Announcement

**nightwalker** · June 3, 2010, 06:17 AM

It looks like you're using the "new" UPB plug-in from the updater. When HS made the change to fix the depreciated items all of the sudden there were lots of these entries in my logs. I have since gone back to the previous version until they get back to it.

**Oman** · June 6, 2010, 10:49 AM

Feature or not

There were many requests to have the plug-in set the device status to unknown if the device command could not be confirmed. This is what it now does. The problem itself has always been there (in that for whatever reason the device can not confirm the command back to the PIM). Now the plug-in reports it as an unknown device status.

Essentially you have poor communications from the device. The PIM would have always retried a few times because it could not get confirmation from your device and assuming your device heard one of these retries it would have ended up in the proper state.

Jon

**nightwalker** · June 6, 2010, 11:34 AM

Curious, after seeing the error , after updating to the newer plug-in, I went and did a communication test on the offending device and the test results were excellent, in the high 50's. After reading this post I went back and ran a complete network communication test as well as tests on individual devices. Again tests were regarded as excellent with most devices reading in the high 50's to a couple being in the high 70's. Comparing the results of several past communication tests things have not changed and seem rather consistent.

I'm not sure what to make of this, Either there is poor communication or there isn't. I can find no evidence of it and the only time I have ever gotten error messages from UPB was when I updated to the new plug-in. At the time RJ said nothing was changed to it beyond removing the depreciated items and he didn't understand why anyone would get those messages.

Seems there were other changes??

**Oman** · June 6, 2010, 12:27 PM

Go back to the old plug-in and turn on detailed logging. See if when you transmit a command to the device you still get a NOACK or missing ACK message. The ACK is a single pulse and not a full command. That is what the PIM is looking for to acknowledge that the device heard the command. Also how many transmissions per command do you have set? You must use at least 2 for the newer firmware devices to work correctly.

There were a few code updates between major releases, and this unkown on NOACK was added between them.

I am using the current version of the plug-in and the only time I get an unknown state in a device is when it is not working. How consistent is the unknown state? Is it always with a particular device?

Jon

Originally posted by nightwalker View Post

Curious, after seeing the error , after updating to the newer plug-in, I went and did a communication test on the offending device and the test results were excellent, in the high 50's. After reading this post I went back and ran a complete network communication test as well as tests on individual devices. Again tests were regarded as excellent with most devices reading in the high 50's to a couple being in the high 70's. Comparing the results of several past communication tests things have not changed and seem rather consistent.

I'm not sure what to make of this, Either there is poor communication or there isn't. I can find no evidence of it and the only time I have ever gotten error messages from UPB was when I updated to the new plug-in. At the time RJ said nothing was changed to it beyond removing the depreciated items and he didn't understand why anyone would get those messages.

Seems there were other changes??

**nightwalker** · June 6, 2010, 02:26 PM

I went back to the old plug-in a couple of days after the new one came out. I posted in that thread about the error messages. It hasn't given me any troubles, I do have it set for 2 transmissions. The only reason I updated was to get rid of the depreciated error message at start up. As far as I can tell I've never had an "unknown state", with the new version of the plug-in I tried I had messages about unknown states but the correct state was always displayed.

I turned on debug logging just for fun and operated a switch to see what I got. it was this. seems fine to me.

6/6/2010 3:09:41 PM UPB PIM-->Received UPB Message 08040100148664F5
6/6/2010 3:09:41 PM UPB Direct Pkt Rep Req: 0 (None) D Len: 8 Ack Req: 0 (None) Trns Cnt: 1 Seq Num: 0 Net ID: 1 Src ID: 20 Dest ID: 0 D Bytes: 1 Msg: !( 4)Core Reports-( 6)Device State! V 100
6/6/2010 3:09:41 PM UPB Processing Direct Message
6/6/2010 3:09:41 PM UPB * Setting HomeSeer device 19 Chnl 0 ID 20 Code ]31 Because of State Report
6/6/2010 3:09:41 PM UPB Setting status and value for ]31 to level: 100
6/6/2010 3:09:41 PM UPB In SetIO:]31 Cmd: 19 Bright: 100 Data1: 0 Data2: 0
6/6/2010 3:09:41 PM UPB Ignoring SetIO (Device Housecode) for ]31 Cmd: 19 Bright: 100 Data1: 0 Data2: 0 because the device table shows that it is being updated by the plug-in.
6/6/2010 3:09:41 PM UPB In SetIO:]31 Cmd: 2 Bright: 0 Data1: 0 Data2: 0
6/6/2010 3:09:41 PM UPB Ignoring SetIO (Device Housecode) for ]31 Cmd: 2 Bright: 0 Data1: 0 Data2: 0 because the device table shows that it is being updated by the plug-in.
6/6/2010 3:09:41 PM UPB Done Processing Direct Message
6/6/2010 3:09:42 PM UPB PIM-->Received UPB Message 08050100148664F4
6/6/2010 3:09:42 PM UPB Direct Pkt Rep Req: 0 (None) D Len: 8 Ack Req: 0 (None) Trns Cnt: 1 Seq Num: 1 Net ID: 1 Src ID: 20 Dest ID: 0 D Bytes: 1 Msg: !( 4)Core Reports-( 6)Device State! V 100
6/6/2010 3:09:42 PM UPB UPB Message is an in-sequence repeat of previous message processed.

Either way I'm not complaining, it's all worked without a hitch for years. I just hate errors in my logs and since RJ had no idea where the errors came from or why I moved back to the old plug-in and never gave it another thought until I saw someone else get the same error message.

Originally posted by Oman View Post

Go back to the old plug-in and turn on detailed logging. See if when you transmit a command to the device you still get a NOACK or missing ACK message. The ACK is a single pulse and not a full command. That is what the PIM is looking for to acknowledge that the device heard the command. Also how many transmissions per command do you have set? You must use at least 2 for the newer firmware devices to work correctly.

There were a few code updates between major releases, and this unkown on NOACK was added between them.

I am using the current version of the plug-in and the only time I get an unknown state in a device is when it is not working. How consistent is the unknown state? Is it always with a particular device?

Jon

**Oman** · June 6, 2010, 07:38 PM

Operating a switch won't show if ACKs work. An ACK is a single-pulse reply that a switch makes when you control it remotely. You have to control the switch directly via HomeSeer (not via a group).

Jon

Originally posted by nightwalker View Post

I went back to the old plug-in a couple of days after the new one came out. I posted in that thread about the error messages. It hasn't given me any troubles, I do have it set for 2 transmissions. The only reason I updated was to get rid of the depreciated error message at start up. As far as I can tell I've never had an "unknown state", with the new version of the plug-in I tried I had messages about unknown states but the correct state was always displayed.

I turned on debug logging just for fun and operated a switch to see what I got. it was this. seems fine to me.

6/6/2010 3:09:41 PM UPB PIM-->Received UPB Message 08040100148664F5
6/6/2010 3:09:41 PM UPB Direct Pkt Rep Req: 0 (None) D Len: 8 Ack Req: 0 (None) Trns Cnt: 1 Seq Num: 0 Net ID: 1 Src ID: 20 Dest ID: 0 D Bytes: 1 Msg: !( 4)Core Reports-( 6)Device State! V 100
6/6/2010 3:09:41 PM UPB Processing Direct Message
6/6/2010 3:09:41 PM UPB * Setting HomeSeer device 19 Chnl 0 ID 20 Code ]31 Because of State Report
6/6/2010 3:09:41 PM UPB Setting status and value for ]31 to level: 100
6/6/2010 3:09:41 PM UPB In SetIO:]31 Cmd: 19 Bright: 100 Data1: 0 Data2: 0
6/6/2010 3:09:41 PM UPB Ignoring SetIO (Device Housecode) for ]31 Cmd: 19 Bright: 100 Data1: 0 Data2: 0 because the device table shows that it is being updated by the plug-in.
6/6/2010 3:09:41 PM UPB In SetIO:]31 Cmd: 2 Bright: 0 Data1: 0 Data2: 0
6/6/2010 3:09:41 PM UPB Ignoring SetIO (Device Housecode) for ]31 Cmd: 2 Bright: 0 Data1: 0 Data2: 0 because the device table shows that it is being updated by the plug-in.
6/6/2010 3:09:41 PM UPB Done Processing Direct Message
6/6/2010 3:09:42 PM UPB PIM-->Received UPB Message 08050100148664F4
6/6/2010 3:09:42 PM UPB Direct Pkt Rep Req: 0 (None) D Len: 8 Ack Req: 0 (None) Trns Cnt: 1 Seq Num: 1 Net ID: 1 Src ID: 20 Dest ID: 0 D Bytes: 1 Msg: !( 4)Core Reports-( 6)Device State! V 100
6/6/2010 3:09:42 PM UPB UPB Message is an in-sequence repeat of previous message processed.

Either way I'm not complaining, it's all worked without a hitch for years. I just hate errors in my logs and since RJ had no idea where the errors came from or why I moved back to the old plug-in and never gave it another thought until I saw someone else get the same error message.

**nightwalker** · June 6, 2010, 08:47 PM

Right you are, i enclosed the on via manual as opposed to the off via remote. My mistake. Here's what you should have gotten, had I sent the correct one.

6/6/2010 3:32:00 PM UPB PIM<--Sending 0A140101FF226400015A to PIM (Try # 1)
6/6/2010 3:32:00 PM UPB PIM-->Received PIM Accept
6/6/2010 3:32:00 PM UPB Waiting for ACK/NAK from PIM
6/6/2010 3:32:01 PM UPB PIM-->Received PIM Ack

**Oman** · June 6, 2010, 09:31 PM

ACK is good. Now the question is: Is anyone getting ACKs with the old version but NAKs with with the new version? This should *not* be happening and would indicate a bug. My guess is that there were NAKs all along but they were hidden (and that is one reason the plug-in changed so that it did not ignore a NAK).

Now an option could be added to the plug-in to ignore the fact that the switch did not acknowledge the command but then there would be no indication of a failed device or communication problem.

Jon

Originally posted by nightwalker View Post

Right you are, i enclosed the on via manual as opposed to the off via remote. My mistake. Here's what you should have gotten, had I sent the correct one.

6/6/2010 3:32:00 PM UPB PIM<--Sending 0A140101FF226400015A to PIM (Try # 1)
6/6/2010 3:32:00 PM UPB PIM-->Received PIM Accept
6/6/2010 3:32:00 PM UPB Waiting for ACK/NAK from PIM
6/6/2010 3:32:01 PM UPB PIM-->Received PIM Ack

**nightwalker** · June 6, 2010, 11:33 PM

I can do another test run of the plug-in that's in the updater. Happily my post from my previous try is still intact, complete with logs. That may be enough to tell you what you need to know. If not I can re-install it and do whatever test you need.

The thread is here http://board.homeseer.com/showthread.php?t=139919

It's post #4

Originally posted by Oman View Post

ACK is good. Now the question is: Is anyone getting ACKs with the old version but NAKs with with the new version? This should *not* be happening and would indicate a bug. My guess is that there were NAKs all along but they were hidden (and that is one reason the plug-in changed so that it did not ignore a NAK).

Now an option could be added to the plug-in to ignore the fact that the switch did not acknowledge the command but then there would be no indication of a failed device or communication problem.

Jon

**joegrace13** · June 7, 2010, 10:51 AM

Well I'm at a loss to explain why HS now shows the correct status of the switch and as always the switch is controllable by HS yet I still get this line in my log when doing so.

6/6/2010 12:00:03 AM ~!~UPB~!~** Retry count exceeded, UPB Message transmission failed.

Dispite the fat that the switch does what it is asked to do.

The UPB plugin is listed as Status: Status=-1 on the interfaces page. Never used to be like that.

I attached the error log.

Attached Files

UPB Error Report.txt (107.6 KB, 2 views)

**nightwalker** · June 7, 2010, 12:17 PM

today I'm running the UPB from the updater and my logs look almost identical to yours. I can't explain all the messages and errors however operation does not seem to be affected, just a lot of annoying messages, especially this one which generates a log looking just like yours does. I already posted it on another thread and referenced it in message #10 of this thread.

Maybe Jon can explain or fix whatever the problem is but until then I'm going back to the previous plug-in where all was good.

Announcement

UPB plug-in error / showing device unknown

UPB plug-in error / showing device unknown

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment