5 ms·
Check your power-on hours: $ sudo smartctl -a /dev/sda | grep -e Power_On_Hours -e ^ID ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE
by lucb1e 4y ago
Check your power-on hours:
$ sudo smartctl -a /dev/sda | grep -e Power_On_Hours -e ^ID
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
9 Power_On_Hours 0x0032 098 098 000 Old_age Always - 9743
Just looking at the raw value, it seems to be 9'743 hours in my case
- borplk 4y agoMine is above 53,000 hours ... time to check my backups!
- lucb1e 4y agoSounds like you're in the clear for this particular bug... ...but always check your backups regularly for data that is dear to you! Protip of the day: that includes things on someone else's server. I remember when Grooveshark went offline from one day to the next and I lost nearly my whole library because I remembered only some artists and had to go through thousands of songs to find which ones I actually liked from them. My browser's localStorage object had the playlists but I didn't use those much. Or when 000webhost cancelled my account because I was using the 100MB(?) to back up some files that were most important to me, rather than for actual webhosting (in my defense, I was 15 at the time), and so when I returned from a holiday with my parents with an actual crashed hard drive, that turned double sour. Backing up things from what they now call the "cloud" is something I learned early, as I have virtually no code I wrote before that summer, only some of the music, only essays with WordArt if they were printed, etc. If you use Telegram, Spotify, Netflix, maybe you have videos uploaded to YouTube and not have local copies anymore... my recommendation is to have backups of things that are important to you, onto a medium that you own (it's only a disaster-case copy anyway), and for copyrighted content like Spotify/Netflix it would simply be enough to just have a list of songs/videos. Maybe Netflix doesn't go offline from one day to the next, but your account might be hacked, or that friend you share it with might be hacked, or the right set of hard disks fail at their datacenter, etc. GDPR data exports are your friend, particularly when they're automated and you don't have to bother support. (They might also reveal, as in my case, that Spotify knows how often you shower because I then connect it to a particular waterproof bluetooth speaker at wakeup time. Data exports are also fun to browse!)
- Pakdef 4y agoI had to adapt a bit: sudo pacman -S gsmartcontrol sudo smartctl -a /dev/nvme0n1p4 | grep -e "Power On" Im at only 2727 hours
- wutbrodo 4y agoJust FYI for anyone for whom this didn't work by default: I needed to use the --all flag with smartctl (and install smartmontools if you don't have it).
- lucb1e 4y agoThe -a flag from my example should be an alias for --all (man smartctl | grep -A1 ' -a' | head -2). Is that not the case in your version?
- wutbrodo 4y agoDerp you are correct, sorry. Hopefully my comment is still useful for anyone who didn't know they needed smartmontools
- climb_stealth 4y agoPro tip: when writing out commands for people to read it helps to use the long form arguments. In this case passing '--all' instead of '-a' to smartctl. It makes it easier to read and more clear what specific options do. Same with calling things in scripts. Short form is for quick and dirty typing things, but not great for reading or comprehension :)
- lucb1e 4y agoFair point, yes, I should have done that!
- jeffbee 4y agoChecked my Samsung 970 Evo 2TB and it says 487 even though it’s been on continuously for years.
- colejohnson66 4y agoReceiving power isn’t the same as being “on”. I’d assume the drive has a sleep state that it goes into after inactivity, and those hours don’t count as “power on” hours.
- jeffbee 4y agoInteresting, thanks. The spec does leave it almost uselessly underspecified: "Power On Hours: Contains the number of power-on hours. This may not include time that the controller was powered and in a non-operational power state." The same drive reports only 329 "controller busy time" minutes.
- rootw0rm 4y agoMy 960 Pro 1TB, which I paid way too much for back in the day is at just shy of 5,000 hours, and I've used the hell out of it since it was new.
- Bilal_io 4y ago960 pro was released in October 2016, even if we assume you purchased it the same month it was released, and you have been running it 24/7 non-stop, that's 50,591 hours counting from October 1st to this hour. I can assume the number is way below 1/3 of that even for people that use it 8 hours a day.
- freebreakfast 4y agoiMac mid-2010. Original disk drive. "9 Power_On_Hours 0x0032 001 001 000 Old_age Always - 74233" 12 years old. More than 8 years of run time. It keeps on purring. Yes, I have redundant backups. I also have a replacement drive ready. I just want to see how far I can take it.
- dylan604 4y agoLuckily it is old enough where you can replace it
- synergy20 4y agoID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 9 Power_On_Hours 0x0032 055 055 000 Old_age Always - 39676 I'm 300 hours from 40K, time to buy new SSD? is this real?!
- JohnBooty 4y agoI mean, is it the affected model? Have you applied the appropriate firmware update?
- synergy20 4y agoI did not find the model info, but thanks!
- yencabulator 4y agosmartctl -i /dev/sda
- lucb1e 4y ago0. Have backups 1. Check your backups
- nomel 4y agoThe two above rules are true regardless of hours on your storage, make, model, or technology. ;)
- adamredwoods 4y agohttps://www.smartmontools.org/ https://www.smartmontools.org/
- anshumankmr 4y agoHow to do this on a Windows PC?
- nodja 4y agoIn an elevated powershell prompt Get-PhysicalDisk | Get-StorageReliabilityCounter | Select-Object PowerOnHours
- anshumankmr 4y agogetting a blank value
- vectorcrumb 4y agoI'm getting a blank value back as well, even with admin privileges.
- kuroguro 4y agoGetting blanks on some of my disks. Two show 64422h and 73318h with no signs of failing :)
- tencentshill 4y ago
- underscore_ku 4y ago
- zh3 4y agoI seem to have the world's oldest SSD (or am I misinterpreting the output?) (shell 1) ~# smartctl -a /dev/sda | grep -e Power_On_Hours -e ^ID ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 9 Power_On_Hours_and_Msec 0x0032 000 000 000 Old_age Always - 933932h+27m+33.940s
- bo0tzz 4y agoThat would make for 106 years of power on time, so it's probably not right...
- user8139471 4y agoThat is one of the most careful uses of "probably" that I have ever seen.
- eps 4y agoMore often than not SMART attributes are completely undocumented and the interpretation of their raw values is a pure guesswork on smartctl devs' part. For your SSD make/model they just have it wrong.
- duckmysick 4y agoFellow time traveller?