Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Non-ECC memory corrupted my hard drive image [video]

youtube.com

Post reply on HN
Print view View original

31–40 of 161 posts

Re: Non-ECC memory corrupted my hard drive image [video]

#31

Post by _0w8t »

H8crilA wrote: post #14

This problem has no ultimate solution. I've seen all components flip bits, CPUs, networking cards, RAM, most often you just can't know for sure what did it. You can remedy it a bit (like with ECC), but ultimately there will always be corruption if you process hundreds of petabytes of data. Get used to it, your computer executes an instruction with a probability extremely close to 1, but not equal to 1. Deep in the ar…

One can process petabytes without bit flips if one use proper checksums and error correction codes. While that does imposes overhead, it is not big and, thanks to Shannon theorem, can be made arbitrary small with sufficiently big blocks.

Re: Non-ECC memory corrupted my hard drive image [video]

#32

Post by vladvasiliu »

eternityforest wrote: post #19
moloch-hai wrote: post #9

We don't have ECC mainly because Intel has long been hostile to "consumer" access to ECC. Apparently this was conceived as a market segmentation scheme: people outfitting servers could get ECC when they pay a huge premium. They would thereby not be tempted to cheap out and buy consumer-grade equipment, otherwise wholly adequate to meet all their needs at a radically cheaper price. That we cannot get laptops or even d…

I used to be a big fan of Intel, up until the latest chips from other companies that seem to have beat them on performance/watt. My next laptop will probably be AMD if the situation hasn't changed.

I'm happy with my AMD based laptop. But I haven't seen any that support ECC.

But I did see a Lenovo model, IIRC, that had some kind of Xeon and ECC. Not sure what the noise and battery life situations on that thing were, though.

Re: Non-ECC memory corrupted my hard drive image [video]

#33

Post by encryptluks2 »

I watched the full video. It was long but very informative. The humor at times made up for the length and the presenter showed a lot of deep knowledge that most people won't have. My biggest gripe is that they just didn't try replacing the RAM sticks in the first place. I get that they wanted to do a root cause analysis, but geez the time and patience they had to do all those memory tests. No wonder they did a video about it, cause otherwise that lost time would have been painful. I was baffled as well that dd and ddrescue work differently in how they utilize the RAM. Caught me offguard.

Onto the discussion of ECC RAM. In a perfect world, all memory would be ECC... but try finding some high performance 16GB sticks of ECC DDR4 RAM like what you'll see on gaming computers. I don't even think they make anything comparable in terms of speed and definitely not costs. I guess you don't really know that you needed ECC until it's too late.

Re: Non-ECC memory corrupted my hard drive image [video]

#34

Post by vladvasiliu »

dale_glass wrote: post #10

Yup, been there. Way back I had a Pentium 133 doing firewall duty in a closet. It did approximately nothing besides iptables, but of course any machine has logs, updates and so on going on. After running fine for months one day it suddenly died. I rebooted it. A few days later it died again. Another reboot. Then it died for the last time and failed to boot at all. Examination showed the disk was corrupt and couldn't…

> Since then I'm very paranoid, always run memtest on any new RAM I buy overnight, and have ECC where it's possible to have it.

Yeah, I do the same, but I've learned that you have to do it regularly.

In one of my desktop machines, the RAM ran fine for like two years. Then, all of sudden, random Firefox segfaults, etc.

Whipped up a memtest ISO, and sure enough, one of the sticks was bad.

Re: Non-ECC memory corrupted my hard drive image [video]

#35

Post by erk__ »

moloch-hai wrote: post #9

We don't have ECC mainly because Intel has long been hostile to "consumer" access to ECC. Apparently this was conceived as a market segmentation scheme: people outfitting servers could get ECC when they pay a huge premium. They would thereby not be tempted to cheap out and buy consumer-grade equipment, otherwise wholly adequate to meet all their needs at a radically cheaper price. That we cannot get laptops or even d…

> That we cannot get laptops or even desk machines with ECC

The Xeon series of laptop processors does support ECC just at a quite large premium.

Re: Non-ECC memory corrupted my hard drive image [video]

#36

Post by OJFord »

vladvasiliu wrote: post #32
eternityforest wrote: post #19

Earlier quoted context omitted.

I used to be a big fan of Intel, up until the latest chips from other companies that seem to have beat them on performance/watt. My next laptop will probably be AMD if the situation hasn't changed.

I'm happy with my AMD based laptop. But I haven't seen any that support ECC. But I did see a Lenovo model, IIRC, that had some kind of Xeon and ECC. Not sure what the noise and battery life situations on that thing were, though.

I realise that's blurrier when it comes to laptops, but AIUI it's more a case of whether the motherboard than supports it than about the AMD chip. i.e. given a desktop CPU, as far as I know you can put it in a motherboard that either does or does not support ECC RAM.

Re: Non-ECC memory corrupted my hard drive image [video]

#37

Post by morelikeborelax »

encryptluks2 wrote: post #33

I watched the full video. It was long but very informative. The humor at times made up for the length and the presenter showed a lot of deep knowledge that most people won't have. My biggest gripe is that they just didn't try replacing the RAM sticks in the first place. I get that they wanted to do a root cause analysis, but geez the time and patience they had to do all those memory tests. No wonder they did a video…

> " I guess you don't really know that you needed ECC until it's too late."

I spent many years on hardware consultation and was amazed at the all the times I had to explain it was just a what if insurance like any other things their business was mitigating against. Sometimes they'd even decided they needed to save costs in non-ecc ram when it was $4 a gb in difference, or (during the FB-DIMM era) there wasn't even an option to avoid it.

Never really understood the resistance towards it.

Maybe the lack of evidence before the Google study and people thinking RAM manufacturers were trying to rip them off or something.

The "never had a problem so why would I need" it attitude with no way to know if an issue was caused by a bit flip was most baffling.

Re: Non-ECC memory corrupted my hard drive image [video]

#38

Post by dale_glass »

vladvasiliu wrote: post #34
dale_glass wrote: post #10

Yup, been there. Way back I had a Pentium 133 doing firewall duty in a closet. It did approximately nothing besides iptables, but of course any machine has logs, updates and so on going on. After running fine for months one day it suddenly died. I rebooted it. A few days later it died again. Another reboot. Then it died for the last time and failed to boot at all. Examination showed the disk was corrupt and couldn't…

> Since then I'm very paranoid, always run memtest on any new RAM I buy overnight, and have ECC where it's possible to have it. Yeah, I do the same, but I've learned that you have to do it regularly. In one of my desktop machines, the RAM ran fine for like two years. Then, all of sudden, random Firefox segfaults, etc. Whipped up a memtest ISO, and sure enough, one of the sticks was bad.

That's the nice thing about ECC, it acts like an always running memory test.

You normally have a scrub time that can be configured in the BIOS, which also adds a regular verification of the entire RAM at regular intervals, just in case something goes wrong in some rarely used part of the memory.

Re: Non-ECC memory corrupted my hard drive image [video]

#39

Post by consp »

dale_glass wrote: post #10

Yup, been there. Way back I had a Pentium 133 doing firewall duty in a closet. It did approximately nothing besides iptables, but of course any machine has logs, updates and so on going on. After running fine for months one day it suddenly died. I rebooted it. A few days later it died again. Another reboot. Then it died for the last time and failed to boot at all. Examination showed the disk was corrupt and couldn't…

Badly socketed ram was one of the reasons my PC started failing after being on for a while. When everything was cool it was all fine, when the case and everything heated up a bit it failed eventually. Re-seating the ram fixed it and ran for quite a time without issues. This was in the early Athlon days though.

Re: Non-ECC memory corrupted my hard drive image [video]

#40

Post by eternityforest »

OJFord wrote: post #36
vladvasiliu wrote: post #32

Earlier quoted context omitted.

I'm happy with my AMD based laptop. But I haven't seen any that support ECC. But I did see a Lenovo model, IIRC, that had some kind of Xeon and ECC. Not sure what the noise and battery life situations on that thing were, though.

I realise that's blurrier when it comes to laptops, but AIUI it's more a case of whether the motherboard than supports it than about the AMD chip. i.e. given a desktop CPU, as far as I know you can put it in a motherboard that either does or does not support ECC RAM.

I'm surprised nobody has made RAM with the ECC logic built into the ram itself, that just looks like normal ram to the CPU.
Post reply on HN