'No Fix - No Fee' and a wayward hard drive

sorcerer

Active Member
Reaction score
77
Location
Preston, Lancs, UK
All computer shops around here offer 'No Fix - No Fee' so I had to as well or I just wouldn't get any work. A regular client called me a couple of weeks ago to say that her Windows 8.1 laptop kept BSODing, so I had a look. The laptop was set to save minidumps but actually wasn't doing it, so there were no previous dumps to look at. On top of that, the fault was very intermittent - I had it on the bench for five days, running all sorts of tests and stressing it out but it only blue-screened on me once in that time and again, did not create a dump file to read, even though it was set to do so - and I'll admit that that in itself had me stumped. I don't know why it wasn't creating and saving a dump file, nor do I have any clue or idea of how to make it do it. What do you think?

I don't like admitting defeat but I eventually had to because I just couldn't find anything wrong. She ended up taking it elsewhere and the technician there diagnosed a faulty hard drive and installed a new one.

During my hardware diagnostics procedure, I always run an extended GSmartControl test and in this case, it completed without any errors. There were no previous errors logged, no attributes highlighted in red or anything, just a clean, error-free result. I also usually run HDDScan v3.3 too and again, no errors, no bad or even slow-responding sectors were found. In other words, the result of my testing showed a good hard drive with no cause for concern.

My question to you venerable folks is, what tool do you think this other tech would have been using to come to the conclusion that the HDD was at fault, because whatever it is, I think I need it in my toolkit.

The client has only had the laptop back for a couple of days so I suppose time will tell if his diagnosis was correct or not. She's booked me to go and install her printer etc., and get it all set up for her as she likes it, so at least I know I haven't permanently lost a client there ;)
 
I personally think that all the drive tests you did were correct. The other 'tech' simply thought of the quickest way of getting the repair completed.

Are you 100% certain that he did install a new drive?, and not simply nuked it?

When you go round, check the hours used on the drive, and you should be able to tell.
 
It's easy for someone to say "bad drive, let's replace an reinstall", without any proper diagnostics, knowing that their chances are extremely good that it will fix the issue. A fresh drive and install usually does fix a vast majority of BSOD and software issues anyhow. Especially if your customer told them that you had it for a week already. The tech could easily assume you tested memory, mobo, etc...
 
If a HDD test tells me the drive is bad, I replace the drive. If it tells me it is good but things are still off then I go by my gut instinct - sometimes that is replace the drive sometimes it is not.
 
Perhaps the drive has intermittent issues that didn't show up in your test that did in his. Assuming that the drive is good, I have seen a healthy drive fail a test here on one system and pass on another, usually due to a faulty cable. When you did a test, did you run a long test for many hours, testing every sector on the drive?

That said, I have also done data recovery from failing drives that were tagged as healthy by the tech who did a nuke and pave without backing up all the client's data. During the mirroring process, we encounter lots of bad sectors showing that the healthy drive isn't so healthy after all.

Too bad you couldn't get your hands on the original drive again so that you could confirm that it is bad and figure out where you went wrong with your diagnosis.
 
gSmartcontrol is a good tool and I have used it plenty of times. I'll usually use CrystalDiskinfo just to look at the SMART then follow up with gSmart later on. First doing a Quick Scan than later running an Extending over night.

I've had a drive come up completely fine one moment and then come up failing the next. Of course all data was previously backed up.
 
It's easy for someone to say "bad drive, let's replace an reinstall", without any proper diagnostics, knowing that their chances are extremely good that it will fix the issue. A fresh drive and install usually does fix a vast majority of BSOD and software issues anyhow. Especially if your customer told them that you had it for a week already. The tech could easily assume you tested memory, mobo, etc...
I am guessing this is what happened. The other shop knew the history, and went the Geek Squad route - simple fix, high probability of fixing the problem, and a nice sized invoice.
 
I had a similiar issue, I took out the HD to use my USB to SATA cables. I used GSmartControl and ran a short test. The results came back with no problems. I put the HD back in to troubleshoot the laptop then BSOD on me within 15 minutes. I took out the HD again to run an extended test to test the drive completely. BAM!!! Results came back with an intermittent error, a couple of badsectors. Had to take out the HD again to backup the contents (docs,pics,vids,etc), put in a new HD I had lying around, complete Windows reinstall, restored the contents. Just to be safe, I had the laptop for an extra day for further testing. No more BSOD's. I eventually learned my lesson to do a long and extended test on HD's. Pretty tricky and mind bobbling.
 
Usually when you are set to get a dump and don't, its because the cause of the BSOD directly relates to writing the Dump to the drive, AKA either HDD or southbridge issues. It is also possible that Windows just wasn't honoring the dump command, because something in it was corrupt.
I would have ran the same HDD tests you did, so you did fine there. But if my gut still pointed me at the drive, I prolly would have made a clone and either A: dropped that on a spare drive just to see how it reacts and/or B: Ran a full surface scan of the drive using something like MHDD.
Biggest problem with intermittent problems is that if you can't reproduce them, you can't troubleshoot them.
I agree with everyone else also, the other tech took the easy route and reloaded. Definitely check the hours on the drive, just to be on the safe side, but odds are they did the job. Only time will tell now if it was a hardware or software problem!
 
I personally think that all the drive tests you did were correct. The other 'tech' simply thought of the quickest way of getting the repair completed.

Are you 100% certain that he did install a new drive?, and not simply nuked it?

When you go round, check the hours used on the drive, and you should be able to tell.
I agree 100% with this. More of a cop out it would seem. Blame the drive and re-install Windows. Unless it was a RAM or overheating issue it should pretty well guarantee to fix the issue, whether right or wrong in the diag.
 
OK - a cop out... but what is the alternative? Spend 10 hours diagnosing, not getting a proper diagnostic or getting conflicting diagnostic results, all the while losing face with the customer, only to have to eat crow and say, "well, I'm going to have to pass this to someone else..." possibly losing the customer forever.

OR, say, "something's not right, and although my tools and utilities tell me everything is fine with this HD, my experience tells me otherwise. Let's drop in a new drive and either fresh install or clone from the old drive, and be done in 2 hours". Everyone is happy, the new drive either solves the issue, and you will know right away, or it doesn't, and you'll know right away.

It is assertive action that is based on making solid decisions that yield the highest level of repair possible in the shortest amount of time (time is money, and my clients want it back next day, fixed). Yes, the cost is higher to the customer, but only slightly so. And, your cost to fix the computer is drastically lower, (2 hours vs. 10 hours).

This is a challenge we constantly face... and it is vital to ID that point in time where further tests and further efforts in fixing a system are wasted efforts and you are better off stopping and taking the "guaranteed" approach.
 
Last edited:
Had you made a sector-by-sector clone to another hard drive, you would have been able to easily determine if the issue was with the hard drive itself. After you get the clone, switch drives, boot off the clone and see if you get the same result. If it works, you know that it was the hard drive. If it doesn't work, it is the file system on the drive...or some other hardware issue (ie RAM or Video) that is causing the problem.
 
I haven't seen CHKDSK mentioned yet? A flakey hard drive can cause file corruption of critical system files and even though the drive may have miraculously corrected itself, the files will still be corrupted. So, even if the SMART tests passed this time, it's still worthwhile running CHKDSK and even SFC.
 
We use gsmartcontrol long test on all computers we repair unless it is in for a virus removal. We have had a few drives pass a full surface test along with good smart data and still be bad. It's rare, but it happens. We usually figure it out by trial and error with intuition.
 
I have to say, if I had to work as a 'no fix no fee' tech, I would have spent - max - 30 mins on diagnostics and then nuke and paved. I dont know how you could make money spending that sort of time on an 'unknown'. The result was an unhappy customer and a dent in your reputation *and* no profit. Whats your time limit on "free" diagnostics? If you dont have one, why not?
 
I use DFT. Drive Fitness Test. It does throw about an 8-12% false positives so we then use diagnostics from the Drive Manufacturer. Do we still have false positives. Perhaps, but the % is small. Also if WD says its bad who am I to argue.
 
We also do gsmartcontrol extended tests, but if it's still flaky and my gut tells me the drive is still at fault, I will run a test in mhdd and usually that can confirm it. I find mhdd significantly more reliable than smart tests.
 
We also do gsmartcontrol extended tests, but if it's still flaky and my gut tells me the drive is still at fault, I will run a test in mhdd and usually that can confirm it. I find mhdd significantly more reliable than smart tests.

I know mhdd can do lots of stuff. What tests/attributes do you use from it to diagnose a failing drive?
 
Back
Top