Windows 7 Freeze and Possible Time Server Problems

[spaceboy1976]

i have found the solution to the time server not starting. i started poking around and comparing to my laptop config and i noticed that i had no scheduled task to synchronize time. i exported from my laptop and imported on the HTPC. I ran it and it started fine but when i clicked on the history tab loads of errors showed up one for each time it started. they don't seem to have been logged in event viewer though? perhaps because the task was missing?

the error acutally says the task scheduler failed to load the task at the service start up so all the info is there its just not reported for some reason?

anyway the error message i was previously seeing in the internet time tab have been replaced by a nice message telling me it will sync on a regular basis.

so, now to see if that was the cause of the problems or just another issue. i think i've followed all instructions so far and i have re run the tool and attached reports. it actually hasn't frozen since i reported at the weekend so things are looking better. Please can you let me know if there are any problems highlighted by my reports?

i'll report back in a week or if i have any further problems. thanks!

 

[spaceboy1976]

ha ha spoke too soon. i have had two freezes this evening, one when trying an encode and a second 20 mins after restarting just playing live tv.

is it normal that i have security events logged every minute? all the HTPC drives are shared and all the time my laptop is connected to the network there are security events recorded for log on and log off between every 30 seconds and every couple of minutes. i have 34000 of these events in the log

 

[CarlTR6]

i have found the solution to the time server not starting. i started poking around and comparing to my laptop config and i noticed that i had no scheduled task to synchronize time. i exported from my laptop and imported on the HTPC. I ran it and it started fine but when i clicked on the history tab loads of errors showed up one for each time it started. they don't seem to have been logged in event viewer though? perhaps because the task was missing?

the error acutally says the task scheduler failed to load the task at the service start up so all the info is there its just not reported for some reason?

anyway the error message i was previously seeing in the internet time tab have been replaced by a nice message telling me it will sync on a regular basis.

so, now to see if that was the cause of the problems or just another issue. i think i've followed all instructions so far and i have re run the tool and attached reports. it actually hasn't frozen since i reported at the weekend so things are looking better. Please can you let me know if there are any problems highlighted by my reports?

i'll report back in a week or if i have any further problems. thanks!

Well done. Great initiative. And I am glad your RAM passed Memtest. Wow, 17 hours!

ha ha spoke too soon. i have had two freezes this evening, one when trying an encode and a second 20 mins after restarting just playing live tv.

is it normal that i have security events logged every minute? all the HTPC drives are shared and all the time my laptop is connected to the network there are security events recorded for log on and log off between every 30 seconds and every couple of minutes. i have 34000 of these events in the log

I suggest that you post this problem in the Installation & Setup section, Installation & Setup - Windows 7 Forums. This is out of my area of expertise; but we have some Gurus there who are darn good with this sort of thing.

 

[spaceboy1976]

Security Events

Hi, I have a case open over on the crashes and debugging forum and i've been sent here to get some advice on one specific area.

The problem PC is running Win 7 64bit and is solely used as a HTPC running Media Center. All hard drives on the HTPC are shared from the root over the home network so i can do admin on moving recorded tv files around and just general monitoring from a laptop.

in the event viewer i see thousands of events logged in the security section and I just wondered if this was ok? Just now i turned it on and the below is a sample of what i see, security events every 30 seconds or so. The logon event tells me that the laptop has succesfully logged on and then the logoff tells me that the session was destroyed?

all my event logs are attached over on the other thread http://www.sevenforums.com/crashes-...7-freeze-possible-time-server-problems-5.html

any help gratefully received, thanks

 

[WindowsStar]

I believe this is normal, for network connections (such as to a file server), it will appear that users log on and off many times a day. This phenomenon is caused by the way the Server service terminates idle connections. -WS

 

[spaceboy1976]

ok, is there anything else i can do to test where the problem is? i had three more freezes yesterday. the only pattern i can see is that its more likely to happen when i'm encoding video, whether thats because its working at 100% or something else i don't know. it also happens at idle though, i was watching a flash video when it froze one of the times last night.

 

[WindowsStar]

The only thing I can think of is you are having an issue with the video driver. It seem every time you a freeze it is when you are using some form of video. Do you have a different video card to try in this machine?? Especially if it is a different brand. -WS

 

[spaceboy1976]

i thought it might have been catalyst control center so i did a clean install and just used the microsoft drivers. it still froze so i just restored back.

it doesn't always happen when playing video so i don't think its this. i've just updated the graphics drivers to the latest version and i'm running prime95 for a few hours as working at 100% does seem to be a factor. this may rule that out at least

 

[CarlTR6]

I also think it is a driver issue. Try some older drivers; sometimes the latest and the greatest does not work the best. As WS suggested, try a different video card.

 

[spaceboy1976]

yeah i really want to try and troubleshoot this one as if i replace anything it'll be to get onboard video and a new processor. i can't afford that for a couple of weeks so i'd like to use that time to try and confirm if that is the problem.

the hd 4850 doesn't really seem to have too many reported problems. prime95 has been running for over an hour now maxing the cpu at 100%. its hot but plenty below the maximums. i'm going to run for another hour and then perhaps overnight.

do you know of any software for stress testing GPU's? i have seen one called furmark

 

[CarlTR6]

Furmark is the one we generally recommend. Here is the tutorial. Be sure to monitor your GPX temps. http://www.sevenforums.com/tutorials/100356-video-card-stress-test-furmark.html

 

[spaceboy1976]

well furmark has shown some interesting things. it caused my pc to shut down within 5 mins of running on two occasions so i opened up the case and ran it again. again shuts down then the gpu reaches 120C.

not unexpected i guess, 120C sounds quite hot. my hd4850 has passive cooling so temps like this aren't unexpected.

however, lid off and a fan extracting hot air from the local area it is leveling out at 107C. so apart from that my gpu is too hot what does this mean? its cleanly shutting down rather than freezing so is this the same thing? i doubt just rendering HD tv gets the gpu anywhere near 100% and even if it does i've seen freezes when video isn't playing.

 

[WindowsStar]

Anything greater than 100C is too hot for a video card. You need to have better air flow or better cooling. -WS

 

[CarlTR6]

Since that is over the boiling point of water, I would say it is a little too warm. The shutdown is to protect your hardware. This indicted that your GPU may be faulty. Do you have another one you can try? Can you borrow one? I think the freezes may be related to the sutdowns in that there is a GPU problem.

 

[spaceboy1976]

i agree that this is too hot but i don't think there's anything faulty with the gpu. as i said its passively cooled, specifically designed for a HTPC and not a gaming rig. its not designed to operate at anything near 100% without much more cooling. but according to ATI it operates perfectly up to 120-130C

but as i said the gpu barely touches 25% playing near full screen hd video. temps around 80C. its not this. totally different reaction by the pc confirms this

 

[spaceboy1976]

first two paragraphs Review - Gigabyte Radeon HD 4850 1GB (GV-R485MC-1GH) | bit-tech.net

 

[WindowsStar]

i agree that this is too hot but i don't think there's anything faulty with the gpu. as i said its passively cooled, specifically designed for a HTPC and not a gaming rig. its not designed to operate at anything near 100% without much more cooling. but according to ATI it operates perfectly up to 120-130C

but as i said the gpu barely touches 25% playing near full screen hd video. temps around 80C. its not this. totally different reaction by the pc confirms this

What ATI is talking about is extreme conditions for a small amount of time. A GPU is never supposed to run hotter than 100C. I have been working with all kinds of computer systems for over 40 years. Anytime you have electronics over 100C you will get errors, shutdowns, reboots, lockups, pausing, et cetera. I did a quick survey of several machines in the office that are running high-end graphic calculations and the GPU never gets hotter than 65C running at 100% CPU. -WS

 

[spaceboy1976]

i agree that this is too hot but i don't think there's anything faulty with the gpu. as i said its passively cooled, specifically designed for a HTPC and not a gaming rig. its not designed to operate at anything near 100% without much more cooling. but according to ATI it operates perfectly up to 120-130C

but as i said the gpu barely touches 25% playing near full screen hd video. temps around 80C. its not this. totally different reaction by the pc confirms this

What ATI is talking about is extreme conditions for a small amount of time. A GPU is never supposed to run hotter than 100C. I have been working with all kinds of computer systems for over 40 years. Anytime you have electronics over 100C you will get errors, shutdowns, reboots, lockups, pausing, et cetera. I did a quick survey of several machines in the office that are running high-end graphic calculations and the GPU never gets hotter than 85C running at 100% CPU. -WS

but this isn't relevant. i know the gpu is not designed to be run at 100% which is what caused these temperatures. right now I'm watching HD video and its using 12% of the GPU and running at 79C. The GPU in my HTPC never, ever runs at anything approaching what the stress test put it through.

a quick survey of forum posts about this card (and make sure its the gigabyte one with passive cooling) will show that these tempeartures are entirely normal.

 

[WindowsStar]

i agree that this is too hot but i don't think there's anything faulty with the gpu. as i said its passively cooled, specifically designed for a HTPC and not a gaming rig. its not designed to operate at anything near 100% without much more cooling. but according to ATI it operates perfectly up to 120-130C

but as i said the gpu barely touches 25% playing near full screen hd video. temps around 80C. its not this. totally different reaction by the pc confirms this

What ATI is talking about is extreme conditions for a small amount of time. A GPU is never supposed to run hotter than 100C. I have been working with all kinds of computer systems for over 40 years. Anytime you have electronics over 100C you will get errors, shutdowns, reboots, lockups, pausing, et cetera. I did a quick survey of several machines in the office that are running high-end graphic calculations and the GPU never gets hotter than 65C running at 100% CPU. -WS

but this isn't relevant. i know the gpu is not designed to be run at 100% which is what caused these temperatures. right now I'm watching HD video and its using 12% of the GPU and running at 79C. The GPU in my HTPC never, ever runs at anything approaching what the stress test put it through.

a quick survey of forum posts about this card (and make sure its the gigabyte one with passive cooling) will show that these tempeartures are entirely normal.

Humm a bit confused here. I thought you were stating that your GPU was overheating, but now it is irrelevant? Humm.

I still think your GPU is running too hot, but that is just MHO.

Note: I entered 85C I meant 65C, typo.

 

[spaceboy1976]

no it isn't relevant. the two situations are completely different. Furmark is a test intended to stress overclocked gaming pc's . If i were intending to use my pc as a gaming pc furmark would be relevant to my situation.

this is a HTPC. the card is designed for a HTPC to deliver video at minimum GPU levels, but with zero sound. there is no fan, it is bound to be hotter than the average card, but this is the trade off for zero sound.

all furmark showed is that my graphics card is not designed to be run at 100% activity. ever. but i knew this.

did you read any of the review i linked to, here is a salient piece

With the Gigabyte GV-R485MC-1GH, the card continued to run without any airflow in an ambient temperature of 26 to 27°C at over 100 degrees. The maximum temperature we recorded on the fins was a finger-melting 98.7°C and, according to RivaTuner, the core hit 106°C during our stress test. We then left the card to settle down and we recorded an idle core temperature about an hour later of 62°C, while the fins were a respectable 57°C.

Despite these rather extreme temperatures, the card kept running and didn't crash - we were pretty impressed with this result considering it was designed to essentially cook the card until it fell over. Gigabyte's Multi-Core cooling solution certainly seems to be up to the task even in environments lacking airflow completely.

as a final test i ran prime 95 for 1.5 hours while playing back HDTV on full screen to simulate the encoding process and the gpu temp has only reached 86C. this is a standard temp for this card. it is not overheating under normal circumstances