HouseBot

Hi Scott,

I have read the other threads about HouseBot crashes and dumps and I figured they don't really match my problem, so I decided to start a new thread.

HouseBot has been dumping on me completely randomly for the past several weeks. Here are the time and dates:



HouseBotServer-10-01-2003.DMP 2,302 10/1/2003 7:14:36 PM

HouseBotServer-10-02-2003.DMP 879 10/2/2003 9:39:18 PM

HouseBotServer-10-03-2003.DMP 1,884 10/3/2003 5:45:08 PM

HouseBotServer-10-07-2003.DMP 1,073 10/7/2003 8:22:14 PM

HouseBotServer-10-17-2003.DMP 4,328 10/17/2003 6:05:50 AM

HouseBotServer-10-25-2003.DMP 1,364 10/25/2003 10:00:30 PM

HouseBotServer-11-01-2003.DMP 1,955 11/1/2003 11:22:54 PM

HouseBotServer-11-07-2003.DMP 1,664 11/7/2003 10:21:46 PM

HouseBotServer-11-11-2003.DMP 1,768 11/11/2003 8:48:24 PM

HouseBotServer-11-13-2003.DMP 2,043 11/13/2003 9:30:32 PM

HouseBotServer-11-14-2003.DMP 2,069 11/14/2003 9:30:56 PM



Below is the latest dump file, which happened a couple of hours ago:





====== Begin Dump - Friday, November 14, 2003 21:30:55 ======

Server Version = 1.51



==================================

======== House Server Thread =======

==================================

Thread Type = Device Thread [Living Room Left Side Lamp]

Thread ID = 594

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 77C43283 01:00032283 C:\WINDOWS\system32\msvcrt.dll



Registers:

EAX:06920DC4

EBX:00000001

ECX:3FFC8371

EDX:00000003

ESI:069FFFFD

EDI:069FFFFC

CS:EIP:001B:77C43283

SS:ESP076FFB80 EBP:076FFB88

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010202

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004C83E3 01:000C73E3 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:24069B4B

EBX:069C8108

ECX:069C8108

EDX:06F419A8

ESI:069155C0

EDI:06F40690

CS:EIP:001B:004C83E3

SS:ESP06CFF23C EBP:06FF4528

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004C83E3 06FF4528 0001:000C73E3 C:\Program Files\HouseBot\HouseBotServer.exe

069CE530 00531688 0000:00000000

004C9B20 004C9A90 0001:000C8B20 C:\Program Files\HouseBot\HouseBotServer.exe





====== End Dump ======



====== Begin Dump - Friday, November 14, 2003 21:30:55 ======

Server Version = 1.51



==================================

======== House Server Thread =======

==================================

Thread Type = Device Thread [System Time]

Thread ID = 42C

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004C83E3 01:000C73E3 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:24069B4B

EBX:069C8108

ECX:069C8108

EDX:06F419A8

ESI:069155C0

EDI:06F40690

CS:EIP:001B:004C83E3

SS:ESP074FF810 EBP:06FE3138

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004C83E3 06FE3138 0001:000C73E3 C:\Program Files\HouseBot\HouseBotServer.exe

069C1248 0053169C 0000:00000000

004CA560 004C9D80 0001:000C9560 C:\Program Files\HouseBot\HouseBotServer.exe

FFFFFF58 E8F18B56 0000:00000000





====== End Dump ======



I am running HouseBot v1.51, Windows XP with SP2, Dual P3-866MHz CPU, 640 MB RAM, and I don't think this matters but I'm running RAID 1 (mirror) on two 30GB hard drives for redundancy.

Any ideas??



Thanks,

- John

John,



If you are using a CM11a, be sure to download version 1.60.



I've never tested HouseBot on a dual CPU. You might want to try changing the affinity for HouseBot to only use a single CPU to see if that helps.



Scott

Hi Scott,

Well I upgraded to v1.60. But now I can't even create a new device !! As soon as I click Create New Device, HouseBot just closes. I tried asigning HouseBotServer.exe to one processor as well, but it still crashes.

Here is the dump after trying to create a new device:





====== Begin Dump - Saturday, November 15, 2003 16:30:11 ======

Server Version = 1.60



==================================

======== House Server Thread =======

==================================

Thread Type = Main UI Thread

Thread ID = 238

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004B6AC0 01:000B5AC0 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:00000000

EBX:00AEAD80

ECX:00000000

EDX:001548D8

ESI:0012F8AC

EDI:0012F90C

CS:EIP:001B:004B6AC0

SS:ESP0012E8A0 EBP:0012F8AC

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004B6AC0 0012F8AC CxImage::`copy constructor closure'+B5050





====== End Dump ======



====== Begin Dump - Saturday, November 15, 2003 16:30:35 ======

Server Version = 1.60



==================================

======== House Server Thread =======

==================================

Thread Type = Main UI Thread

Thread ID = F8

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004B6AC0 01:000B5AC0 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:00000000

EBX:00AEAD80

ECX:00000000

EDX:001548D8

ESI:0012F8AC

EDI:0012F90C

CS:EIP:001B:004B6AC0

SS:ESP0012F220 EBP:0012F8AC

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004B6AC0 0012F8AC CxImage::`copy constructor closure'+B5050





====== End Dump ======



====== Begin Dump - Saturday, November 15, 2003 16:33:13 ======

Server Version = 1.60



==================================

======== House Server Thread =======

==================================

Thread Type = Main UI Thread

Thread ID = 6CC

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004B6AC0 01:000B5AC0 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:00000000

EBX:00AEAD88

ECX:00000000

EDX:001548D8

ESI:0012F8AC

EDI:0012F90C

CS:EIP:001B:004B6AC0

SS:ESP0012F220 EBP:0012F8AC

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004B6AC0 0012F8AC CxImage::`copy constructor closure'+B5050





====== End Dump ======



====== Begin Dump - Saturday, November 15, 2003 16:33:34 ======

Server Version = 1.60



==================================

======== House Server Thread =======

==================================

Thread Type = Main UI Thread

Thread ID = 7B8

Exception code: C0000005 ACCESS_VIOLATION

Fault address: 004B6AC0 01:000B5AC0 C:\Program Files\HouseBot\HouseBotServer.exe



Registers:

EAX:00000000

EBX:00AEAD88

ECX:00000000

EDX:001548D8

ESI:0012F8AC

EDI:0012F90C

CS:EIP:001B:004B6AC0

SS:ESP0012F220 EBP:0012F8AC

DS:0023 ES:0023 FS:003B GS:0000

Flags:00010246



Call stack:

Address Frame

004B6AC0 0012F8AC CxImage::`copy constructor closure'+B5050





====== End Dump ======



I wonder if it's Windows XP or something. Do you have any other ideas? Should I go back to v1.51 ??



Thanks,

- John

John,



It's crashing in a different place now. After you create the Device, are you clicking on a Theme or does it crash immediately after creating the Device?



Also, are you using a CM11a?



Was there something that you changed in the configuration before this started happening (if you can remember)?



I'd like to try and reproduce the problem here. Could you send my your HBData.mdb file (found in \HouseBot\Config) for starters? Maybe there's an anomalous configuration setting somewhere. Send it to [email protected].



If you want to reinstall version 1.5x, you should be able to install over the 1.6, but you will first need to delete the HouseBotServer.exe file.



Scott

John,



If you want to reinstall an older version, I would suggest you install 1.52 instead of 1.51. Here's a link 1.52.



Scott

Hi Scott,

It crashes right after I click on Create new Device. Or right click and select Create New Device. It doesn't even go into the device creation window. It just crashes. I am doing nothing to the themes either.

I am using the CM11A, and I have not changed the configuration at all for quite some time.

Strange thing is, I installed 1.60 on my laptop (Win98), then copied everything onto there, and I created a device there with no problems. Then I copied back to my PC and the device was there. I just can't seem to add a device on this PC.

So it appears that maybe there's something on this PC that HouseBot doesn't like. Not sure if there are any other problems, but this is all I have found so far. I will email you the mdb file right now.

Thanks Scott.



- John

John,



I think I've fixed the 1.60 crash you were seeing. The fix will be in the 1.61 update that I plan on releasing next week. If you're interested in trying it before that, let me know and I'll pass a beta copy along to you.



If you installed version 1.52, I'm interested in knowing whether it fixed your original problem. There was fix in 1.52 that solved something very similar to what you were seeing.



Scott

Hi Scott,

I just went ahead and installed 1.52 just to see, and yes it did solve the problem. I created a device successfully without any problems.



I will wait for you to release 1.61 next week, unless you think it might be helpful in your development to install the beta now. If not, I will continue to run 1.52 until the release.



Thanks a bunch,

- John

Hi Scott,



I guess I should have started two separate threads since there are really two issues here. So I'd like to clarify.



In my response, version 1.52 seems to have resolved the "Create New Device" crash problem, but I am not sure whether it has resolved the random dump problem.

If you installed version 1.52, I'm interested in knowing whether it fixed your original problem. There was fix in 1.52 that solved something very similar to what you were seeing.

I assume by "original problem" you mean the random crash problem. Sometimes it takes a few days to dump. So I guess I'll let you know.

I think I've fixed the 1.60 crash you were seeing. The fix will be in the 1.61 update that I plan on releasing next week.

By this, did you mean the random crashing, or the crash due to "Create New Device"? Or both?



Sorry about the confusion.



- John

John,



By Original Problem I meant the random crashes. In looking at your original dump files I saw a similar problem that was a fixed in 1.52. So hopefully you will be ok.



I really doubt you will see the Create New Device problem with 1.52. In 1.60, I enhanced/reworked that dialog. I can't say that I completely understand why you are seeing the problem in 1.60, but I've added some defensive code so at least it won't crash. I'll be interested to hear how it works for you.



Scott

Hi Scott,

I have been running HouseBot 1.61 now for about a week and no crashes

Only thing is, I get some errors in the error log, but it doesn't seem to affect the operation of HouseBot or any of its tasks:



-MR26A has recieved an invalid unit code(0xb0) etc..

-Error reading COM port ....

-Due to an error device [xxxxx] did not change state ...(but it actually did)



etc.....



So I'm not sure what that means. Should I just ignore these entries since they don't seem to really affect anything?



Thanks,

- John

John,



It's good to hear that 1.61 is working better for you.



Sometimes I'm overly conservative and probably log too many things as errors. If the MR26 receives some garbage, or a trashed signal, it's possible that you may see some anomalous errors. It's also possible that there's a bug in the Interface that has not yet been uncovered.



If all of the errors are related, it might point to a single problem so you can ignore everything except the initial error message.



Feel free to send me the error logs, if you want.



Scott

Hi Scott,

First off, Merry Christmas and Happy New Year. All the best for you and your family in the new year, and thanks for all the hard work.



Sorry it took so long to send the errors, but I realized I wasn't logging, and then before you know it the holiday season was upon us, so I kinda forgot



Well here are a few examples of the errors I'm getting. But like I said, they don't seem to be affecting HouseBot's functionality.



Dec 30 2003,07:50:09PM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x80]"

Dec 31 2003,10:07:13AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x81]"

Dec 31 2003,11:39:12AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x41]"

Dec 31 2003,02:13:21PM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x80]"

Dec 31 2003,07:10:41PM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x31]"

Dec 31 2003,08:55:12PM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0xc6]"

Jan 01 2004,12:32:12AM,X10 CM11A Module,Error,"Error reading COM port. Error = Insufficient data in read buffer"

Jan 01 2004,12:32:13AM,Property:Power State,Error,"Due to an error, Device [Front Yard Camera] Property [Power State] DID NOT change."

Jan 01 2004,12:32:23AM,X10 CM11A Module,Error,"Error reading COM port. Error = Insufficient data in read buffer"

Jan 01 2004,12:32:24AM,Property:Power State,Error,"Due to an error, Device [Front Yard Camera] Property [Power State] DID NOT change."

Jan 01 2004,12:32:35AM,X10 CM11A Module,Error,"Error reading COM port. Error = Insufficient data in read buffer"

Jan 01 2004,12:32:35AM,Property:Power State,Error,"Due to an error, Device [Front Outside Spot Light] Property [Power State] DID NOT change."

Jan 01 2004,12:32:46AM,X10 CM11A Module,Error,"Error reading COM port. Error = Insufficient data in read buffer"

Jan 01 2004,12:32:46AM,Property:Power State,Error,"Due to an error, Device [Front Outside Lights] Property [Power State] DID NOT change."

Jan 01 2004,02:25:24AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x82]"

Jan 01 2004,06:57:49AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x1]"

Jan 01 2004,09:44:38AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x90]"

Jan 01 2004,09:47:01AM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0xc4]"

Jan 01 2004,12:12:53PM,X10 MR26A Module,Error,"MR26a has received an invalid unitcode value [0x2]"



The "invalid unit codes" are pretty self-explanatory. But for the other ones, even though the errors say that the "power state did not change", it did. So I'm not too concerned about it, but it would be nice to get "real" errors in the error logs. Maybe these are real errors I don't know...It does occur occasionally for other devices as well.

I'm still running v1.61 and it has been rock solid let me tell you !!



Thanks,

- John

John,



Thanks for the info and I'm glad that 1.61 is stable for you. I recall that when I used to use the MR26, I was seeing these types of errors as well. I upgraded to the W800 and didn't receive any errors.



If you are receiving the error AND the device state is still changing, it's probably because the MR26 received a good signal and a bad signal right after one another. The MR26 repeatedly reports the same information.



It may very well be that the sensitivity of the MR26 is such that I need to buffer the error message and not send it if a valid command is received right after it.



Since it is functionally working for you, I'm going to add it low on my list for right now. I'll get to it, but I've just got some bigger items ahead of it.