waffles Posted July 17, 2017 Share Posted July 17, 2017 Sometimes my setup (ISY994 and approx. 94 devices) exhibits some flaky responsiveness: commands are being missed, long delays, etc. So I thought this points to a PLM that is slowly failing. I did a PLM count and got consistently 58. Then I replaced the PLM with a spare one I had for a few years: 529 counts (consistently and after PLM restore). Afterwards, I tried a 3rd, seemingly new PLM and got 276 counts (again, consistently and after PLM restore). Any thoughts as to why I get so different results and how I can determine what the correct count should be? Link to comment Share on other sites More sharing options...
Techman Posted July 17, 2017 Share Posted July 17, 2017 Sometimes my setup (ISY994 and approx. 94 devices) exhibits some flaky responsiveness: commands are being missed, long delays, etc. So I thought this points to a PLM that is slowly failing. I did a PLM count and got consistently 58. Then I replaced the PLM with a spare one I had for a few years: 529 counts (consistently and after PLM restore). Afterwards, I tried a 3rd, seemingly new PLM and got 276 counts (again, consistently and after PLM restore). Any thoughts as to why I get so different results and how I can determine what the correct count should be? The PLM link count can vary depending on activity within the ISY. It's best to run your count when you know there is no or little ISY activity Link to comment Share on other sites More sharing options...
waffles Posted July 18, 2017 Author Share Posted July 18, 2017 Thanks Techman; Following your advise, recounted the links for the second and third PLM again. This time I got 290, instead of 529 for the second; and again 276 for the third one. Should this tell me anything? The count results seem to be rather random. Are they an indicator if a PLM works OK or is failing?? If not, how else can I determine if a PLM is OK? Link to comment Share on other sites More sharing options...
mwester Posted July 18, 2017 Share Posted July 18, 2017 Link counts are pretty much useless for anything. So in order: Should this tell you anything? Nope, nothing useful. Are they an indicator if a PLM works OK or is failing? Not really. That you get a count at all tells you that the PLM is working well enough that it can talk to the ISY, but outside of a lab environment, that's all it tells you. How else can I determine if a PLM is OK? That's a really tough question -- the only answer that makes engineering sense is to have a spare in hand that you compare with. There are simply no useful diagnostics in the PLM itself that the ISY can access. Regarding the link count -- the only way you get meaningful numbers is if you turn off the breakers to every insteon device that's wired in, unplug every insteon device that plugged in, and pull the batteries out of every insteon device that's battery-powered. The fundamental problem is that the PLM can't answer the question "How many links do you have" - instead, the ISY can only ask it to start sending all the links -- and the ISY counts them. The problem with that is that the PLM is so utterly stupid that any incoming message for any reason interrupts the sending of the links, and the PLM loses track of where it was. Just to ensure that this mis-feature is utterly infuriatingly frustrating is that it works just fine if you have very few devices -- few devices means sending the link table takes only a small amount of time, and also means that with few devices there's not much insteon traffic being sent. The result is that when you first try this you get faked into thinking it actually is a useful feature. Of course, when you need it is when you have lots of devices - and that's exactly when the probability of getting a useful count drops close to the likelihood of the Chicago Cubs winning the World Series (might happen - it's happened before - once)... Link to comment Share on other sites More sharing options...
stusviews Posted July 18, 2017 Share Posted July 18, 2017 An unusually low link count is often the sign of an impending PLM failure. A very high (>900) link count indicates that you are near the limit. Other than these two extremes, I agree that the count is meaningless. But agreement alone is not synonymous with truthfulness nor accuracy Link to comment Share on other sites More sharing options...
waffles Posted July 18, 2017 Author Share Posted July 18, 2017 Thanks all for your replies. I saw some older posts that UDI was working on releasing their own PKM version. Did this ever happen? Link to comment Share on other sites More sharing options...
mwester Posted July 18, 2017 Share Posted July 18, 2017 Thanks all for your replies. I saw some older posts that UDI was working on releasing their own PKM version. Did this ever happen? Nope. Rumor has it that several beta units exist. But, apparently SmartHome/SmartLabs decided not to sell UDI the chipsets they needed, so the project failed to see the light of day. There's no explanation for why that decision was made -- I expect UDI is bound by a non-disclosure. Personally, I suspect that SmartHome/SmartLabs was rather enjoying the revenue stream they gained by selling replacement PLMs every two years -- but take that with a grain of salt, because I'm known to be a bit cynical. (Sad, though, that reality is often closer to the predictions of a cynic than an optimist... ) Link to comment Share on other sites More sharing options...
Brian H Posted July 18, 2017 Share Posted July 18, 2017 The needed firmware chips {RF and Main Function firmware} did not appear. Maybe Richmond Capital Partners may change things. Now that they own SmartLabs/Smarthome . Though I only see a slim chance of that happening. I see the other side of the coin. SmartLabs/SmartHome may have made added profit by selling PLMs every few years, but how many users DUMPED Insteon because of this poor reputation. I probably would not be using Insteon if the UDI products where not being sold and I know how to rebuild the 2413 power supplies. Link to comment Share on other sites More sharing options...
apostolakisl Posted July 18, 2017 Share Posted July 18, 2017 The needed firmware chips {RF and Main Function firmware} did not appear. Maybe Richmond Capital Partners may change things. Now that they own SmartLabs/Smarthome . Though I only see a slim chance of that happening. I see the other side of the coin. SmartLabs/SmartHome may have made added profit by selling PLMs every few years, but how many users DUMPED Insteon because of this poor reputation. I probably would not be using Insteon if the UDI products where not being sold and I know how to rebuild the 2413 power supplies. I can't say that I would have dumped Insteon if it weren't for ISY, but I definitely would have ceased adding anything new from them a long time ago. In short, Smarthome has sold me about $1000 worth of stuff over the last few years that they would not have sold had I not had an ISY. As far as the link count, I assume if you took your ISY and PLM to location where no Insteon devices existed (like take it to your friends house or your office), plugged them in, and did a link count, you would get an accurate count in the complete absence of any traffic. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.