Computer question. Technical advice requested.

[mav]

For the last couple of days my computer has been acting strange during boot up. When I got home this evening my computer would not boot into windows. I tried several things including trying to boot off the windows disk, but nothing was happening. Out of desperation, I switched ports on my master drive. That seemed to work initally. However, it was taking forever (~5 minutes) to load up Windows.

Once I got into Windows, I thought I might have a virus even though I thought it was highly unlikely. My Avast program was turned off, and my computer would not allow me to turn it back on, nor would it allow me to go to the Avast website and download the program. I managed to get the program through another site. I removed the old program and installed the new program including the updates. The Avast program will not work at all. It will not scan anything. It's just crazy.

I managed to get a copy of AVG. I installed it and ran a full scan. It found absolutely nothing. Malewarebytes didn't find anything as well. Everything seems back to normal, but this was a serious WTF? moment. Anybody have an ideas, outside of it being a PC, on what could have been going on?

Thanks.

[MacGyver]

If I'm reading your description correctly, you've likely got a hard drive starting to go bad. This is a great time to make sure you've taken the time to back everything up. You're probably going to need them pretty soon.

[Symbolic]

You probably have a virus. Backup all the data you can (if you can) to an external HD or DVD. Reformat your entire system afterwords. You don't want any trace of this thing left behind. This happened to me last week. I had no control over my computer and I have no idea how the virus got past my AV.

[bubbadavis]

Haunted computer? It is Halloween...

Based on the evidence you've laid out, I gotta agree with MacGyver. Sounds very much like what my Dad's laptop started doing - the Windows event logs showed detected bad sectors on the hard drive once I got the thing to boot.

[BPE]

More than likely a virus. Press the Windows key and R. Type 'msconfig' and disable all Startup items on the Startup tab. Backup data. Either take to a repair shop, friend in the know or reload OS yourself. Or maybe time for a new PC. You said in your post you switched ports for your master drive. This terminology typically refers to IDE technology. If you have a system with an IDE drive it is time to upgrade and worth your money and saved time to do so. Best of luck. Though, you shouldn't need luck, just a little work to get it all sorted out. And hang up Avast and AVG. Find AntiVir at http://free-wc.com. It may not correct your current issue but will most definitely prevent it from happening again.

[BPE]

That should have been http://free-av.com - sorry.

[mav]

When I get through running my backup, I will check my hard drive. I wouldn't think that was the problem because I just installed two new drives about 2 months ago when I updated my both of my old video cards.

When I mentioned "port," I was referring to one of SATA connections.

I was hoping to get a few more years on this computer before I scrap it and build a new one. I am currently running:

EVGA X58 3X SLI LGA 1366

i7 920 OC to 3.0 GHz

2 x 2TB HD

2 x GTX 570

1000 pws

12 GB DDR3 1600

Coolermaster V8 heatsink

Antec 1200 case

There are times when I really hate computers.

Edited November 1, 2011 by mav

[bubbadavis]

I've had drives die prematurely. I've also had them last way past when they should have died. Just because its a newer drive doesn't mean that isn't the problem (even if they shouldn't die early).

This is one of those problems that's hard to diagnose second hand - too many possibilities with similar symptoms.

[MacGyver]

MTBF isn't what it used to be. As a forensics firm, we go throught dozens, if not hundreds of hard drives a year. Though it pains me to say it, they really don't make them like they used to anymore. Nothing makes me madder than having a pretty new drive fail. Since ours generally contain evidence we can't just send them back - it's off to the shredder (which in and of itself is some consolation).

[eapking]

My vote is also the hardrive.

[bubbadavis]

MTBF isn't what it used to be. As a forensics firm, we go throught dozens, if not hundreds of hard drives a year. Though it pains me to say it, they really don't make them like they used to anymore. Nothing makes me madder than having a pretty new drive fail. Since ours generally contain evidence we can't just send them back - it's off to the shredder (which in and of itself is some consolation).

We have drives at work that must be decommissioned via physical destruction (according to policy). We usually just take them out and put .308 and .45 holes in them.

[Jonnin]

virus scanners that work on a booted PC suck, I have not seen one that really works well to date. Take the drive out and mount it into a clean system as a slave drive, and scan it from there with a high quality product, if you have access to all the stuff needed to do that. You will find your virus if that is the issue. If the drive is going bad, it should tell you, modern PCs with modern drives tell you if the disk is acting up, or many of them do. And you can run the old stuff, check for bad sectors, scandisk, chkdsk, whatever it is called now, I use it and still forget the new name.

We are required to grind off the media from the platters when we destroy a disk. Overkill, really. I have some good magnets from older disks though.

[Guest Lester Weevils]

Yep, modern mtbf ain't good.

I've started running RAID 1 on a pair of boot drives except for laptops. The PC's with the Intel chip mobo's seem to usually native-support RAID 1 pretty nice nowadays with no special extra drive cards or non-intel RAID software, as do the Mac Pro's.

I had a RAID 1 boot array degrade within a year on the quad core HP PC with WD 1 TB drives. One drive went down and the other one was fine. So as soon as that happens, one rushes to Best Buy or wherever and gets a replacement drive for the bad one, install it and tell the RAID to rebuild itself.

Similarly have experienced degrading of two different RAID 1's on the Mac Pro since 2008. I run four drives in that one configured as two RAID 1 volumes. Worked out OK as long as I hurried up and replaced the bad drive ASAP and did a rebuild.

The big problem that makes redundant RAID "not as safe as it ought to be" is that all the drive manufacturers occasionally release a run of bad drives. OTOH it usually works best to have a pair of identical drives in a RAID 1. So if you buy a matched pair of new drives out of a bad production run and then one of them prematurely fails, then it is not-bad odds that the remaining good one in the pair could fail at any time. Which is why I try to replace a bad one immediately. Stop everything, turn off the puter, and go buy a new drive.

Perhaps statistically one is a "little safer" once one of the drives in a RAID 1 has failed? If you have ferinstance a one year old drive and a brand new drive in the pair, the odds seem less likely that both might fail within the same day or two?

I guess if somebody's biz involved a whole bunch of elaborate multi-drive redundant RAID's, then maybe it would be worth going to some kind of effort to "spread out" the manufacture date of the drives? So as to avoid ferinstance installing an 8 disk RAID out of the same carton of drives, and then if there is a defect maybe they all fail like clockwork within a few days and you lose data anyway. Which would definitely ruin the purpose of going to all that trouble and expense.

[Jonnin]

Yep, modern mtbf ain't good.I've started running RAID 1 on a pair of boot drives except for laptops. The PC's with the Intel chip mobo's seem to usually native-support RAID 1 pretty nice nowadays with no special extra drive cards or non-intel RAID software, as do the Mac Pro's.I had a RAID 1 boot array degrade within a year on the quad core HP PC with WD 1 TB drives. One drive went down and the other one was fine. So as soon as that happens, one rushes to Best Buy or wherever and gets a replacement drive for the bad one, install it and tell the RAID to rebuild itself.Similarly have experienced degrading of two different RAID 1's on the Mac Pro since 2008. I run four drives in that one configured as two RAID 1 volumes. Worked out OK as long as I hurried up and replaced the bad drive ASAP and did a rebuild.The big problem that makes redundant RAID "not as safe as it ought to be" is that all the drive manufacturers occasionally release a run of bad drives. OTOH it usually works best to have a pair of identical drives in a RAID 1. So if you buy a matched pair of new drives out of a bad production run and then one of them prematurely fails, then it is not-bad odds that the remaining good one in the pair could fail at any time. Which is why I try to replace a bad one immediately. Stop everything, turn off the puter, and go buy a new drive.Perhaps statistically one is a "little safer" once one of the drives in a RAID 1 has failed? If you have ferinstance a one year old drive and a brand new drive in the pair, the odds seem less likely that both might fail within the same day or two?I guess if somebody's biz involved a whole bunch of elaborate multi-drive redundant RAID's, then maybe it would be worth going to some kind of effort to "spread out" the manufacture date of the drives? So as to avoid ferinstance installing an 8 disk RAID out of the same carton of drives, and then if there is a defect maybe they all fail like clockwork within a few days and you lose data anyway. Which would definitely ruin the purpose of going to all that trouble and expense.

I think the most useful raid purpose is to keep a web server up 24-7, rather than preserve against data loss it is used to reduce/eliminate downtime (along with other tricks, such as physically seperate backup servers and so on). Those are often hot swap drives so you do not even turn the machine off to fix it. Second most useful thing about a raid is for speed, in disk intensive applications. For a home PC, raid zero is fine, and backup to a portable usb hard drive, or no raid at all and backup the same way. I wouldnt fool with raids at home for backup/data safety, just copy the files now and again -- a batch or script file with 7zip to get the important stuff and a full mirror once in a blue moon is sufficient. The speed benefits of raid0 are questionable, so its just not something I can recommend to most users.

[Guest Lester Weevils]

I think the most useful raid purpose is to keep a web server up 24-7, rather than preserve against data loss it is used to reduce/eliminate downtime (along with other tricks, such as physically seperate backup servers and so on). Those are often hot swap drives so you do not even turn the machine off to fix it. Second most useful thing about a raid is for speed, in disk intensive applications. For a home PC, raid zero is fine, and backup to a portable usb hard drive, or no raid at all and backup the same way. I wouldnt fool with raids at home for backup/data safety, just copy the files now and again -- a batch or script file with 7zip to get the important stuff and a full mirror once in a blue moon is sufficient. The speed benefits of raid0 are questionable, so its just not something I can recommend to most users.

Hi Jonnin. Those are good points. I suppose it depends on situation. RAID 1 is not my only backup strategy, but my working setup is near a half-terabyte. If I have an image of my working setup with all programs, data and settings, it takes a long time to restore/copy the image to a new internal hard drive. I haven't lately put a stopwatch to it, but an estimate of about 12 to 24 hours seems about right.

Alternately, if for whatever reason the last image is not suitable for restore or if I screwed the pooch and negligently did not back up everything I should have, then it takes me 2 or 3 days to set up everything from scratch before I'm ready to work. Maybe I'm real slow, but thats just how long it takes to install everything. That is one reason I rarely buy a new computer except when the old one is dead or hopelessly old and slow. Too much aggravation to spend the setup time unless the benefit outweighs the aggravation.

By comparison, a degraded RAID 1 from failure of one drive in the pair, it takes me a lot less than 1 hour to swap in a drive, plus the downtime going to buy a new drive. But even with a failed non-raid drive you still have to go get a replacement drive.The RAID 1 rebuilds in the background, so as soon as the new drive is installed I can go back to work. I used to use removable-bay hard drives and still think they are a good idea, but a drive problem maybe once a year, it isn't worth the trouble anymore.

[mav]

I ran dskchk last night and every thing was fine. My computer seems to be back to normal now. My guess is this was some software error and/or a bad SATA connection ( I have had a few go bad on other motherboards). I will probably update this motherboard and the memory late next year.

Thanks for the input folks, I appreciate it.