Live data from Hacker News

SSD will fail at 40k power-on hours (2021)

cisco.com

41–50 of 276 posts

Re: SSD will fail at 40k power-on hours (2021)

#41
post #9
post #7

Earlier quoted context omitted.

Wow, thanks for sharing. I didn't realize how closely related they were. (TLDR For anyone wondering, "recent HN issues" means HN very likely went down yesterday because of this same bug, when two (edit: two pairs, four total) enterprise SSDs with old firmware died after 40,000 hours close together. An admin of HN and its host both like this theory. See details in that thread.) Edit: If you want to discuss that theory…

the chance of two SSD's failing at the same time under normal circumstances is extremely slim. So this might actually be a good cause of this incident.

Especially since one pair was a nearly unused backup server that had a totally different use profile.

Re: SSD will fail at 40k power-on hours (2021)

#42

Quoted post unavailable.

It's a device to identify certain kinds of people who would have a problem with less loaded language without any loss in clarity.

I would love to see some examples of Cisco documentation that ever offended anyone.

Re: SSD will fail at 40k power-on hours (2021)

#43
post #33
post #28

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

Mine is above 53,000 hours ... time to check my backups!

Sounds 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!)

Re: SSD will fail at 40k power-on hours (2021)

#44

Quoted post unavailable.

My reaction was "If you want to write some documentation with bias-free language, just write the documentation with bias-free language." Why the need for a long paragraph explaining "Look how great and sensitive we are!" I understand, and agree with, the desire to use inclusive language, but so much of this has just devolved into performative nonsense. Edit: Thought I'd leave my original statement above up, but after…

Else you get questions, like, "why don't you say master/slave like everyone else?!@!!"

Re: SSD will fail at 40k power-on hours (2021)

#45

Earlier quoted context omitted.

It's a device to identify certain kinds of people who would have a problem with less loaded language without any loss in clarity.

I would love to see some examples of Cisco documentation that ever offended anyone.

I miss old Cisco documentation, with IP addresses and router names like SanJose3 and 408 phone numbers on PRIs etc.

Re: SSD will fail at 40k power-on hours (2021)

#47

Quoted post unavailable.

My reaction was "If you want to write some documentation with bias-free language, just write the documentation with bias-free language." Why the need for a long paragraph explaining "Look how great and sensitive we are!" I understand, and agree with, the desire to use inclusive language, but so much of this has just devolved into performative nonsense. Edit: Thought I'd leave my original statement above up, but after…

Saying that and usernames like "DoneWithAllThat" and "hn_throwaway", yeah, it checks out.

Re: SSD will fail at 40k power-on hours (2021)

#48
post #30

It's been over two years since this was first identified... since this apparently affected many makes and models of SSDs, it would be nice to know if my laptop could be affected and if there's anything I could do about it.

This will not affect your laptop, all of the models affected by this are enterprise SAS SSDs. Of course your SSD might have some other firmware bug that would eat your data, all you can do is search for the model number and see if the manufacturer has issued any notices/firmware updates.

> This will not affect your laptop

That’s just your presumptive opinion, right?

Edit: sorry, probably put that offensively. mikiem said about the HN drives: “These were made by SanDisk (SanDisk Optimus Lightning II) and the number of hours is between 39,984 and 40,032...” - https://news.ycombinator.com/item?id=32031428 Without knowing parts of a codebase are shared between SanDisk devices, it is hard to say that enterprise SAS devices have absolutely no code shared with consumer devices. So just the commenter’s opinion unless the commenter has knowledge of writing Sandisk firmware. “HPE and Dell both used the same upstream supplier (believed to be SanDisk) for SSD controllers” https://www.anandtech.com/show/15673/dell-hpe-updates-for-40...

Re: SSD will fail at 40k power-on hours (2021)

#49
post #3

Cisco is not a SSD manufacturer. They write industry-wide bug. Does that mean that more than one SSD manufacturer is affected (because they use partially the same firmware)? Further down they mention only Sandisk. Or is the industry-wide just their newspeak for saying any Sandisk of affected model, regardless whether installed in a Cisco box or somewhere else?

The problem seems to be widely experienced.

The Cisco report turned up in response to a post I'd made of the HN issue on the Fediverse:

https://mastodon.infra.de/@galaxis/108622795822100862

Re: SSD will fail at 40k power-on hours (2021)

#50
post #44

Earlier quoted context omitted.

My reaction was "If you want to write some documentation with bias-free language, just write the documentation with bias-free language." Why the need for a long paragraph explaining "Look how great and sensitive we are!" I understand, and agree with, the desire to use inclusive language, but so much of this has just devolved into performative nonsense. Edit: Thought I'd leave my original statement above up, but after…

Else you get questions, like, "why don't you say master/slave like everyone else?!@!!"

At this stage I think such questions can just be ignored.
Post reply on HN