Retrieve information from NVMe storage devices that stopped functioning under heavy load or stress.
Retrieve information from NVMe storage devices that stopped functioning under heavy load or stress.
My primary concern is trying to find a way to get data off my drives that have seemingly failed. The not so short version is I took some nvme drives out of another system and put them in updated system with new motherboard and CPU and after it was determined stable and it causes the system to shut down. Not just shut down like hey I crashed let me restart. Hard power cycled... You have to cycle the switch on the PSU to remove power from the Motherboard before it will power the Motherboard again.(It lights the power and reset button when off). I'm assuming there is a circuit protection device that resets upon power loss(circuit breaker resets on board). I have already installed the drives in the new system and made it to the OS and tried to copy paste etc. It will start but shortly after starting it will come to a crawl and then the system powers down. If I power cycle it to reset the motherboard it will not boot into windows. It shuts off before you can even get into bios and will repeatedly due that as long as your willing to reset it. I can let the drive sit for hours and it will work momentarily but do the same thing. Hard to recover 2 TB's worth of data with a drive that is flaky and works for maybe 30 seconds. I thought maybe if I could restrict the bandwidth, dial it back to pci-e 3, or in some way limit the HD so it wouldn't heat up as much, thus maybe work with some active cooling or something I might be able to recover the files. It's not worth sending off to a data recovery site unless it's just dirt cheap or something. Not sure if a disk cloning things on amazon that do it outside the OS might work better? I figure it won't be much different but at least with a external power source maybe it'll shield my new motherboard from failure. All of this is speculation of course as it could be a fault on the drive causing the chipset to heat up. All I can tell you is the HD 100% causes the system to fail. I hadn't even considered that I had 2 drives bad until I finished building the somewhat newer or updated system. I had beat my head against the wall for 3 days with the old system testing variables trying to narrow it down and assumed it was a motherboard problem and it might still be bad since 2 nvme drives that were on the same controller failed at the same time. Ok, The long version for those much smarter than me or just plain interested as it has nothing to due with my issue of data recovery. It's more for people who may search the web like me that might find a useful morsel in there that helps them. Also helps verify the issue with the data recovery though. My old system was perfectly stable meaning no random resets, errors, crash's of programs. Nothing unusual. I wanted to power my system up remotely with USB(keyboard or mouse). It wasn't a option in my Bios so I decided to check the bios page just to see if they added a feature. Didn't expect to but figured it was worth a click with the mouse to see. Upon landing on the Bios page for my motherboard I discovered that Asus had put out a bios update for my board for the Intel microcode. It was released like July 22nd and I was like welp let's do that anyway. Grabbed some updated drivers while I was there. Prepared the Bios USB stick, proceeded with the update and all was well. Booted up and decided to install the drivers. Shortly after I started getting some errors and crash's and just plain feeling unstable. I looked in the Event Viewer and noticed some programs having issues related to the drivers I just installed. Mumbled to myself and proceeded to uninstall said drivers and reboot. Then it get's interesting. It boot's to windows and a 20 minutes later it turns off and powers back up. Not paying much attention but just figured windows crashed. I had noticed the hard "click" in the background but didn't pay much mind to it. Then about 5 minutes after that it happened again. Then it proceeded to power down and try to restart. Sometimes it would post, get past the bios screen and start to load windows, sometimes it would just get to the part that says hit Del or F2 to enter bios. Then it just consistently wouldn't get to the enter bios screen. Basically system would power up.. display would have the screen saver active but not register the video yet, and it would power cycle again. At this time I assumed the bios I put in was the issue. I flashed bios again with the flash bios button on the motherboard with the USB stick in the correct port like you do when you don't have a compatible CPU. Made sure to roll the bios back to the same version I had previously and sure enough same problem. At this point I figure I'm screwed. All my parts are under warranty but we know how long that can take and I need my PC. I patiently plug through different scenarios over the next couple days and with just the CPU, Memory, and HD you couldn't get to bios. I ordered another Motherboard and CPU and I had already socketed the chip and installed the nvme's and the board was ready to go in. I decided oh what the hay. I hit the power button and bam.. there it sits on Bios. Put nvme back in and couldn't get it to post and would just reset constantly.. no HD and it goes to bios like it should... there was still a sata HD connected to the sata port BTW. I just mean it wouldn't post with the NVME's installed. I tried taking them out one at a time but nothing changed(I hadn't tried removing them all and just adding a different one each time. I hadn't realized I had more than 1 bad drive at the time. Both these drives had the OS/MBR on them so I was trying to boot with them. I never tried the 4 gig gen5 drive by itself because it didn't have an OS. Figured something on the board was bad related to HD's.. put them all in the new board and installed it. Powered up and boom, powered down immediately, AGAIN! Now to bring this picture into perspective. I'm currently running the same system just a new board and CPU. Hmm So I remove the two gen4 nvme's and leave the 4TB gen5 in it's dedicated slot. Everything is right in the world again. Install windows, install drivers, install used programs, install games. It's all chooching like it should. I ran it like that for 24 hours, with resets, stressing it in games etc. Feels 100% stable. Now i know one day does not make a system stable but 24 Hrs compared to 5 seconds is a heck of a lot MORE stable. Happy with the current status I grabbed my 2TB drive to grab some data off it. System boots up, start a file transfer and system powers down. Took me a second to figure out you had to kill the power at the PSU to get the board to reset(I recalled the clicking when it powered down and put 2 and 2 together). Apparently the old motherboard would just reset automatically and this one is... well.. different. There is pro's and cons to both as it's annoying to power it down and wait and then power back up.. but if it resets constantly it's just risking worse damage every time it turns it back on and at least when it makes you power down the board you KNOW something went wrong. So bad drive.. alright. Grab the other nvme drive.. the 1TB 990 Pro and guess what. Same exact thing. Tried it in a different spot... same thing. Only this one just wouldn't let it boot at all. This is where I figured out my mistake as I always left one of the gen4 drives in as it had the OS. So I had a 4TB 9100 Pro in it's slot operating off the CPU controller. I had two Gen4 drives the 1TB 990 Pro and the 2TB 980 Pro and both those were on the same controller. I know it's annoying somebody but I just hadn't got around to installing windows on the new drive yet. I reinstall windows every so often and it had only been in there a couple of months. Just hadn't got there yet. Now is where I draw assumptions. I'm guessing since two nvme's both failed at basically the same time that the only common factor they have is the motherboard/controller. So I'm going to assume the motherboard is bad as are the two related nvme drives. This is an intel system. I had ran the intel degradation test one of the times i was lucky enough to boot from a cold boot as soon as it started up. It passed. I ran memtest out of the bios as when I could manage to get into bios from again.. a cold start.. it was usually pretty stable there and memtest ran fine. It tested the ram and the cpu inside it and found no errors. I know this is far from comprehensive but I had to use what I had access to and some info is better than none. Now for the System in question Seasonic Vertex GX-1200W Asus Z690-E Gaming Wifi ---------> Replaced with Asrock X870E Taichi and 9800X3D Intel 13600K never overclocked limits enforced every since the Intel 13th Gen Fiasco. I never noticed any problems with mine but as soon as the first microcode update came out limits were enforced. Wasn't OC'ed before that. MSI Liquid Suprim 4090 --------- Used Seasonic 12 pin cable that came with PSU Corsair Vengeance 32GB CMK32GX5M2X6600C32 Samsung 4TB 9100 Pro, Samsung 1TB 990 Pro(Failed), Samsung 2TB 980 Pro(Failed), 4TB Sata SSD Backup. 360 AIO for CPU and of course graphics had it's own AIO. Usual run of the mill peripherals If anyone is interested I can post a short vid of the power cycling with the Q code display as I recorded it to try and decipher what was going on and where the process was stopping to troubleshoot
I question whether it would last long enough if it runs on Windows, perhaps something you can boot from to inspect it? It might only stay up for a minute before shutting everything down. I added Samsung Magician and planned to capture data during a cold boot, hoping to see reports and other details. I didn’t want to put too much strain on the drive, since that could cause it to shut down.
You wish to verify them at a secondary level, not the main one. Employing an enclosure/adapter or treating them as 'slaves' on your board.
If these drives are definitely bad, skip the hassle. Try recovering data with familiar techniques, but the signs suggest recovery chances are slim. Once you've checked all online resources without success, consider moving forward.