-10 error clarifications

[johnnyt]

I occasionally get -10 errors in ISY error log. I was told it indicates a defective PLM. The official error code description is "-10    UNEXPECTED DEVICE RESPONSE".

It has occurred intermittently with 4 different PLMs of different vintages that I've gone through with no noticeable issues. It's possible I just missed the issues when they happened but the point is that overall my different PLMs have been working. Furthermore I've had 3 of the four PLMs refurbished to upgrade the capacitors and other components designed to make them last longer by this ebay service with positive feedback and it didn't eliminate the errors. The guy who did the refurbishing didn't really know what this ISY error meant. He was familiar with ISY but not a user of it himself.

Does anyone know what they mean in practice? Why wouldn't upgrading capacitors and other pieces not fix them?

Any info would be appreciated.

[Michel Kohanim]

@johnnyt,

PLM is dying or will die. It's just a sign of deterioration in performance when the PLM responds with junk and thus -10.

With kind regards,
Michel

[johnnyt]

1 hour ago, Michel Kohanim said:

@johnnyt,

PLM is dying or will die. It's just a sign of deterioration in performance when the PLM responds with junk and thus -10.

With kind regards,
Michel

wouldn't refurbishing them fix this? 

[Brian H]

If it was a power supply capacitor issue. I would think replacing them would help.

If is the serial port chip then it may not help. At V2.3 they changed the chip to one with better ESD ratings and V2.4 and up has a new serial daughter board with the better chip and some signal protection on them.

If you are having many PLM's giving the -10. You may want to verify the cable between the PLM and the ISY controller is 100% OK.

[johnnyt]

I went through error logs going back to before I got the PLMs refurbished. In about a month of testing 4 PLMs prior to refurbishing 3 of them, I saw about 23 "-10" errors in about 3 1/2 weeks. After the refurbishing, the first refurbished one I tested (a v 1.7) simply would not hear anything anymore. While it may have had some -10 errors, it was hearing insteon events before I refurbished it. I expect I will sent it back but waiting to see what else I might need to send in again.

The second one I tested (v 2.6) has now been in production for at least 6 weeks and I found 29 "-10" errors in that time, so about 4.8 errors per week. That's more than I thought was happening. It seemed less frequent when I was spot checking things. I've replaced it (today) to see if the third one I had refurbished (a v1.C) does any better.

I do see the ebay PLM fixer has a "Communication Repair" service that I did not think was my situation since things were working in general (I actually really notice anything not working). Before I go potentially throw out good money after bad, would the -10 errors fall into this category? (The first part was done as part of the refurbishing so the part after "Also" is the key stuff I didn't get done.)

 

Quote

Communications Repair: Included in this repair is the replacement and upgrade of all electrolytic capacitors failed or not. Also replaced are failed components including the serial or USB communications chip, RJ45 jack, USB jack, diodes, resistors, and other items with the exception of the main processor as these are not obtainable (but we have yet to see one fail).  

 

[johnnyt]

On 12/5/2022 at 3:51 PM, Brian H said:

If you are having many PLM's giving the -10. You may want to verify the cable between the PLM and the ISY controller is 100% OK.

A couple days ago, on your suggestion, I replaced the cable to the PLM with a brand new cable that I first tested using my managed switch 'cable test' function. While the old cable also tested okay, I was nonetheless hopeful about the new cable because I took the opportunity to check/restore all my keypadlincs (about 8 of them), which almost all had some bad links (common after a while or when replacing PLM, which I did 3 days ago). I figured that was a good stress test and no errors were reported.

But, alas, last night I got hit with a couple -10 errors during query all, which is when majority of them occur. I think I'll send one back to refurbisher see if the comms repair was the issue I should have sent them in for in the first place...

[MrBill]

It's possible that there is another failing component that doesn't often fail.  You've tried 4 different PLMs so that suggests it could be the device at the other end of the cable..i.e. the 994.  while it's uncommon to have serial port issues there... it's not impossible.  

also this was an interesting post yesterday on the range of issues with bad PLMs:

 

[johnnyt]

32 minutes ago, MrBill said:

It's possible that there is another failing component that doesn't often fail.  You've tried 4 different PLMs so that suggests it could be the device at the other end of the cable..i.e. the 994.  while it's uncommon to have serial port issues there... it's not impossible.  

also this was an interesting post yesterday on the range of issues with bad PLMs:

<snip>

Thanks! Interesting other post, especially about the repair service in Toronto area (more than just PLM too) as I'm in Canada. Missing info on what one gets for how much, and history/depth of feedback compared with NY-based ebay service I've used but will keep close eye on that.

RE: 994i serial port. Interestingly I replaced the 994i I was using just last weekend with an older one to fix unrelated issues but I still see -10 errors with this other one. Would be even more rare to see that problem with two separate units.

[MrBill]

Have you opened a ticket?  sounds like it might be time:  support@universal-devices.com

[IndyMike]

I will start by confessing that I didn't know what the -10 error meant.  After reading your posts, I went back through a number of error logs that I had saved. 

FWIW, the -10 errors are rather common on my system.  I have not noticed any performance issues over the time frame covered below. The 4.8 errors per week are lower than my most recent log, but higher than the previous.  

I am not sure what to make from this.  I have never experienced a PLM failure.  I've replaced several units over the years due to various improvements, but no failures.  I seem to be blessed by "good" power (Hate letters to the normal Email please).   

I also have logs going back to the 2011 time frame that show the -10 error.  They seem to come in clumps, with no particular day/time preference.  If I weren't old and lazy I would try to run an analysis on the data to see of there was a correlation with other events going on in the system...  I'd be interested to know if there appears to be a time/day preference for your errors.

If you are having errors across several PLM's, MrBill's advice is sound.  Submit a ticket to try to determine if the errors are due to hardware or environment.

Firmware Start Date End Date Days Errors Errors/Day V5.3.4 7/15/2022 11/29/2022 134 125 0.9328358 V5.3.4 10/5/2020 7/13/2022 638 270 0.4231975 V5.3.4 12/8/2019 3/7/2020 89 239 2.6853933

[johnnyt]

9 minutes ago, MrBill said:

Have you opened a ticket?  sounds like it might be time:  support@universal-devices.com

oh yes. before I posted here. The answer I got that way was basically the second post above.

 

[johnnyt]

39 minutes ago, IndyMike said:

I will start by confessing that I didn't know what the -10 error meant.  After reading your posts, I went back through a number of error logs that I had saved. 

FWIW, the -10 errors are rather common on my system.  I have not noticed any performance issues over the time frame covered below. The 4.8 errors per week are lower than my most recent log, but higher than the previous.  

I am not sure what to make from this.  I have never experienced a PLM failure.  I've replaced several units over the years due to various improvements, but no failures.  I seem to be blessed by "good" power (Hate letters to the normal Email please).   

I also have logs going back to the 2011 time frame that show the -10 error.  They seem to come in clumps, with no particular day/time preference.  If I weren't old and lazy I would try to run an analysis on the data to see of there was a correlation with other events going on in the system...  I'd be interested to know if there appears to be a time/day preference for your errors.

If you are having errors across several PLM's, MrBill's advice is sound.  Submit a ticket to try to determine if the errors are due to hardware or environment.

Firmware Start Date End Date Days Errors Errors/Day V5.3.4 7/15/2022 11/29/2022 134 125 0.9328358 V5.3.4 10/5/2020 7/13/2022 638 270 0.4231975 V5.3.4 12/8/2019 3/7/2020 89 239 2.6853933

very interesting. thanks for posting, @IndyMike.

It's UDI support that brought the -10 errors and what they mean to my attention when looking into something else. Even the ebay guy that refurbished 3 of the 4 PLMs I have is a little surprised that all three of my refurbished PLMs are seeing "so many problems". The reality is that I'm not noticing any problems in real life. There may be a bad/missed command at the time of the -10 that recovers thanks to retries? Overall, though, my insteon is working as well as it has been in general. While I don't understand why devices (mostly KPLs) end up with bad links over time, hey, I've been living with it for more than a decade now so that's Situation Normal AFU.

So what is this problem really? is it it a 994i issue, will any repair one can do today fix it? Also, does the PLM/ISY recover from it, meaning they are really just warnings, if that? Also, if it is a warning sign of a degraded unrepairable PLM, could it still function for another year or two? Looks like it has worked fine for you for 3 years!

Maybe just having 4 spares with good capacitors will allow me to continue using Insteon until I can move off it (which is the plan, but I did want to drive my multi thousand dollar insteon device investment into the ground)

Edited December 8, 2022 by johnnyt

[IndyMike]

2 hours ago, johnnyt said:

So what is this problem really? is it it a 994i issue, will any repair one can do today fix it? Also, does the PLM/ISY recover from it, meaning they are really just warnings, if that? Also, if it is a warning sign of a degraded unrepairable PLM, could it still function for another year or two? Looks like it has worked fine for you for 3 years!

I, for one, have no clue what the implications of the -10 error are.  I have apparently had this occurring for around 12 years with no observable problems.  At this point I am in about the same position as you.  I have a number of spare PLM's and very little incentive to investigate this further.  I do have one advantage - advanced age.  I'm thinking I'll allow CRS to take over (Can't Remember S#*5).  Tomorrow I'll be feeling great again.

[IndyMike]

OK, I lied.  I couldn't let things go.   I ran a histogram on the last Error data set.  My errors peaked in late summer and dropped off significantly during fall.

Two things were changed in this time frame:

1) I removed my Schlage Z-wave lock from the system in Mid August.  It was a problematic 300 series device and may have been loading the ISY with retries (programmatic) .

2) I was running a ~20 minute timer program cycling my AC to prevent Evaporator Freeze up (15 minutes on/5 off).  This would have stopped in the September time frame (New HVAC).

I am thinking these errors are environmental.  I don't remember the flow control between the ISY/PLM and am wondering if asynchronous events (motion sensors) may be colliding with synchronous events (timer programs). 

Either way, I am generating far fewer errors now than I was in July with no change to PLM/ISY994 hardware.  I like that direction.

Thoughts, comments, catcalls appreciated...

 

[johnnyt]

Thanks @IndyMike. 

Very interesting to see you were getting 8-16 errors a week for 8 weeks, and no less than 1-2/week the rest of the past 5 months (without a PLM replacement I assume.) If my 4 PLMs can last at least 6 months each once the errors start coming up, and retries make them benign from a functionality perspective (though I don't know if that's the case), that would give me 2 years to plan/execute my move off insteon. I can probably live with that.

Also very interesting that you wonder if it isn't because of heavy ISY load. I began to see my ISY overloaded (underpowered) at least 2-3 years ago. And despite multiple efforts at rewriting entire segments of programs with generous "waits" and constant considerations for timing of actions, I continue to intermittently but repeatedly suffer with slow response and even spontaneous reboots suspected to be because of overloads. Today, with 173 insteon nodes, 152 zwave nodes, and 990 programs, I don't think I can go 2 weeks without some kind of problem (crash, spontaneous time changes to 2 years in the future, spontaneous reboots, etc.). While I don't believe all my problems are overload related, I'm really hoping the move to IoP will eliminate many of my problems. (Just waiting for the zwave migration to be there, which, unfortunately, suffered another setback last week.)

All that said, I'm seeing less than 5 errors/week. If it was an overload issue I'm thinking I would be seeing at least as many as you, if not more.

You also made me curious and I went back looking for old error log files and with dngrep (for Windows) and excel, plus some internet how to's, I pulled together a historical view of -10 errors. I wasn't able to figure out how to do a chart by week when my data was by day for a straight apples-to-apples comparison but here's what I have between June 2019 and May 2021.

 

Except for one day in Dec 2019 and a period between April and June 2020 I'm not seeing that many -10 errors. I'm guessing I likely replaced a PLM in that time period but I can't confirm exact dates when I've replaced PLMs. Fall 2022 was pretty consistent with the above, and never more than 1-2 in one day for the latter.

I'm still left wondering if -10 errors recover with retries (i.e. are they generally benign) and how long / how many -10 errors I can take with a PLM before it needs to be fixed or replaced. Also, for that matter, what the fix for -10 errors is, e.g. capacitor? comms? processor? etc.

Thanks again for your insights.