Getting stuck on an old wives tale

Topic

New Reply

Years and years ago I had a hard drive die.  I was working on a paper for college at the time and hadn’t made a backup. And of course, the computer ha
[See the full post at: Getting stuck on an old wives tale]

Susan Bradley Patch Lady/Prudent patcher

Reply | Quote

Replies

I do not recall having a drive die over the past decade. But then again, it is our policy to replace drives when they have been in service for three years.

But like you Susan, we keep spare SSD drives around as well as external hard drive enclosure cases.

We also have multiple backups of each of our PCs including:

• A clone of the C drive when a computer is taken out of its box and started up for the first time – before any updates or apps are installed or data transferred from the machine it is replacing,
• An Acronis True Image for Western Digital backup of the C drive to another internal drive at the close of business each day,
• An Acronis backup of the C drive to an external drive after the close of business on Fridays,
• Coping all data files from the C drive to an external drive using File Explorer after the close of business each day, and
• A monthly clone of the C drive to an external drive just before running Windows Updates.

The work stations we use for quantitative analysis are also backed up to external drives daily.

We have spare work stations on the shelf that mirror the configurations of the PCs we have in use. If a machine goes down all we need to do is pull one of the backups off the shelf and install the appropriate data files.

All software, product keys, recovery media, login information, etc. are stored in a tub dedicated to each computer and stored is a separate building on our campus.

And yes, I am working late on Saturday night/Sunday morning while four of our workstations are running our quantitative analysis programs – the results will be posted to our clients just after 7:30 AM.  The question comes to mind as to who backs me up if I experience a failure?

Reply | Quote

Wise advice from Susan. Drives eventually fail, and they do so unpredictably.

One way that I’ve found to inject a significant degree of predictability into the situation, is to use Hard Disk Sentinel. The program constantly monitors the health of my drives (HDD and SSD both) and gives a percent assessment of the drive’s performance and health. I’ve long since given up trying to make heads or tails out of S.M.A.R.T. data; the report HDS gives is much more straightforward.)

For example, on the PC where this is being typed, the storage HDD is shown at 100% Performance, but only 38% Health, with 40 weak sectors on the disk’s surface. As the number of bad sectors grows, the health drops. Thanks to this program, a replacement HDD stands ready to pop into place if and when the current disk’s health dips further.

Historically, I’ve replaced drives only when they were showing obvious signs of distress, such as slow file operations. But experience shows that by that point it could already be too late; Hard Disk Sentinel provides an earlier warning of approaching failure.

Keep Running Windows 7 Safely for Years to Come  
Make Windows 10 look and work like Windows 7

3 users thanked author for this post.

AlexEiffel, oldfry, SueW

Reply | Quote

At 38% I would have replaced it already.

cheers, Paul

2 users thanked author for this post.

wavy, Cybertooth

Reply | Quote

I have definitely been considering just going ahead and replacing that drive. However, Hard Disk Sentinel is still marking the drive as “Acceptable” and I’m curious to see what it would take it tip into non-acceptable range. You’re right though, 38% makes it sound awfully low already.

Keep Running Windows 7 Safely for Years to Come  
Make Windows 10 look and work like Windows 7

Reply | Quote

Paul T wrote:

At 38% I would have replaced it already.

And I would run it to failure, having current drive images at the ready to prevent any loss of data.  We all have our own reasons for doing the things that we do with our systems, we don’t need anyone’s approval, and we don’t all have to do the same things.

Always create a fresh drive image before making system changes/Windows updates; you may need to start over!

We were all once "Average Users". We all have our own reasons for doing the things that we do with our systems, we don't need anyone's approval, and we don't all have to do the same things.

Computer Specs

Reply | Quote

Cybertooth wrote:

Historically, I’ve replaced drives only when they were showing obvious signs of distress, such as slow file operations. But experience shows that by that point it could already be too late; Hard Disk Sentinel provides an earlier warning of approaching failure.

In this thread, I used SpinRite and SeaTools to extend the life of a HDD.  All of my SSD’s are Samsung, and I use Samsung Magician to monitor them.

Always create a fresh drive image before making system changes/Windows updates; you may need to start over!

We were all once "Average Users". We all have our own reasons for doing the things that we do with our systems, we don't need anyone's approval, and we don't all have to do the same things.

Computer Specs

2 users thanked author for this post.

oldfry, Cybertooth

Reply | Quote

I used the freezer trick once also. Not on my own drive, but friend of me got ‘mad’ at his laptop for some reason and slammed it on his desk. Was a laptop from work, and had some files he needed like really bad.

By the time he got to my house, drive had not been used for at least 30m.  I attached it via usb to my laptop and mounted it read only to no avail > ticking sounds we all know and fear.

Good 20m in the freezer, in a plastic bag after letting it cool to room temperature (not sure if needed, but I feared condensation would get it wet somehow) Then attach it again and look for the files. Got to locate them but ticking started again.. could not copy.  Prepared the copy commands in a .sh file and back to the freezer. After another 20m in the freezer, quickly attached it, mounted and copied the files he needed.

So yeah, I can only confirm this worked for me/him, but I wouldn’t recommend it for any large number of files or data. Also don’t slam your laptop on your desk, please

1 user thanked author for this post.

cyberSAR

Reply | Quote

Years ago I had a drive in here that did the same thing a couple times. Back in freezer, set thermostat down to get office as cool as possible. Set a bag of ice in ice chest and created a sling for the drive to sit about 1″ above ice bag. Hooked it up and was able to copy the drive without it going out again. Fun times!

Never Say Never

Reply | Quote

I use a laptop with M.2 Nvme SSD drive.
I don’t have a backup M.2 Nvme.
I create full X2 monthly image backups + daily incremental backups to external SSD.
I keep a copy of a full image backup on laptop’s drive D in case I’ll need to restore while away from home with my laptop.

Lenovo’s repair shop is not far away so a with a quick ride the laptop can be fixed.
I have an iPad and iPhone to keep me connected.

1 user thanked author for this post.

oldfry

Reply | Quote

I have had a few hard drives fail over the years. The most recent was in 2014 or 2015. It was a Seagate 500GB that was about five years old, and it had never shown any signs of imminent failure. No reallocated sectors or other such errors showing in SMART. It just quit during an installation of Windows. One minute working, the next deader than a doornail.

Hard drives do not always give you warning before they die.

I had another failure back in the day, of a 40 MB Seagate ST251-1 MFM drive that was in my first self-built PC. SMART didn’t exist yet, but when I got the first read error, I soon noticed that the bad sector count was growing. I don’t remember if I had to rescue any data, but if I did I would have done so right then, as the next step was to use Norton Disk Doctor (back when Norton was a must-have for PCs… they were all DOS tools) and do a surface scan. Each pass returned more errors, and while it never was bricked like the 500GB drive above, it was effectively quite dead. Fortunately, both of those Seagates were under warranty when they died. The 500GB had a month or two left of the 5 year warranty!

I remember having more than just these two, but the details of the others have faded in time. That one with the 40 MB drive was 33 years ago!

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon
XPG Xenia 15, i7-9750H/16GB & GTX1660ti, KDE Neon
Acer Swift Go 14, i5-1335U/16GB, KDE Neon (and Win 11 for maintenance)

Reply | Quote

[Berserker79]

Well, thankfully I only had one hard drive fail on me and it was many, many years ago, back when we had only mechanical HDDs, SATA didn’t exist yet and SSDs were more sci-fi than reality. I still remember the drive was a Fujitsu IDE drive (an early UDMA33 model and if that sounds unfamiliar to any of you it’s because that’s really old tech by now :)) with a weird 1.7 GB size. Lost everything on that drive back then, today I hope to avoid any kind of disastrous loss like that by creating regular backups should one my SSDs fail.

BTW, regarding “SSDs tend to not warn” about imminent failure, a while back I think it was Will Fastie who posted here about a little freeware tool called “ClearDiskInfo” that can read SMART data and allow a user to see with ease whether there is a risk of an imminent failure. While I guess there is no guarantee that the prediction may be accurate/reliable, it seems better than nothing and I always perform a check once in while to keep on eye on the health of my drives.

• This reply was modified 1 month, 3 weeks ago by Berserker79. Reason: Corrected typo, ooops :)

Reply | Quote

I don’t think I’ve ever had a hard drive fail as such, rather I’ve had the contents of a hard drive totally messed up by a Windows update (according to the technician who dealt with it for me). This was part of a home desktop with the only important stuff backed up so it didn’t cause me any major worries.

I’m intrigued by the reports of using a freezer to produce a temporary fix for an ailing hard drive. My main concern is always when the freezer fails, as I don’t have any backups for the frozen food! That actually happened just a few months ago, but happily family and neighbours came to the rescue and were able to look after the contents while I ordered the next day delivery of a replacement. It was a reminder that it isn’t just computers that fail, we all have a lot of other appliances that can be just as problematic when they fail and just as expensive to replace, if not more so.

Reply | Quote

Over the years I have had several HDD failures which I replaced and reloaded with an image of Windows and apps because I learned the importance of backups with the first failure. Most of the HDD failures occurred long ago when SSDs first came out…they weren’t very good then. I went thru 3 in 2 years. I learned that SSDs don’t make those grinding noises (usually bushings and bearings on motor) before they go out like platter HDDs often do. However, the SSDs that I had that failed all showed the same glitch before they died. All the SSDs would hang on a boot up; but total power off, wait, power on, and reboot would get it working again. With time, the frequency of hang boots would increase until it would not boot up again. At least, that was my experience with HDD failures.
BTW:
I keep all documents and other personal data on a separate drive from Windows OS drive. I take backup images of the Windows drive and backup up documents and other personal data with incremental copies on a separate drive. This allows access to individual files in the backup drive without having to load or access the entire personal data backup to retrieve one file.

HTH, Dana:))

Reply | Quote

0 hard drive have failed on mine.

My Windows 95 hard drive works fine except motherboard failed in 2014 after a lightning storm hit the power line. Using external case for it.

My Windows XP hard drive works fine.

My Windows 7 hard drive works fine.

I am not sure how people keep saying that hard drives fail. I have not had a single failure of a hard drive in over 30 years but I’ve only had 3 computers in my life time.

Reply | Quote

Jump on the Way Back Machine.

How many of you remember Steve Gibson’s SpinRite? I can’t even count on how many “drive failures” I fixed with that utility. They weren’t actually hardware failures but marginal sectors/tracks that it would bring back to life and move elsewhere on the drive. As drives got bigger and bigger it became less useful as the run times went through the roof. I still have a copy on a bootable USB drive. The last time I tried it on a 1Tb drive (USB 3.0 enclosure) the time estimate was 3 DAYS!

May the Forces of good computing be with you!

RG

PowerShell & VBA Rule!
Computer Specs

1 user thanked author for this post.

bbearren

Reply | Quote

It turns out that Steve is still working on SpinRite.  Version 1 appeared in 1987, the current version 6 appeared in 2004.  He is working on version 6.1 now (2023)

Reply | Quote

RetiredGeek wrote:

How many of you remember Steve Gibson’s SpinRite? I can’t even count on how many “drive failures” I fixed with that utility. They weren’t actually hardware failures but marginal sectors/tracks that it would bring back to life and move elsewhere on the drive. As drives got bigger and bigger it became less useful as the run times went through the roof. I still have a copy on a bootable USB drive.

AlphaCharlie wrote:

It turns out that Steve is still working on SpinRite.  Version 1 appeared in 1987, the current version 6 appeared in 2004.  He is working on version 6.1 now (2023)

Here are more details: SpinRite Development Roadmap. In particular, note SpinRite 6.1 will be a free upgrade to anyone who already owns v6.0.

with some limitations for USB & NVMe, SpinRite v6.1 will offer:

• Native AHCI and IDE support, with BIOS support fallback for USB, NVMe and others.
• Operates on the physical drive surface to provide compatibility with ANY drive format.
• Practical use on drives of =ANY= physical size. (~ 0.5 terabytes per hour)
• Support for drive advancements: Hybrid drives and large physical sector drives.
• Absolutely maximum possible performance – no missed revolutions.
• Support for (older) Apple Mac Intel hardware which supports Boot Camp / CSM.
• Downloadable as an .ISO for boot media prep on non-Windows systems.
• Diskette (floppy disc) support removed. SRv5.0 or v6.0 may be used for diskettes.
• Minimal changes will be made to SpinRite’s user interface so that its new functions can be provided as soon as possible. But the many changes will be very apparent in SpinRite’s performance and its ability to see much more deeply into its client drives.

The development version is reportedly even showing impressive benefits for SSDs.

In my own experience, SpinRite 6.0 became less useful as drives got larger because of some sort of divide overflow bug. FWIW, during 6.1 development the bug was identified and a patch was made available for 6.0 users.

3 users thanked author for this post.

Casey S, RetiredGeek, Cybertooth

Reply | Quote

I’ve had a number of HDD failures, but I’ve always been fully prepared with drive images.  One failure was so severe that the PC would not even POST.  There were four drives and I disconnected them one at a time until the machine booted.  Then I replaced that drive with a spare, restored its drive image, and all was well.

I had a house fire in 2011 that destroyed two PC’s but all my drive images were safe and, over the course of a few months (bought a Dell, built a DIY), both were fully restored.  On the other hand, I’ve never considered HDD anomalies (not outright failure) to be a deciding factor to toss a HDD.  There are tools that can extend its life and, with a strict regimen of drive imaging, loss of files is not an issue.

Always create a fresh drive image before making system changes/Windows updates; you may need to start over!

We were all once "Average Users". We all have our own reasons for doing the things that we do with our systems, we don't need anyone's approval, and we don't all have to do the same things.

Computer Specs

Reply | Quote

Just downloaded, installed, and ran the Western Digital Dashboad S.M.A.R.T. app. on 1 of our work stations.

Results – Drive Health = 99%

I guess I can sleep a little easier knowing the drive manufacturer’s software says all is well.

Reply | Quote

I’ve had several HDD “failures” over the last 2.5 decades, but in essentially every case, they gave some warning that not all was well ahead of time by starting to whine, changes to the sound of seeking, or starting to throw read or write errors. It seems there was always some sign of distress shown that would concern me enough to investigate with appropriate tools to check drive health.

I only had to use the freezer trick once, and it worked! I unplugged the “dead” drive cables (thankfully not the system drive) so I could use the system while the new drive I ordered came through the mail. Cloned existing system drive onto the newer, bigger, faster better one I’d just ordered, then prepped the old system drive as the receptacle for whatever could be saved from the dead/dying one once I had determined the cloning was successful and everything was in order.

It worked. It took two trips to the freezer as the drive warmed up and started glitching before I finished, but it worked with only a few files of little significance corrupted and lost.

I guess I’ve been lucky with SSDs. The only failure I’ve had was somewhat soft, in that the Intel branded 120 gig drive didn’t go suddenly dead, but developed what I came to call a “rotten spot”. Troubles were intermittent, the system would lock up, freeze or occasionally crash. Strangely the factory drive health monitoring tool said it was ok. Thinking it was a corrupted Windows install, I ordered a newer bigger SSD since prices had come down with the intent of passing the Intel SSD on to a secondary system still using an HDD as a boot drive. I learned better while backing up files off the Intel drive.

While the Intel software claimed the drive was fine, it also refused to secure erase it, claiming it was security locked or something. I know it hadn’t gone fully read only, since when I tried to use DBAN to wipe it, it would go to about 62% or so and then freeze the machine. I ended up secure wiping it by putting on a safety eye shield, grabbing my two biggest pair of Vice-Grips, and flexing the drive back and forth like some sort of exercise gadget until one piece became two. 💥

 

1 user thanked author for this post.

Cybertooth

Reply | Quote

On the small business network I manage, we’ve had two SSD’s within the past two years fail.  Both were from the same manufacturer, and installed in the same make/model of PC.  We utilize backup software that runs nightly and creates “bare metal recovery” bootable ISO / USB images.

In both instances, I was able to drop in a new SSD and recover from the previous night’s image.  Downtime was less than an hour if memory serves me correctly.  Users were ecstatic, as they were historically accustom to the IT guy having to rebuild and setup a workstation from scratch as the only backups in the past were the servers.

Reply | Quote

Ah, the days of shaking 450MB-1.2GB IDE hard drives to temporarily resolve “stiction” long enough to rescue their data off with Binary Resarch Ghost (Symantec didn’t always own it) 1.x off a boot floppy. I remember it. The place I worked also froze a few of those horrible Quantum Bigfoot drives (Making a 5.25″ drive to lower platter density yet retain capacity and save cost is a horrible idea when it turns out you need a much stronger motor and still can’t spin the drive fast enough for any performance), with varying levels of success.

Aside from those early days, my most recent failures have been awhile ago, largely because I’m only using hard disks now on my NAS at home (whose HGST 6TB enterprise drives are still chugging away). My last knowns were Seagate Cheetah server drives in the mid-twenty-teens (one model was notoriously unreliable, and often in servers that were poorly cooled which compounded it) so bad I created a script to identify client Dell servers that had them so we could proactively monitor and replace, and also Seagate SATA drives (1.5TB) that I used in my HP Mediasmart EX490 Windows Home Server I used in the late twenty-tens, which also had a design flaw. I’ve had a few first-gen 6TB WD Reds fail too, but like the previous drives, I had RAID, monitoring and scheduled consistency checking/data scrubbing to alert me, and a backup.

I don’t use hard drives in desktop systems any more, as SSDs from 512GB even up to 4TB are quite reasonably priced, and even if they have a fixed lifespan, I’ve had increased reliability and consistency. I kind of look forward to the day when I can build a NAS out of storage-grade SSDs for a reasonable price.

We are SysAdmins.
We walk in the wiring closets no others will enter.
We stand on the bridge, and no malware may pass.
We engage in support, we do not retreat.
We live for the LAN.
We die for the LAN.

Reply | Quote

AlphaCharlie wrote:

It turns out that Steve is still working on SpinRite.  Version 1 appeared in 1987, the current version 6 appeared in 2004.  He is working on version 6.1 now (2023)

For how many years has he been ‘working ‘ on that??

🍻

Just because you don't know where you are going doesn't mean any road will get you there.

Reply | Quote