I have tried many combinations of waits and scene mods, but always ended up with some failures.  At several points it was a single device within a scene that wa not responding.

 

For the motion sensors, I splurged and got the pro version, which allows turning off automatic writes to these devices.  In my case, the main reason is I'd rather not dig my way through several feet of snow to put these things into linking mode.  Hoping that the new motion sensors will not require physical intervention to allow this.  (which rasies the question - when will version 5 firmware be ready?)

 

Thanks

 

Jack

Edited January 18, 2018 by jrainey

[larryllix]

Thanks for the suggestions larryflix.

 

I have tried many combinations of waits and scene mods, but always ended up with some failures.  At several points it was a single device within a scene that wa not responding.

 

For the motion sensors, I splurged and got the pro version, which allows turning off automatic writes to these devices.  In my case, the main reason is I'd rather not dig my way through several feet of snow to put these things into linking mode.  Hoping that the new motion sensors will not require physical intervention to allow this.  (which rasies the question - when will version 5 firmware be ready?)

 

Thanks

 

Jack

allows turning off updates.

 

Careful with this.

I have MSes that are not capable of using delayed update writes, and my system ground to a halt sometimes when an MS would see motion. ISY was trying to use my auto-update programs to update a device, ad nauseum, and Insteon comms were tied up for 20-30 seconds at a time, before other any other traffic could go through.

The busy box was almost always present in the admin console.

[jrainey]

Hi All

 

Received my new PLM today, installed it, restored and everything went without a hitch.  System is up and running with no apparent glitches.

 

The test will be tomorrow, when the kitchen sink light should turn off.

 

If anyone is interested, the PLM info is as follows:

 

                             Firmware Rev.                    9E

                             Hardware Rev.                   2.4

                             Manufacture date               2917

 

Thanks

 

Jack

Edited January 18, 2018 by jrainey

[paulbates]

Hi All

 

Received my new PLM today, installed it, restored and everything went without a hitch.  System is up and running with no apparent glitches.

 

The test will be tomorrow, when the kitchen sink light should turn off.

 

If anyone is interested, the PLM info is as follows:

 

                             Firmware Rev.                    9E

                             Hardware Rev.                   2.4

                             Manufacture date               2917

 

Thanks

 

Jack

 

Jack

 

One thing to keep in mind is that the ISY can record some of the corrupted data from the failed PLM, and that gets backed up..., and restored to the new PLM. If you have problems with the scene tomorrow, I would remove the affected items from the scene(s) and re-add it

 

Paul

[jrainey]

Thanks Paul -- will watch for it.

 

Jack

[jrainey]

Hi all,

 

Well, the new PLM has been in place for one complete day, and I can report that the system is up and running without issue during that entire time.  I have pretty high confidence that the issue has finally been discovered and fixed, and that this will be the last post on this topic.  (meanwhile keeping my fingers crossed just in case)

 

Thanks to all who have read the topic and offered troubleshooting suggestions  I have learned a lot.

 

I will add a edit to the first post noting that the 'strange problem' was a result of a slow motion failure of an old PLM.

 

Again,  thanks to all,

 

Jack