Yes the dreaded ST3000DM001-1CH166. Terrible that one. We've had 17 out of 24 fail in a consumer NAS fail in less than 3 years. It's also a very widely deployed external drive since it's cheaper than WD and WD has its own external drive products.
0 failures on the spinning disk, but I will give it a few years.
You haven't used any recent Seagates then:D. When I look to my right I see two spinners being recovered with ddrescue:(. And it's been like that for the last 2 years. These large spinning consumer drives are just crap.
As I posted above, SMART 177 isn't a percentage, it's a P/E count. Documentation claims 2000 P/E cycles, independent testing shows close to 6,000. It's far less than even 1% used.
You're absolutely right. I wonder what other SMART values will be really important to SSD's. Where is blackblaze's SSD data. My guess is it will be even more vendor specific as it is now.
To be fair you've written ~950 GB and used 8%. If we simply extrapolate that you'll get ~12 TB or about 100 full drive writes. Personally I'm also dissapointed in the consistency of the TLC NAND drives. I get performance degradations on all my Samsung drives even though I've applied all the performance fixes. I can't fathom though why you are putting the smallest (slowest) and cheapest drives in your server. A Professional SSD (Sandisk or Intel) would be a lot more appropriate.
Mmm good thinking but you're assuming the swap partition is on the same drive as the system. While this is assumption is right most of the time, so is it that the system partitions (/,/usr,...), from which you'll be fetching that "new memory hungry application", also lie at the start of the drive.
I don't think that there's any evidence that the linux swapfile performs better - and in any case why would it being unfragmented be an advantage? Memory access is random, and so swapfile access is random, and so why does having it non-contiguous cause an issue?
That's exactly why you want a swap partition at the start of the drive where seek time is as low as possible. You don't want a bad situation (swapping) to become even worse (swapping and seeking all around the drive).
Fucknose why anyone would actually want that though.
I believe it was Alan Cox that uses a high swappiness (99) to keep interactive programs always fed with real memory. The idea is that the programs with low cputime (deamons,...) gets paged out before the interactive programs.
That's not true, they are not eliminated. There is plenty of crappy kernel mode software out there (drivers, filesystem filters). And the following (mostly hardware) problems also often cause BSOD's:
- Overheating GPU's.
- Broken stick of RAM, often due to dusty working environment, overheating or abuse.
- Bad Hard drive causing corrupt system files.
- ACPI Power issues (hibernate,suspend,...).
Now most of these problems do crash all systems. But personally I find the kernel panic and short stack trace on the console a lot more helpful than the BSOD (not counting the MEMORY.DMP or Minidump files since these need windbg).
Are the ways to abuse the system worse than the things the system is supposed to protect you against? I don't know. But it's a valid question - one that is likely to have different answers at different times and places. The answer may vary for different people too. For that reasons, bitcoin is a good thing. Whether the wild west mentality is better is one thing, but having the option of the wild west mentality can't really be bad.
What about spending ridiculous amounts of electricity to print your own money (mining) ?
Who wouldn't want to be able to carry around their own movie screen?
Hello ? Smartphone, tablet, laptop, portable DVD players,... I don't like wearing glasses. But what I really don't like is wearing bulky glasses. I'm pretty sure I'm not alone in this sentiment.
History is full of people saying things will fail that are now commonplace. How many people said the iPod or iPad would be a flop? People said the same about the GUI, homeless carriage, and any number of things.
And the line of vaporware wraps a few times around the earth.
VR combined with an automated (and reliable, working with a good percentage of the population) induction of an alternate mental state will lead to gaming that matches the reality levels of a lucid dream.
I think that's still pretty far in the future. We don't really understand the brain enough to do that. And even if we did there are legal, health and general safety issues that need to be explored beforehand.
I don't really think it's all about competence. If you enforce the use of "complicated tools" (GPG)
you'll be perceived as working against the company. And your job could be on the chopping block next quarter. Also while I think password hashing and salting should be mandatory I've received numerous calls from higher-ups wanting to know their "forgotten passwords". So even that can that can be seen as you working against the bosses:(.
Presumably when you are very far away and invisible to the enemy, you fly back to where ever you came from, reload, and repeat until the enemy has no planes left. I think this is generally the idea behind the whole F-35 concept.
Yeah, but once radar tech catches up you lose that tactic.
Planes don't have an infinite amount of AAM missiles. At some point in time they will have to use their gatling guns. And quick bombing raids without additional air support are supposed to be the JSF's forte IIRC.
Slashdot Mirror
So poison the oceans, suffocate the atmosphere, irradiate the land because in a few decades I won't be around anyway.
Selfish religious assholes !
But you are equipped to put forward your prescient feeling of a mini ice age coming really really soon as something all of us should plan for ?
I wonder if it is possible to start a class action lawsuit ?
Yes the dreaded ST3000DM001-1CH166. Terrible that one. We've had 17 out of 24 fail in a consumer NAS fail in less than 3 years. It's also a very widely deployed external drive since it's cheaper than WD and WD has its own external drive products.
You haven't used any recent Seagates then :D. When I look to my right I see two spinners being recovered with ddrescue :(.
And it's been like that for the last 2 years. These large spinning consumer drives are just crap.
Have you tried different brands ?
You're absolutely right. I wonder what other SMART values will be really important to SSD's.
Where is blackblaze's SSD data. My guess is it will be even more vendor specific as it is now.
Right that value represent the raw maximum erase cycles (~3.000 cycles TLC, ~10.000 MLC, ...)
Let me quote wikipedia:
Now my own raid0 drives:
To be fair you've written ~950 GB and used 8%. If we simply extrapolate that you'll get ~12 TB or about 100 full drive writes.
Personally I'm also dissapointed in the consistency of the TLC NAND drives.
I get performance degradations on all my Samsung drives even though I've applied all the performance fixes.
I can't fathom though why you are putting the smallest (slowest) and cheapest drives in your server.
A Professional SSD (Sandisk or Intel) would be a lot more appropriate.
Mmm good thinking but you're assuming the swap partition is on the same drive as the system. /usr, ...), from which you'll be fetching that "new memory hungry application", also lie at the start of the drive.
While this is assumption is right most of the time, so is it that the system partitions (/,
That's exactly why you want a swap partition at the start of the drive where seek time is as low as possible.
You don't want a bad situation (swapping) to become even worse (swapping and seeking all around the drive).
Apparently It could have been Andrew Morton (https://lkml.org/lkml/2004/4/28/263).
I believe it was Alan Cox that uses a high swappiness (99) to keep interactive programs always fed with real memory. ...) gets paged out before the interactive programs.
The idea is that the programs with low cputime (deamons,
That's not true, they are not eliminated. There is plenty of crappy kernel mode software out there (drivers, filesystem filters).
And the following (mostly hardware) problems also often cause BSOD's:
- Overheating GPU's.
- Broken stick of RAM, often due to dusty working environment, overheating
or abuse.
- Bad Hard drive causing corrupt system files.
- ACPI Power issues (hibernate,suspend,...).
Now most of these problems do crash all systems.
But personally I find the kernel panic and short stack trace on the console a lot more helpful
than the BSOD (not counting the MEMORY.DMP or Minidump files since these need windbg).
Actually it's not paper but rope.
What about spending ridiculous amounts of electricity to print your own money (mining) ?
What about rickshaw businesses run by homeless people ?
Hello ? Smartphone, tablet, laptop, portable DVD players, ...
I don't like wearing glasses. But what I really don't like is wearing bulky glasses. I'm pretty sure I'm not alone in this sentiment.
And the line of vaporware wraps a few times around the earth.
I think that's still pretty far in the future. We don't really understand the brain enough to do that.
And even if we did there are legal, health and general safety issues that need to be explored beforehand.
At least he has a sense of humor.
I don't really think it's all about competence. If you enforce the use of "complicated tools" (GPG) :(.
you'll be perceived as working against the company. And your job could be on the chopping block next quarter.
Also while I think password hashing and salting should be mandatory I've received numerous calls from higher-ups wanting to know their "forgotten passwords".
So even that can that can be seen as you working against the bosses
Yeah, but once radar tech catches up you lose that tactic.
Now they are trying another tactic: having international politicians lobby the rest of the world to buy them.
Planes don't have an infinite amount of AAM missiles. At some point in time they will have to use their gatling guns.
And quick bombing raids without additional air support are supposed to be the JSF's forte IIRC.
Sad but true. If you want to get taken seriously you need to put your custom on.