johnnyt

Looks like the HomeSeer ISY plug-in is the cause of my performance problem with notifications, not the SD card. see http://forum.universal-devices.com/viewtopic.php?f=27&t=7939&p=60301#p60301 I'm still not clear why replacing my SD Card did improve my performance in some areas (with HS plug-in running) and the notification problem didn't occur with firmware 3.1.11 and before.

I think I've found the problem for system busy after a save. I disabled my HomeSeer ISY plug-in and the system did not go busy after a save. Now what? I guess I need to talk to the HS ISY plug-in developer but have you seen this before? Is there something I should tell him (or myself ) that would help me resolve this issue?

Got 2 strange errors when I saved a program tonight (the program saved okay so probably just a coincidence). I couldn't find them in the list of error messages on the wiki. First, while the system was busy after the save, the attached screenshot shows a message about and open failed for an error log file. See attached screenshot below. In the error log at around this time I got the following messages. Thu 2012/01/26 07:44:35 PM System -170001 [uDSockets] HTTP:33 error:6 Thu 2012/01/26 07:44:39 PM System -170001 [uDSockets] HTTP:33 error:6 Thu 2012/01/26 07:44:44 PM System -170001 [uDSockets] HTTP:33 error:6 and a little later: Thu 2012/01/26 07:52:24 PM System -170001 [uDSockets] HTTP:29 error:6 Thu 2012/01/26 07:52:29 PM System -170001 [uDSockets] HTTP:29 error:6 What do these things mean?

If not a garbage collection issue, any idea what would have caused my SD card to begin impacting performance the way it did without actually failing or showing any bad blocks? Sure, here it is: If $sLightOut is 1 And ( Time is Sunset - 30 minutes Or ( From Sunset - 1 hour and 30 minutes To Sunset - 29 minutes (same day) And Control 'MISC (Non Lighting) / Dusk and Low Bat Sensors / Front Door Dusk-Dawn' is switched On ) ) Then Send Notification to 'JT' content 'Dark Out' $sLightOut Init To 0 $sLightOut = 0 Else - No Actions - (To add one, press 'Action') I have about 4 other programs that would re-evaluate

Thanks for the replies. I will re-evaluate my need for persistence in my programs. Interesting. I think I may have evidence that there is a visible performance problem related to writing to nvram, at least in more recent firmware releases. I've found that after I moved to 3.1.16 from 3.1.11 I needed to add a "wait" after a "send notification" command and before updating a state variable and saving it. Otherwise I would get the standard notification instead of the custom notification I was supposed to get. I was told it's because the system is busy but it happens every time without fail so the busy condition seems to be writing and saving a state variable. See http://forum.universal-devices.com/viewtopic.php?f=25&t=7693&p=58831#p58831

Thanks for the great insights. Sent from my iPod touch using Tapatalk

Have read a couple of posts that suggest or seem to suggest avoiding writing to non volatile memory. I have several programs I use as variables, many created before variables were available and many simply because I got used to that paradigm. In the then clause I "enable run at startup" and in the else clause I "disable run at startup" to preserve the state through a reboot. If I was using variables I would use the "init to" command to do the same thing, which also writes to nvram I believe. Am aware of some of the drawbacks with solid state disk and ran into what I now think was a performance problem due to perhaps a lack of garbage collection on it (didn't really get an answer when I posted on that but it might be because it was near the bottom of a long post. see http://forum.universal-devices.com/viewtopic.php?f=7&t=7768&p=60001&hilit=garbage+collection#p60001) What are some recommendations when it comes to this?

Setting a variable does seem like a cleaner/better idea than disabling the program and re-enabling it later but I end up in the same boat as before. I still need to reset the variable to 0 at an appropriate time later, with "appropriate" being several conditions, including time period, status change like nighttime variable change and "switched on or switched fast on", perhaps others. Since it isn't clear to me anymore if/when I can include "control" and "status" in the same program, this will just add one more program. Does the order of the execution matter with conditions, i.e. does it make a difference if I put "control" conditions first or last? Could this be because my insteon network is less than ideal but not bad enough that things aren't visibly not working for me so I don't realize it?

Oops. I meant "true=open" Sent from my iPod touch using Tapatalk

The then above should read "turn garage light on" The "2 garage doors open" is really just a check of a program variable "false=open" Sent from my iPod touch using Tapatalk

Thanks for the replies. A couple of clarifications: 1. am running firmware 3.1.17 2. I want this program (and others just like this one) to automatically turn a light on if it's off (garage light in this case) when something is true (in this case when either of my two outside garage doors are open at night) except if I manually turn the light off (Off or Fast Off). Benoit, you bring up a great separate conversation I'd like to have about saving (or not) to nvram. Have read a couple of posts now about avoiding that where one can and I'd like to hear more about that because I do a lot of writing to nvram and have recently encountered speed problems that were (in part) do to the SD card after only a little over one year of use. I'll start a new thread at some point for that, unless someone knows of one (I haven't done any searching for that yet.)

I don't quite get the "control not switched off" logic when I use it as part of multiple conditions. I assumed (wished) it would trump everything but it doesn't. Here's a program that I don't want to run true if I switch the garage light off but it runs true anyway. Both "program" variables are True when this happens, so the program certainly does what it's supposed to when I don't touch the garage light, which is to turn the garage light on when it's night time and the outside garage door opens (is open). If Status 'Main Floor / Garage Light' is not On And Control 'Main Floor / Garage Light' is not switched Off And Control 'Main Floor / Garage Light' is not switched Fast Off And Program 'Nightime Flag' is True And Program 'Garage Door Outside Open Flag' is True Then Set 'Main Floor / Garage Light' On Else - No Actions - (To add one, press 'Action') To do what I want in other scenarios I've found myself having to write several programs with the one for the "control not switched off" disabling the one that runs the command then have another program run later to re-enable it. The problem with that has been all the fiddling needed to cover all the conditions needed for re-enabling the program. Am assuming there's a clean one-program way to do this and that seeing it will provide me insight into the way things are evaluated when there's a "control is not switched off". Any advice would be appreciated.

Hi Michel, Performance of the restore is significantly better. It went from taking 8 1/2 minutes on the old SD card (with the failure mentioned above about 1:20 into the process) to about 3 1/2 minutes (no failure). Am only one day into the new card but have seen a marginal improvement in the best backup time and no ~5 min backups as would happen occasionally before. In other areas performance improvements varies from marginal (loading programs) to fair (initial startup). It didn't fix the need for me to add a "wait 2 secs" when I have a program that sends a notification then updates a state variable (http://forum.universal-devices.com/viewtopic.php?f=25&t=7693&p=59009#p59009) I hope to post with more info to my other thread after I've done more testing.

upgraded to 3.1.16 and the problem happened again. However, I followed my test with the replacement of my SD card (to solve a different problem) and installed 3.1.17. I don't know whether it was something to do with the (old) SD card or something in 3.1.17 but things are fine now.

Hi marksidell, I have this problem too. Affects both my XP and W7 machines (LAN machines). I don't access ISY remotely so I don't know if it occurs after a while that way but it didn't occur when UDI tried connecting remotely a few times. (like bringing your car to the mechanic and everything works fine when he tries it ) The only way I am able to avoid it is to use the ISY Finder via the link http://www.universal-devices.com/99i//admin.jnlp. I tried several things. The details are at http://forum.universal-devices.com/viewtopic.php?t=5245 If you find anything that fixes it for you, please post back. Thanks.

I have 2 bathrooms in which I use KPL's to control the fan and am very happy with them. In one bathroom I use the KPL to also control the light. An inline relay replaces the bathroom fan switch but you could leave/use the switch. I just worried people would gravitate to it first to turn the light on instead of looking for the KPL button that says "Light ON". I use one scene with all the needed secondary buttons (e.g. C-On, D-15 mins, E-30 mins) as controllers and the fan relay as a responder (no choice on that front). Each button in the scene is set to turn the relay to 100% and the other buttons to 0%, which will turn them off. (Note that this approach doesn't work to turn off a secondary button on another KPL - to do that you would need to have a program send an OFF to a scene with the other KPL's secondary button in it. This is not a problem here, am just mentioning it to maybe save you some time figuring that out later) Then I have programs that do the following: 1) if KPL.C (Fan On) status is ON, wait 45 mins then turn the scene off (this is the max I let the fan stay on) 2) if KPL.D (15 mins) status is ON, wait 15 mins then turn the scene off 3) if KPL.e (30 mins) status is ON, wait 30 mins then turn the scene off By having mutual exclusion and checking the buttons for "status" ON you would never have more than one program running at a time. In the second bathroom, I use the KPL load to control the fan and a regular switch for the light. This is idiot proof for lighting but I also don't have any lighting scenes in this bathroom like I do in the other one. I might add that for both bathrooms I use humidity sensors (via a HomeSeer based solution right now) to turn the fans on automatically when someone takes a shower (hum delta > 15% with rest of house) so I also have a "disable" button to keep that from happening when desired (i.e. when someone wants to take a long bath). That button is also in my scene mutually exclusive of the others, and the scene is turned off an hour later in that case. I found the KPL cost effective because the button labeling adds clarity and I use the buttons I don't need for the fan or light to do other things and/or let me know about other things going on in the house. In my case, in bathroom 2 (master bathroom) I can turn the furnace fan on high (or know that something did, since I have programs that will do that) and I can turn all the lights downstairs off or on (each it's own button). I also have a button that tells me if ISY and HomeSeer are no longer communicating (ON is bad).

Am sure that's what I did the 1st time. Either the routine quit on me or I didn't wait long enough for it to finish. I just found myself looking at a summary tab with a bunch of programs that said " - not saved -" and a system that was "Ready" (none of the usual "do not change or unplug attached devices...") Maybe there are pauses in the process? Will try again when next weekend. Sent from my iPod touch using Tapatalk

have had to do a bunch of restores tonight and each time I get an error during the "clearing old configuration" step. something about java null exception. Each time I simply restart the restore and it moves past that and completes no problem. I did try to search for this before posting but all the words I used were too common... no results am running 3.1.11

factory reset did it, although the load didn't turn off right away - I needed to turn it on first, then off. thanks for that tip. is this a fluke or is my KPL on it's way out? Also I still need to be able to replace one of these things. can you confirm the steps I have to follow?

so the wiki instructions are wrong? do I just need to link the new KPL and replace old one with new one? am sure I tried that and got a whole bunch of < Not Specified >, which is when I went for the instructions... will try again (am in the middle of another restore right now - tried the reverse replacement, i.e. new with old instead of old with new, which didn't work either this is painfully slow. started this process about 3 hours ago. it shouldn't be the complicated...

I have a KPL relay where the load is stuck ON (even when I air gap the thing - is that possible? it doesn't seem right - how do I turn the light off??) I linked a new one in (spare I kept in inventory of the same type and version -> a 6 button KPL relay v36 converted to 8 button operation), did a find/replace on all the programs, saved programs, then did a right click on bad device to "replace with" good device. New device names disappear from my network (replaced with old device names, presumably) but all my programs end up with where the new device should now be. Have done this three times now. Have re-read the wiki instructions each time to make sure I'm taking all the right steps in the right order. What am I doing wrong? Am running firmware 3.1.11.

Was debugging something and noticed programs were executing almost 30 secs ahead of the clock in the GUI. The "last run time" would update to the expected new run time, e.g. 5:30:00 PM but the clock at the top left of the console was at about 5:29:34 or something like that when the "last run time" was updated. I tried a manual time sync and it didn't change the time, and there's no indication the automatic syncs aren't working (every 6 hours in my case). I did notice the GUI is a few seconds ahead of my PC clock. In this case the GUI had been running about 30 minutes or so when I noticed this, so if it gets it's time from either the PC or ISY at startup, it certainly loses track of it quickly. Any idea what's going on? Edit: am running ISY 3.1.11, Java 1.6 update 29, W7

I use Midon Design pre-built kits. I have ten "MD3020E" temp/humidity sensors, some in custom enclosures, others hidden in cold air returns. See mid way down the page @ http://www.midondesign.com/Sensors/sensors.html All are read by the "Temp08", which reports the data out an RS232 serial port http://www.midondesign.com/TEMP08/TEMP08.html The Temp08 also supports other sensors and relay devices. From the manual at http://www.midondesign.com/Documents/TEMP08UserGuide.PDF: · Multiple DS2438-based Humidity Sensors, including, of course, the Midon Design MD3020E Humidity Sensor. · Multiple DS2438 based Barometric Sensors · Multiple DS2438 general purpose sensors for analog voltage input · Multiple DS18S20, DS18S20-PAR, DS18B20, DS1822, DS1920 temperature sensors as well as temperature reading from any DS2438 · One DS2450 based Weather Station for wind speed and wind direction · Multiple DS2450 voltage sensors · Multiple DS2423 based rain gauges · Multiple DS2423 general purpose counters (for use with Lightning sensors and other types of counter inputs) · Multiple DS2401 or DS1990 1-Wire serial numbers (only presence will be reported) · Up to 20 1WIO relay interface modules (or equivalent DS2408 based relay module) available from Midon Design · Up to 10 1WIO LED modules (available from Midon Design) · Up to 20 1WIO Input modules (available from Midon Design)

What I'd really like to see is energy (pardon the pun) spent on support for a (wired) one-wire temperature and humidity sensor solution. I think temp/hum (and thermostats, actually) should be kept off the insteon network, especially if each has to be polled every minute, times # of sensors (& stats). Teken, I currently use a HomeSeer and one-wire based sol'n (that I would like to move to ISY but can't right now) for what you'd like to do. I routinely turn on house ventilation to balance out the temperature in the house. However, unless one lives in a particularly dry region, the biggest bang for the buck IMO is with respect to managing humidity in the house. I'm able to turn on the bathroom fans when the humidity hits ~12% higher in one compared with the humidity on the main floor (far from bathrooms, kitchen, laundry room) then off again when it comes back down. In winter I turn on/off the furnace humidifier and HRV based on indoor humidity levels and the outside temperature. In summer I also control the HRV around temp and humidity outside, turn on/off a dehumidifier, and lower the AC setpoint when the humidity in the house is high but the temp is not quite high enough to otherwise trigger the AC. I have to say the automagic bathroom fans have been the most visible automation tricks in my bag next to motion/scene lighting. While people don't see much of the stuff going on, I consider humidity control in general to have provided the highest ROI since, unlike much of the other automation stuff (that I like too), it contributes directly to the house's and my family's health.

Hi, I'm looking for advice on the value of converting from using physical COM ports and RS232-to-USB converters to a RS232-to-IP serial server. I ask in this forum because, one, people here are brilliant , and second because my medium to long term plan is to move everything to ISY from HomeSeer, possibly replace my DSC alarm with ELK, and add other ISY supported devices for energy metering, etc. Am a bit constrained in the short (and possibly medium) term waiting for a good wired thermostat sol'n and a 1-wire temp & humidity sol'n but am hoping the medium to long term promises good things in this area. How much are serial devices a part of ISY controlled solutions, either directly supported or through 3rd party products/utilities? The HomeSeer forum has a number of folks who seem to swear by serial servers, particularly for virtualization, which I plan to go to when I refresh my HA server imminently (for HomeSeer, DSCLink, shared files, Windows Home Server for backups, etc.) See http://tinkertry.com/vzilla/, although mine will be a scaled back version. I'm not sure if a serial server is a must have or a nice to have (and how nice) for ISY and other reasons. FWIW, I currently have 5 serial devices, which would mean >$400 for a good 8-port device (Quatech brand seems to come highly rated, with people saying you get what you pay for so don't cheap out on a device). People also keep saying they get stuff like this for less than $50 on ebay - always seems everyone finds what they want for $50 on ebay in these forums It's certainly not been my experience. My RS232 devices include APC UPS with smart signalling (a special cable), Midon Design's Temp08 (a 1-wire temp/hum sensor reader), RCS TR40 thermostat, W800 X10 RF receiver, and DSC Alarm panel interface (PC5402). For the DSC panel I'm currently using a software solution with DSClink, described here http://forum.universal-devices.com/viewtopic.php?f=51&t=6921. I could keep using that and only buy a 4-port SS but would have no room to grow. Plus I'm wondering if the odd (daily) missed messages from the DSC panel are caused by the SS s/w solution or the RS232-to-usb converter I'm using to get the messages to it. Either way, I see some value in standardizing to a single h/w SS solution and having room to grow if it's a worthwhile expense. Is it? Any advice would be appreciated.