New Firmware 3.10A old comm/time problems back

[pete_c]

I've left the network disconnected (real PITA) for the last few days and the time has remained in sync and all of the automation events are running fine.
 
This is a mickey mouse solution to my issues and do not make sense to me at this time.  Priority though is sort of functioning panel first.

 

[ano]

HAIku Helper, WebLink 3, and a few other programs offer the option to update the Omni's time clock periodically. Are you sure one of these programs isn't going wacko and updating the time wrong?  

 

[pete_c]

Only the Omnitouch 5.7e's are connected 24/7.   I will disconnect them for my next test when reconnecting the network cable.
 
Afterwards will test connecting to a network switch that isn't connected to the rest of the network just to see what happens.
 
I did test some with Snaplink and did see the thermostat comm error come up concurrent with an online offline connection meaning to me that the network connectivity went south. 
 
This made me time sync once and remove the network connection afterwards.  Most important is that the events continue to run in a timely fashion related to a correct time.
 
The time sync issue is major as its typically 20 minutes off by the hour when it happens.  I am particular about time and use an internal server for time sync (which works way better than the internet).
 
The network errors after a bit of time though are really bad with this new version of firmware 3.11C.  
 
Historically I would disconnect the network for about a day or so; reconnect it and I would be fine for literally months.  This is the first time I have seen it go back to bad network connectivity, way off time sync and comm issues literally hours after reconnecting the network interface.
 
I have configured 6 cameras via Video hub about 2-3 months ago such that I can see them on the legacy Omnitouch screens.  I just noticed too that I configured 10 IP cameras via PCA and wondering if that is my issue.  I will remove the 10 IP cameras and see if that does anything today.

 

[pete_c]

Removing the IP cameras from the configuration helped a bit. 
 
This time I got a 2 day run time before I saw the disconnecting network, wrong time and serial comm link thermostat issue.
 
I am now considering maybe downgrading the firmware to the B version as it seemed to be a bit more stable for me.
 
This will be the first time I have downgraded the HAI OPII firmware since installation of this newer panel.
 
With said firmware I could disconnect the network for a day and reconnect and all would be fine for longer than two days.

 

[pete_c]

I tried the disconnect of the LAN and then reconnect after a day or so.  It worked again for about 2 days. 
 
Downgrading today to version B to see if the same occurs.

 

[ano]

If you don't care about Zigbee, I'd go back to 3.10D It was an oldie but a goodie.

 

[pete_c]

Thanks Ano.
 
I was actually wanting to test Zigbee; but now prefer a functioning HAI OPII panel (with network connectivity).  That said yesterday I downgraded to 3.11B. 
 
I am not sure though when the IP camera stuff was added as I only recently had added the IP cameras in the PCA configuration. 
 
I did notice that I had to look at my PCA setups to find the 3.11B firmware and could not find it on the HAI site.
 
I looked for FW 3.10D and cannot find it. (found everything almost but 3.10D)
 

04/21/2010  03:25 PM         2,097,268 HAI_OmniPro_II_Version_3.4_(English).cff
03/01/2011  03:04 PM         2,097,268 HAI_OmniPro_II_Version_3.7_(English).cff
09/05/2011  12:42 AM         2,097,268 HAI_OmniPro_II_Version_3.8a_(English).cff
04/16/2009  11:34 PM         2,097,268 HAI_OmniPro_II_Version_2.16a_(English).cff
04/16/2009  11:34 PM         2,097,268 HAI_OmniPro_II_Version_3.0a_(English).cff
04/21/2012  10:13 PM         2,097,268 HAI_OmniPro_II_Version_3.10a_(English).cff
07/15/2012  10:03 PM         2,097,268 HAI_OmniPro_II_Version_3.10b_(English).cff
11/18/2012  05:35 PM         2,097,268 HAI_OmniPro_II_Version_3.11_(English).cff
11/20/2009  06:31 PM         2,097,268 HAI_OmniPro_II_Version_3.2b_(English).cff
03/06/2013  12:07 AM         2,097,268 HAI_OmniPro_II_Version_3.11b_(English).cff
03/30/2013  09:52 PM         2,097,268 HAI_OmniPro_II_Version_3.11c_(English).cff
11/07/2010  06:16 AM         2,097,268 HAI_OmniPro_II_Version_3.6a_(English).cff
11/24/2010  09:59 PM         2,097,268 HAI_OmniPro_II_Version_3.6b_(English).cff

 
Can you attach a link to said firmware?

 

[ano]

Unfortunately I only have it in PROM form because at that time I didn't have the board that could upgrade the FW.  Hopefully someone here can send it to you, and I bet HAI would send it to you if you asked. 

 

[pete_c]

Thanks Ano.
 
Decided to try the downgrade in "baby" steps first.  Today just downgraded to 3.11B for now. 
 
Interesting though my thermostat ambient temperature is at 103 F.
 
I did a warm reset of the panel and that didn't work so I guess I will need to do a reset of the Omnistat2 instead.
 
A factory reset of the Omnistat2 brought the ambient temperature back to normal.

 

[pete_c]

It seems that my network, comm and time sync issue went away with the "B" version of the firmware.  I left the network disconnected for 2 days and reconnected it and its been fine now for at least 2 days.  This time I left the Omnitouch 5.7e's disconnected (pulling at straws now). 
 
Only network connection is Snaplink and coreserver for the time being. (and spot checking letting run for only a few hours at a time on my home office desk little tablet).

 

[pete_c]

The problem has returned with version "B". 
 
The first time I saw this issue was when I upgraded my Omnistat RC-80 to the Omnistat2 a couple of years back. 
 
That said tomorrow will be downgrading thermostat back to the RC-80. 
 
Will run it a bit to see how it behaves.
 
That said will request Omni Pro II firmware 10D (before Zigbee). 
 
I did find some diagnostic stuff relating to my communications issues to the thermostat.
 

1) Look at the Pro II board. On the lower left above the top terminal strip, you will see the blinking D24 status led. Approx ¾” to the right of the status led you will see a small silver switch with a white button. Just above that switch is an output fault led. Is it lit?

2) If it is let, press and release the small switch. Did the led go off? If so communication should be restored.

3) If the led did not go off, power down the controller and disconnect the wires from output 8.  Power up the controller, press the reset. Is the led off? If so, there is a problem either with a thermostat or with the wiring. If not, the controller is defective and needs repair.

4) If the output fault led was never on, power down the controller and disconnect any wires from output 8. Power up the controller and check the voltage from output 8 to ground. There should be a pulsing voltage going from 0VDC to approx 8 to 12VDC every 1 to 1.5 sec. If the voltage is missing and the controller setup is correct, the controller is defective and needs repair.
 

 

[ano]

All the consoles, Omnistats, wireless receivers, and a few other things commiunicate on shared buses.  If ANY device on that bus is defective, It will screw up all the other devices.  Remove one external device at a time, and see if that fixes it.  Omnistats are a good place to start but it could be anything on the bus.  If it still happens, its probably a bad board. I think you ruled out a software but by running with the programming cleared.
 
You don't need to replace a Omnistat2 to test if its a problem.  Just pull off the three wire HAI connector. It will still operate as a plain old thermostat, but if that is the problem, the panel should start working again. Same with consoles. Just disconnect the connector.  (But leave at least one to turn off your alarm, of course.)
 
Although it seems to work better without the Ethernet port, from what I have seen, what appears to be an Ethernet port problem on the panel, is really another problem in disguise. The Ethernet port is just a symptom of the other problem. 
 
3.10D seemed to be the most stable to me because it is before major Zigbee changes, but having said that 3.11C seems to work fine too. Some versions between those two were a bit iffy.

 

[pete_c]

Thanks Ano.
 
The initial issue cropped up when I was at firmware 3.10 something or another and changed the thermostat from the Omnitstat RC-80 to Omnistat2.
 
That said HAI CS was great as they helped me and replaced the thermostat some 3 times; only to see the comm issue crop up each time.
 
The networking issue really didn't appear until most recent firmware.  Historically it was the comm issue with the thermostat and the time going off.  The networking issue might have been there and I may have not noticed it.
 
Someone on the forum had mentioned to disconnect the network cable for a time and reconnect it.  That always worked. 
 
I will start today by removing the thermostat to see what happens.  (I did move the console and Omnitouch 5.7's to an Omnitouch Hub such that it is easy to remove the consoles one by one).  I also have temperature and combo sensors connected and said devices appear to be fine but will also test these.
 
I did notice similiarly described problems (comm with Omnistat2) with support / installation forum on the Worthington site.  The common theme was the Omnistat2 connected serially.  Diagnostics always involved either ruling out an Omnistat2 or the main OPII boards.  No real mention though of the networking or time issues.
 
Just removed thermostat this morning as its been in the 60's/70's here in the last few days.  Noticed this morning the network connectivity issues / time was off by around 40 minutes.

 

[ano]

You can rule out the standalone temp or temp/humidity sensors. Each of those has a separate zone, so you wouldn't see the problems your getting.  It seems like it has to be a console, serial OmniTouch, Omnistat, or wireless receiver. All those items share a bus.

 

[pete_c]

Thanks Ano.
 
With the Omnitstat2 now disconnected for about an hour I am not seeing any wierd networking disconnects or time being off. 
 
I think I had measured some minutes per group of 10 minutes off time before using coreserver.
 
I was watching the core server stuff and noticed a time thing going on between the thermostat and the panel.  That said I wonder if its a "conflict" of sorts between the serial legacy bus, legacy network and new Zigbee OS (with no radio) in my Omnitouch2. 
 
As much as I really hate to post this; but I never saw anything like this (almost 10 years) with the Omnistat RC-80. (much more basic set up though) and this issue did crop up with the introduction (here in my home) of the Omnistat2.
 
Really this stuff is built to last forever; especially the panel.  I used to just have consoles and the only issue with one over 10 years was one-two blue LED lights going out (and that was covered under warranty).  Personally I toasted one Omnitouch 3.5"  screen by reversing the power leads to it; then making it worse by trying to update the firmware when it was already disfunctional.  I replaced that one though with a Omnitouch 5.7 screen which has been fine now for many years.
 
I have the two types of wireless HAI devices (actually two and one) but never have connected them in the midwest (so no wireless except for Z-Wave on the bus).  In Florida I have wireless (still testing) and have no issues with it; but I am not using it really.