| Author |
|
grif091
Super User
Joined: March 26 2008
Location: United States
Online Status: Offline
Posts: 1357
|
| Posted: November 30 2008 at 13:21 | IP Logged
|
|
|
I have the same problem running 2.1. Besides the performance problem, when enough Group Cleanup Directs have been repeated the device Clear Failed column is shaded red and PH stops communicating with the device.
I emailed support in case the forum was not being watched closely.
Edited by grif091 - November 30 2008 at 13:22
__________________
Lee G
|
| Back to Top |
|
| |
nadler
Super User
Joined: February 25 2006
Location: United States
Online Status: Offline
Posts: 354
|
| Posted: November 30 2008 at 13:43 | IP Logged
|
|
|
Lee
When the clear failed column is shaded red it means the
Insteon device is not communicating with PLC/PLM. This
is a feature of PH, not a bug. It lets you know there
is a communication problem. And it PH doesn't keep
trying when it knows it won't get through. If you
haven't tried this already, turn polling off by
unchecking the status scan box and increase the max
number of failures above the default value of 5. This
doesn't correct the problem, it just means there will be
less traffic on the power line and fewer communication
failures.
My experience after many years of X-10 and Insteon is
that there is always some powered device in my home
causing problems when a box turns red in PH. It's
usually a new power strip or a computer fan gone bad, or
a new refrigerator, or a vacuum cleaner was turned on or
something similar. Once the problem in the line is
eliminated no more 'red'. I also noticed when I
switched from a PLC to a PLM I had increased
communication issues which I was able to track down to
my computer's power supply. To correct that, I just
switched the computer to a different outlet then my PLM
and the problem went away.
Noel
|
| Back to Top |
|
| |
Viper
Groupie
Joined: January 14 2007
Location: United States
Online Status: Offline
Posts: 88
|
| Posted: November 30 2008 at 13:54 | IP Logged
|
|
|
Lee,
Ouch! You are right, I hadn't noticed that. So this is a bit more serious than I first thought.
I just up'ed my Max. Failures to 25 and implemented a trigger at 00:10:00 everyday to ph_insteonclearfailed("") until the problem is fixed.
Thanks for pointing this out.
Tom
|
| Back to Top |
|
| |
grif091
Super User
Joined: March 26 2008
Location: United States
Online Status: Offline
Posts: 1357
|
| Posted: November 30 2008 at 13:57 | IP Logged
|
|
|
nadler, thanks and you are quite correct about the red shading normally being a communication problem. However, in this case my post was a follow up to the one just before mine where PowerHome is retrying a Group Cleanup Direct command that was successfully issued and ACKed, not NAKed. I posted the log file in the email to support, I should have included it in my post as it shows the repeated Group Cleanup Direct commands being issued even though each command is ACKed. Also in my case the Group Cleanup Direct is energizing a relay which is actually happening as one would expect when the device is ACKing the command. The post prior to mine was documenting it as a performance problem as the PLM Group had several devices, each of which worked but had the Group Cleanup Direct repeated three times for each device. I mentioned the red shading as an additional symptom beyond the performance consideration. Appreciate your post. Under most circumstances it is a device communication problem.
EDIT: viper, that is why I sent an email to PowerHome support. I figured a performance issue was one thing but when PH stops communicating with a device(s) over this it has a different severity. That is nice workaround.
Edited by grif091 - November 30 2008 at 14:02
__________________
Lee G
|
| Back to Top |
|
| |
Viper
Groupie
Joined: January 14 2007
Location: United States
Online Status: Offline
Posts: 88
|
| Posted: November 30 2008 at 14:06 | IP Logged
|
|
|
Noel,
Lee is right, that this is a new problem. Every one of the devices that were controlled via Insteon Group commands from PH2 was showing failed counts and one had hit the max failed limit. Normally, the only time I see failures is when I've had a device go bad. I had no failure counts with PH 2.0.0.
Tom
|
| Back to Top |
|
| |
nadler
Super User
Joined: February 25 2006
Location: United States
Online Status: Offline
Posts: 354
|
| Posted: November 30 2008 at 14:23 | IP Logged
|
|
|
Lee, I didn't catch the connection. I'm sure Dave will
chime in with his thoughts.
|
| Back to Top |
|
| |
BeachBum
Super User
Joined: April 11 2007
Location: United States
Online Status: Offline
Posts: 1880
|
| Posted: November 30 2008 at 14:40 | IP Logged
|
|
|
Another wrinkle, the motion detector is having problems with shaded red also. I’m get NAKs from what appears to be an OFF sent although it’s not coming from my macros. I do have jumper 4 on (ONs only sent no OFFs).
2008-11-30 13:12:49.578 RX RECEIVEINSTEONRAW=0A 3D B3 00 00 01 C7 11 01
2008-11-30 13:12:49.935 RX RECEIVEINSTEONRAW=0A 3D B3 00 00 01 C7 11 01
2008-11-30 13:12:52.837 RX RECEIVEINSTEONRAW=0A 3D B3 0D 4F EE 41 11 01
2008-11-30 13:13:05.305 TX 02 62 0A 3D B3 0F 13 00
2008-11-30 13:13:05.321 RX SENTINSTEON=0D 4F EE 0A 3D B3 0F 13 00 06
2008-11-30 13:13:09.328 RX INSTEON TIMEOUT=0A 3D B3
2008-11-30 13:13:10.306 TX 02 62 0A 3D B3 0F 13 00
2008-11-30 13:13:10.320 RX SENTINSTEON=0D 4F EE 0A 3D B3 0F 13 00 06
2008-11-30 13:13:14.322 RX INSTEON TIMEOUT=0A 3D B3
2008-11-30 13:13:15.307 TX 02 62 0A 3D B3 0F 13 00
2008-11-30 13:13:15.320 RX SENTINSTEON=0D 4F EE 0A 3D B3 0F 13 00 06
2008-11-30 13:13:19.328 RX INSTEON TIMEOUT=0A 3D B3
Also when I link the PLM as a Responder I have to declare a button number. If I used RemortLinc I didn’t.
Should we start a new thread for 2.1?
__________________
Pete - X10 Oldie
|
| Back to Top |
|
| |
grif091
Super User
Joined: March 26 2008
Location: United States
Online Status: Offline
Posts: 1357
|
| Posted: November 30 2008 at 15:38 | IP Logged
|
|
|
Pete, new thread with some specific symptom would not be a bad idea. There is a Group Broadcast, Group Cleanup Direct with an ON command from 0A.3D.B3 Group 1 (motion sensor motion ?). In response PH issues a Direct OFF command to 0A.3D.B3 which, if that is the motion sensor, would TIMEOUT as the RF receiver is sleeping.
Does the Event Log where detailed information is logged about messages received, triggers fired, macros invoked etc show anything around the 13:12:52 time frame when the Group Cleanup Direct was received. There is a 12.5 second time interval between when the Group Cleanup Direct is received and when PH sent the Direct OFF command. If the Direct OFF that PH sent was initiated from when the Group Broadcast was received, there is approx 15 seconds between it and the Direct OFF command.
__________________
Lee G
|
| Back to Top |
|
| |
BeachBum
Super User
Joined: April 11 2007
Location: United States
Online Status: Offline
Posts: 1880
|
| Posted: November 30 2008 at 16:01 | IP Logged
|
|
|
Here’s the capture..
__________________
Pete - X10 Oldie
|
| Back to Top |
|
| |
grif091
Super User
Joined: March 26 2008
Location: United States
Online Status: Offline
Posts: 1357
|
| Posted: November 30 2008 at 16:23 | IP Logged
|
|
|
The event at 13:13:05 (Insteon Out) sent the Direct OFF to MOTION_SENSOR_1. There are three triggers that were fired, INSTEON_CHANGE, MOTION_SENSOR_1, and INSTEON_CHANGE a second time. Plus Macro HALLWAY_ALERT was invoked. Something in there directed PH to send the Direct OFF back to the motion sensor. Is it possible for the device ID for the Direct OFF command to come from a variable that was set to the MOTION_SENSOR_1 by mistake, either in one of the triggers or macro logic? Or maybe something PH sets as the result of an event which you then change to the device you want to use but PH set it back because of the sequence of in-bound messages. Just throwing out ideas without the benefit of seeing the triggers and macro.
__________________
Lee G
|
| Back to Top |
|
| |
BeachBum
Super User
Joined: April 11 2007
Location: United States
Online Status: Offline
Posts: 1880
|
| Posted: November 30 2008 at 16:52 | IP Logged
|
|
|
The Insteon Change is from Insteon Device Change Any, Any, Any. It is used for my error recovery routines. The gates are used to control the amount of traffic triggers release if allowed. Hallway Alert is from the firing of the Motion Sensor 1 with Group In, 1 On and is gated from firing again on the second Group In. Hall Alert is nothing but a TTS and changes the global for the gate. Your right there was an Insteon Ioff issued by that macro ( ph_devicebtn(5,””motion_sensor_1”,ioff,”0”) It was intended for the purpose of setting the Motion Sensor to OFF in the Device Screen. My BAD!!!  Thanks Lee as usual. 
__________________
Pete - X10 Oldie
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: November 30 2008 at 21:08 | IP Logged
|
|
|
The group cleanup commands not being properly ACK'ed is indeed a bug. The new Insteon code now includes automatic retries (up to 3) for all direct commands. The group cleanup command code does not properly interpret the ACK so the command is always retried up to 3 times resulting in a slowdown of about 15 seconds.
Another bug has also come to light. Version 2.1 had a bad compile and the COM port activeX control for the CM11A did not build properly resulting in NO CM11A support for version 2.1.
Ive already gone in and coded the fixes for both of these issues and am ready to roll out release version 2.1a. Before I do though, I'd like to hear of any other bugs that I can add quick fixes for. If I don't hear of any others, I'll roll out 2.1a tomorrow night.
Thanks,
Dave.
|
| Back to Top |
|
| |
crisx
Groupie
Joined: September 14 2006
Location: United States
Online Status: Offline
Posts: 72
|
| Posted: December 01 2008 at 00:45 | IP Logged
|
|
|
Super insignificant, but the EZRain device is now called EZFlora Irrigation Controller.
Simplehomenet Forum Post
|
| Back to Top |
|
| |
TonyNo
Moderator Group
Joined: December 05 2001
Location: United States
Online Status: Offline
Posts: 2889
|
| Posted: December 01 2008 at 10:32 | IP Logged
|
|
|
I'm seeing some strangeness here. Trying to narrow it down from the W800, virtual X10 devices, or another partial-macro execution thing.
When I left the house today, the new DS10-type device on my garage door was in the log (there was also a new byte, 196), but the virtual X10 device was not updated via 'X-10 Raw'.
Also, in another macro, my ph_sendsmtpemail did not fire.
The new DS10-type device is a part that Radio Shack had for a short time that was smaller, used AAA's, and could handle two contacts.
|
| Back to Top |
|
| |
mustangcoupe
Super User
Joined: August 08 2003
Location: United States
Online Status: Offline
Posts: 308
|
| Posted: December 01 2008 at 16:06 | IP Logged
|
|
|
Dave,
I am still having problems with the PHservice and not showing up in the task bar. Even with this release. Any ideas? If needed I can let you remote in to try and help out with an instance where it doesn't work.
__________________
Todd Hannemann
HA newbie 
 PowerHome - UPB - ELK
|
| Back to Top |
|
| |
TonyNo
Moderator Group
Joined: December 05 2001
Location: United States
Online Status: Offline
Posts: 2889
|
| Posted: December 01 2008 at 21:02 | IP Logged
|
|
|
Update: My code was hosed on the wireless device. No issues with W800 or virtual devices.
A partial-macro execution thing is still in the wings.
|
| Back to Top |
|
| |
TonyNo
Moderator Group
Joined: December 05 2001
Location: United States
Online Status: Offline
Posts: 2889
|
| Posted: December 01 2008 at 21:26 | IP Logged
|
|
|
Also, I found that the 196 comes from a Tamper bit (0x40).
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: December 01 2008 at 22:51 | IP Logged
|
|
|
Crisx,
I'll update the EZRain description in the next version so I don't have to have a new database Upgrade Util  . Also, any user who desires to can update the Description in the Insteon Types screen. It won't affect anything to do so.
Todd,
I was able to reproduce your situation by creating multiple users on my PowerHome machine and RDP'ing as a different user from the currently logged console user. Unfortunately, I have not found a way to "fix" the problem. I'll keep playing, but this will most likely need to wait until I separate the backend processing from the UI so that the service part of PowerHome has no UI and the UI is a separate program that can be launched and connect to the service.
Tony,
Glad to hear it wasnt a 2.1 issue (so far)  .
Dave.
|
| Back to Top |
|
| |
Viper
Groupie
Joined: January 14 2007
Location: United States
Online Status: Offline
Posts: 88
|
| Posted: December 02 2008 at 09:43 | IP Logged
|
|
|
Dave, RE: Group Cleanups and 2.1a
Thanks for the fast turn around. The macro that used to take 2 minutes and 42 seconds, now takes 6 seconds. Nearly all the time was spent waiting for the timeouts to occur.
Nice job!
I also figured out the install oddity I had. When an uninstall is done, the pwrhome.ini file gets modified and those two lines I mentioned before are removed. I kept images of the files before the uninstall this time and did a quick comparison with the files after the install. That's what I get for not following directions....
Tom
|
| Back to Top |
|
| |
Active Topics 
Memberlist 
Search 
Help
Register 
Login
PowerHome Messageboard : PowerHome General
Topic: Release 2.1 available
