Discussions related to HomeSeer Beta builds 3.0.0.369->3.0.0.423

[Jobee]

I also saw handles moving up slowly and moved to .387 but now now I cannot get it to start. The startup process gets about 3/4 of the way and sits for a while and then comes back with it shut down. Details give:

Problem signature:
Problem Event Name: APPCRASH
Application Name: HS3.exe
Application Version: 3.0.0.387
Application Timestamp: 5a134419
Fault Module Name: clr.dll
Fault Module Version: 4.7.2117.0
Fault Module Timestamp: 59cf5105
Exception Code: c00000fd
Exception Offset: 000187fa
OS Version: 6.3.9600.2.0.0.272.7
Locale ID: 1033
Additional Information 1: 1403
Additional Information 2: 1403c701083761d4e0bcdfb0e98a06e4
Additional Information 3: 9921
Additional Information 4: 9921e24fe9c708a7058981d0e16ee4a5

The handles for HS3.exe are still at about 108 when it crashes.

Also noticed this in the hs_sentry.log:

11/20/2017 9:19:08 PM:Error accessing web server: The operation has timed out
11/20/2017 9:21:08 PM:Error accessing web server: The operation has timed out
11/20/2017 9:23:08 PM:Error accessing web server: The operation has timed out
11/20/2017 9:25:08 PM:Error accessing web server: The operation has timed out
11/20/2017 9:27:08 PM:Error accessing web server: The operation has timed out
11/20/2017 9:28:16 PM:Error accessing web server: Unable to connect to the remote server
11/20/2017 9:28:37 PM:Error accessing web server: Unable to connect to the remote server
11/20/2017 9:28:37 PM:Cannot access HomeSeer, shutting down system and re-starting...
11/20/2017 9:28:42 PM:Starting the HomeSeer HS3 Application
11/20/2017 9:29:03 PM:Error accessing web server: Unable to connect to the remote server

[fvhemert]

Results after update to 3.0.0.383

I have updated to HS3 Standard Edition 3.0.0.383 (Windows) yesterday and I am happy to report the system has become more responsive.

I am no longer seeing the delays in status updates or device changes and HSTouch is way more responsive too.

I have checked the log for PI queue warnings but over a 24 hrs period they did not occur.

Looking at the HS3 handles it still shows the same behavior. A steady increase as soon as HS3 is started. I am currently at 2876 handles with a system uptime of almost 6 hrs. (The HS3 process is restarted every night at 4 am)

[MNB]

I also saw handles moving up slowly and moved to .387 but now now I cannot get it to start. The startup process gets about 3/4 of the way and sits for a while and then comes back with it shut down. Details give:



Problem signature:

Problem Event Name:APPCRASH

Application Name:HS3.exe

Application Version:3.0.0.387

Application Timestamp:5a134419

Fault Module Name:clr.dll

Fault Module Version:4.7.2117.0

Fault Module Timestamp:59cf5105

Exception Code:c00000fd

Exception Offset:000187fa

OS Version:6.3.9600.2.0.0.272.7

Locale ID:1033

Additional Information 1:1403

Additional Information 2:1403c701083761d4e0bcdfb0e98a06e4

Additional Information 3:9921

Additional Information 4:9921e24fe9c708a7058981d0e16ee4a5



The handles for HS3.exe are still at about 108 when it crashes.



Also noticed this in the hs_sentry.log:



11/20/2017 9:19:08 PM:Error accessing web server: The operation has timed out

11/20/2017 9:21:08 PM:Error accessing web server: The operation has timed out

11/20/2017 9:23:08 PM:Error accessing web server: The operation has timed out

11/20/2017 9:25:08 PM:Error accessing web server: The operation has timed out

11/20/2017 9:27:08 PM:Error accessing web server: The operation has timed out

11/20/2017 9:28:16 PM:Error accessing web server: Unable to connect to the remote server

11/20/2017 9:28:37 PM:Error accessing web server: Unable to connect to the remote server

11/20/2017 9:28:37 PM:Cannot access HomeSeer, shutting down system and re-starting...

11/20/2017 9:28:42 PM:Starting the HomeSeer HS3 Application

11/20/2017 9:29:03 PM:Error accessing web server: Unable to connect to the remote server

[mrhappy]

I think I may have traced my issue, I think I carelessly edited an event and used the trigger of 'device has been x for at least' rather than 'device has been x for exactly'. I know in HS2 historically the advice was to use the 'exactly' trigger with care as HS could miss it if it was doing something else so perhaps I had just slipped into old habits.

I'm fully aware that the 'for at least' is going to run continually until either the condition is satisfied or it will just run and run in a loop if it is a sole trigger. I had a multi trigger event (a few OR IF's) so it would only misbehave when one of the other rarer triggers was satisfied but most of the time it worked fine. I didn't have the 'cannot re-run' type setting set on this event as for the other triggers I didn't envisage needing it.

I think it then just ran and ran as fast as HS could run it and because it was being logged it took HS down with it when the SQLite DLL hit an issue (whether record count or speed of writing I don't know). I am loathed to do any experimenting further in case I trash something again but this is my thoughts at the minute.

Perhaps;

1) This trigger is a condition only (not sure the HS API allows for this) as if you are careless with this as a trigger could cause issue? Not sure as a sole trigger whether it is really that commonly used?

2) If it is kept as a trigger then a forced limit in the code so it ran no more than once a second?

Not sure whether other people are using this in their events or not??

[MNB]

Well, I updated to v.387 yesterday everything look great and at 8:43 a.m., I found my system with "HS3 has stopped working" message (and non-responding). Again the log did not have any error message(s) and at the time it was doing a TTS event. So based on my @bout page my System Uptime with v.387 is: 0 Days 12 Hours 52 Minutes 38 Seconds. Looking at the Win Event app it indicated the infamous “clr.dll” application Error.

With v.387 I also can not start up HS3 with a clean computer restart I have the same observations as @JoBee, the HS3.exe startup process gets about 3/4-90% of the way (green progress bar) where it sits for a while (~5 seconds) and then comes back with "HS3 has stopped working" message along with the HS Log stopped. If I simply restart the HS3.exe app again after the crash/hang it will complete the startup process and then minimizes the HS3 icon and appears to be working. Based on this mornings "HS3 has stopped working" I have ~ 12 Hours until the next crash/hang.

@Rich, what do you want from us to help you discover root cause, obviously this is a repeatable fail condition. Mike

[Stuart]

just curious, what if you run it without Hs sentry

Stuart

[MNB]

just curious, what if you run it without Hs sentry



Stuart

I will give it a try later this afternoon, what makes think it might be a factor?




Sent from my iPhone using Tapatalk

[Jobee]

just curious, what if you run it without Hs sentry

Stuart

I turned mine off. My .387 kept crashing after ~4 hours so I reverted back to .384. I still received the dll error. Will try to go back up to .387 and see if I can get it to run by restarting it.

One note: When startup gets to 90% started it sits for about 5 minutes not just 5 seconds. From the time I start the HS3.exe to when I hear the "Welcome to HomeSeer" is now at 12 minutes.

I run on Server 2012R2 and there was a MS update for .NET:

2017-11 Security and Quality Rollup for .NET Framework 3.5, 4.5.2, 4.6, 4.6.1, 4.6.2, 4.7 on Windows 8.1 and Server 2012 R2 for x64 (KB4049017)
Installation date: 11/15/2017 4:21 PM
Installation status: Succeeded
Update type: Recommended

Install this update to resolve issues in Windows. For a complete listing of the issues that are included in this update, see the associated Microsoft Knowledge Base article for more information. After you install this item, you may have to restart your computer.

More information: 
[url]http://support.microsoft.com/kb/4049017[/url]

Help and Support: 
[url]http://support.microsoft.com[/url]

[MNB]

I think I may have traced my issue, I think I carelessly edited an event and used the trigger of 'device has been x for at least' rather than 'device has been x for exactly'. I know in HS2 historically the advice was to use the 'exactly' trigger with care as HS could miss it if it was doing something else so perhaps I had just slipped into old habits.



I'm fully aware that the 'for at least' is going to run continually until either the condition is satisfied or it will just run and run in a loop if it is a sole trigger. I had a multi trigger event (a few OR IF's) so it would only misbehave when one of the other rarer triggers was satisfied but most of the time it worked fine. I didn't have the 'cannot re-run' type setting set on this event as for the other triggers I didn't envisage needing it.



I think it then just ran and ran as fast as HS could run it and because it was being logged it took HS down with it when the SQLite DLL hit an issue (whether record count or speed of writing I don't know). I am loathed to do any experimenting further in case I trash something again but this is my thoughts at the minute.



Perhaps;



1) This trigger is a condition only (not sure the HS API allows for this) as if you are careless with this as a trigger could cause issue? Not sure as a sole trigger whether it is really that commonly used?



2) If it is kept as a trigger then a forced limit in the code so it ran no more than once a second?



Not sure whether other people are using this in their events or not??

[psmaaswinkel]

I just posted build 387 with some minor changes. I would like to release this next week.

Thanks Rich, just update my HS3 running on Windows10. Any issues will be reported..

Regards,

[w.vuyk]

One note: When startup gets to 90% started it sits for about 5 minutes not just 5 seconds. From the time I start the HS3.exe to when I hear the "Welcome to HomeSeer" is now at 12 minutes.

If I install the update >=.368 my installation runs for many minutes too. 12 minutes I have seen, but also more then 20 minutes. Here running on windows 10.

Thanks

wim

[Mr_Resistor]

Please note...extended start-up times are not necessarily new to these beta builds and depend on a number of other factors.

I running v .318, and my startup clocks in around 7-8 minutes. But I'm also running over 400+ Events, 70+ Plug-ins and over 4,000 devices.

Just adding my .02

[Jobee]

Please note...extended start-up times are not necessarily new to these beta builds and depend on a number of other factors.

I running v .318, and my startup clocks in around 7-8 minutes. But I'm also running over 400+ Events, 70+ Plug-ins and over 4,000 devices.

Just adding my .02

Good to keep in mind. I have 14 plugins and about 50 events with 328 devices.

[bartbakels]

Running on 387 now for 8 hours and no issues yet with 1200 devices and about 600 events. However i see an increase in cpu usage


Btw starts up in about 60 seconds on win 7 64 bit next to blueiris with 7 cams which 4 are recording

Regards

Bart

[bartbakels]

Running on 387 now for 8 hours and no issues yet with 1200 devices and about 600 events. However i see an increase in cpu usage


Btw starts up in about 60 seconds on win 7 64 bit next to blueiris with 7 cams which 4 are recording

Regards

Bart

OK To early.....


Just crashed my Homeseer by running my sleep command, alot of zwave command executed, and HS3 crashed with only this popup;


Application popup: HS3.exe - System Error : A new guard page for the stack cannot be created.


rolling back to 357 as we speak.

only application error eventlogs was on the plugins who crashed, not on hs3 itself, but it was clearly crashed nd closed...