|
Moey posted:I have used "WD Drive utilities" before shucking for the past 8 easystore drives. Run a full scan on them (took like 28 hours maybe) per drive with my most recent 14tb disks. I think this is going to be what I start with, because this is mostly a "failed to resilver" when I added them to a pool situation. Along with the other one that just wouldn't spin up at all right out of the box. 3 of them are working and have been part of the pool for weeks, so.....yay? But I'd like to finish replacing the 5-6 year old drives in this pool and have a known good spare on hand and be done with this project. I also got the 10 Gbe stuff I had asked about previously and need to shut everything down and install cards.
|
#
?
Feb 4, 2022 22:45
|
|
|
|
| # ? Jun 10, 2024 15:01 |
|
|
Another thing to keep in mind is that hard drives have an optimal temperature operating range because they’re engineered to perform in those temperatures. There’s a ton of stories of how old servers running for 6 years get moved but upon boot-up 10% of the fleet has dead drives - it’s not only because of damage in transit.
|
|
#
?
Feb 4, 2022 22:47
|
|
|
Sir Bobert Fishbone posted:Is there a way to suppress this error and keep smartd/smartctl from yelling at me? Try sudo /usr/sbin/update-smart-drivedb to update the smartmontools database, which is independent of actual software updates. Failing that, their docs say to do smartctl -s on /dev/sdX to turn on smart monitoring for a particular drive X. So I'd just do off to disable it for that drive only. Possibly come back to this in the future to re-enable it, because these problems are generally because the smart tool doesn't have the accurate decoding info for that drive in its database. So a future update will probably cover the drive and produce accurate information.
|
|
#
?
Feb 4, 2022 22:49
|
|
|
This is the dumbest thing, but I got a low disk space warning from my Synology and it made me realize that on the last five years I never figured out how to see how much free space is left on the volume from a pc. Without the rigmarole of logging into the thing that is. Is there a way to expose that information in explorer or anything? Feels like something you could reasonably want to easily keep an eye on.
|
|
#
?
Feb 4, 2022 23:13
|
|
|
Flipperwaldt posted:This is the dumbest thing, but I got a low disk space warning from my Synology and it made me realize that on the last five years I never figured out how to see how much free space is left on the volume from a pc. Without the rigmarole of logging into the thing that is. Is there a way to expose that information in explorer or anything? Feels like something you could reasonably want to easily keep an eye on.  That's my Linux server but anything mounted over SMB looks the same no matter what's hosting it. Note that if there are any kind of volume management shenanigans involved this number may be wonky, when I first switched to using multiple datasets in ZFS I had to tell Samba to use a special script to figure out free space because things were no longer straightforward, but I'd imagine any mainstream NAS platform handles that automatically. wolrah fucked around with this message at 23:42 on Feb 4, 2022 |
|
#
?
Feb 4, 2022 23:40
|
|
|
Sir Bobert Fishbone posted:I have what looks on its face to be the dumbest SMART error. This is the smartctl output for a 1TB Gigabyte nvme drive on my Proxmox server, which has begun emailing me every day that the drive is failing. The room that box is in is somewhat chilly, sure, but certainly not as cold as the drive seems to think it is. Smashmouth, [h]eat the drive
|
|
#
?
Feb 5, 2022 00:00
|
|
|
wolrah posted:Mount it as a mapped drive on your PC?
|
|
#
?
Feb 5, 2022 00:37
|
|
|
If anyone is avoiding shucking, this seems like a solid deal on WD Gold 16tb (2 for $540). https://slickdeals.net/f/15605173-2x-wd-gold-16tb-wd161kryz-for-540-or-less
|
|
#
?
Feb 5, 2022 10:20
|
|
|
nevermind
|
|
#
?
Feb 5, 2022 16:40
|
|
|
Moey posted:If anyone is avoiding shucking, this seems like a solid deal on WD Gold 16tb (2 for $540). I was too slow on this, it is now expired. It would have been a good upgrade from the 2x8tb drives in my NAS.
|
|
#
?
Feb 5, 2022 20:04
|
|
|
Smashing Link posted:What is everyone's favorite versioned backup software + platform? I went through a big eval a year or so ago and settled on Duplicacy + Backblaze with some local external disk backups for good measure. After running it a while w/ cron I bit the bullet and paid for the web app version (quite cheap in the grand scheme of things). It's awesome running in Docker and makes scheduling, management, etc. so simple. Setting it up w/ RSA for Backblaze was a piece of cake. I just wish I had done it sooner rather than put up with CrashPlan's BS for way too long.
|
|
#
?
Feb 5, 2022 20:37
|
|
|
Moey posted:If anyone is avoiding shucking, this seems like a solid deal on WD Gold 16tb (2 for $540). Back in stock, but coupons are no longer stacking, leaving the final price at $600 for 2 WD Gold 16TB drives. It hurts my soul spending this much money on 1s and 0s, but those linux isos aren't going to store themselves.
|
|
#
?
Feb 6, 2022 09:23
|
|
|
SuitcasePimp posted:I went through a big eval a year or so ago and settled on Duplicacy + Backblaze with some local external disk backups for good measure. After running it a while w/ cron I bit the bullet and paid for the web app version (quite cheap in the grand scheme of things). It's awesome running in Docker and makes scheduling, management, etc. so simple. The fact that there is a paid version is actually somewhat reassuring to me as it suggests hey will continue to maintain things. I will have to compare with Borg and report back.
|
|
#
?
Feb 7, 2022 00:11
|
|
|
just learned after installing my new 4TB WD Black drive that WD has implemented a new feature called Preventive Wear Leveling (PWL) that causes drives to have an audible click every 5 seconds. as cool as that is, its going to drive me up the loving wall, are there any manufacturers of HDDs who don't implement technology like this?
|
|
#
?
Feb 7, 2022 04:21
|
|
|
Too Many Birds posted:just learned after installing my new 4TB WD Black drive that WD has implemented a new feature called Preventive Wear Leveling (PWL) that causes drives to have an audible click every 5 seconds. I'm assuming you're using this in a desktop rather than a NAS, which is why you're irritated by the noise? You might want to look at 5400 RPM drives instead. That is, if you can put everything that cares about speed on SSDs, and the HDD is just 4TB of single-drive whatever storage. They're quieter, and with area density so high on modern drives they have surprisingly decent raw transfer speed. Seeks are bad of course, and most of 'em are SMR so they absolutely suck for RAID / NAS / etc. But as a place to dump media files on a desktop they work fine.
|
|
#
?
Feb 7, 2022 05:24
|
|
|
yeah its in a desktop, i wanted something 7200 RPM for HDD since i'm working with a lot of bigger photo files in Lightroom/Photoshop on a regular basis, and i like switching between them with little to no lag, which i think is the seeking speed boost that i want.
|
|
#
?
Feb 7, 2022 05:31
|
|
|
Too Many Birds posted:just learned after installing my new 4TB WD Black drive that WD has implemented a new feature called Preventive Wear Leveling (PWL) that causes drives to have an audible click every 5 seconds.
|
|
#
?
Feb 7, 2022 10:10
|
|
|
my mom's external drive died. we did the usual checks, like trying it on another machine, to no avail, and trying a different enclosure, also to no avail. she now wants to send it off to a hard drive recovery company. any recommendations? i do not know a single thing about doing this.
|
|
#
?
Feb 7, 2022 16:17
|
|
|
abelwingnut posted:my mom's external drive died. we did the usual checks, like trying it on another machine, to no avail, and trying a different enclosure, also to no avail. she now wants to send it off to a hard drive recovery company. any recommendations? i do not know a single thing about doing this. Are the contents of the drive worth multiple thousands of dollars? That’s how much you’d likely be paying
|
|
#
?
Feb 7, 2022 16:21
|
|
|
Possibly as low as hundreds rather than thousands, but yeah. Data recovery is expensive. And anyone who promises you a price up front should be avoided -- the reputable companies only give you the price once they have the drive in their hands and can see how bad it is. They charge by the hour, and if they have to do something like surgically extract the platters to read them that's skilled clean-room work. OTOH most of them do free evaluations so that's fine. Ontrack and DriveSavers are both good and have low-ish starting minimums, so if you just have a controller failure that can be easily fixed by swapping out the controller board, you might get away for relatively cheap.
|
|
#
?
Feb 7, 2022 17:16
|
|
|
Controller swaps still work? I thought there's unique calibration data stored in each controller nowadays, based on factory testing/burn-in?
|
|
#
?
Feb 7, 2022 17:40
|
|
|
Buff Hardback posted:Are the contents of the drive worth multiple thousands of dollars? yea, i told her the process could easily cost a few thousand. she's still willing to talk to some companies. Klyith posted:Possibly as low as hundreds rather than thousands, but yeah. Data recovery is expensive. we'll give drivesavers a call, thanks.
|
|
#
?
Feb 7, 2022 19:07
|
|
|
I had a shucked Easystore fail on me prematurely recently. The drive was like 6 months of use when it started to throw SMART errors. It was 1 of 2 in my 218+, so I replaced the drive and that's all solid. Once that was runjning I sent the bad one into WD. I didn't have the shell or anything left and figured gently caress it. Well today (about 1 month later) I received a replacement Easystore! I'm on Ubuntu with my main desktop, what should I do to try and stress the drive to make sure this one is good? I'm going to upgrade to a 4 bay and I'll use this 3rd 14t in there. So I just want to plan ahead and do what I can to check the drive.
|
|
#
?
Feb 8, 2022 17:27
|
|
|
Yeah, if I had a drive fail I'd probably just shuck another Easystore to replace it asap and send the bare failed drive in to see if they'd warranty it. No harm in trying, worst they'll say is "no". Then I'd have a spare drive anyway.
|
|
#
?
Feb 8, 2022 17:46
|
|
|
In my WD red saga, the NewEgg return won't even read SMART data using WD Dashboard. Tried an "erase" and its kerchunking/powering down and back up again. The extra one I ordered doesn't power up, just like the original I returned to NewEgg. Two of the other ones that are now in service are throwing: Device: /dev/da16 [SAT], failed to read SMART Attribute Data. Device: /dev/da17 [SAT], failed to read SMART Attribute Data. So maybe one of them is good? It's past the return window so I opened a case with WD on the first two I mentioned and included the issues I'm having with the rest. This has been really, really stupid. It shouldn't be this hard. I'm hoping the ID the serial numbers as a bad batch or something and do the right thing, which would be an advance return so I can test/swap them before sending them back for an RMA.
|
|
#
?
Feb 8, 2022 17:56
|
|
|
deong posted:I'm on Ubuntu with my main desktop, what should I do to try and stress the drive to make sure this one is good? I'm going to upgrade to a 4 bay and I'll use this 3rd 14t in there. So I just want to plan ahead and do what I can to check the drive. If you have time and don't care about any data on the disk, badblocks seems to be the most complete way to beat up the drive. It will write and verify four full patterns of bytes, AA, 55, FF, and 00, to the disk. The one of my 12TB disks that I started a week ago is done, the other three that I started scanning Friday evening are all in the middle of writing the "FF" phase. so they're a bit past half way while having been effectively maxing the disks out the whole time. My post on the last page has an example of the command I used which is needed for any larger disks as the default block size causes a 32 bit issue, obviously change out /dev/sda for the correct drive path. wolrah fucked around with this message at 20:38 on Feb 8, 2022 |
|
#
?
Feb 8, 2022 20:36
|
|
|
wolrah posted:If you have time and don't care about any data on the disk, badblocks seems to be the most complete way to beat up the drive. It will write and verify four full patterns of bytes, AA, 55, FF, and 00, to the disk. The one of my 12TB disks that I started a week ago is done, the other three that I started scanning Friday evening are all in the middle of writing the "FF" phase. so they're a bit past half way while having been effectively maxing the disks out the whole time. awesome. That's what I was seeing as well. This drive is a spare until I upgrade the synology; so no data etc. Thanks
|
|
#
?
Feb 8, 2022 22:35
|
|
|
Is it possible to clone a TrueNAS OS drive to a new drive and just keep plugging along as usual? I'm just asking with an eye toward the future. I've got a Fractal Node 304 coming in and will be populating it, at first, with 2x10tb, 2x16tb, and an old SATA SSD (X25M-G2 160gb). However, my ITX board only has 6 SATA slots, so if I want to expand down the line (not sure I will as that's a shitload of data but who knows, I'm sure (like physical space) the more I have the more garbage I'll fill it up with like offline GOG game installers or what have you) then I can just clone the TrueNAS installation off the SATA SSD onto an M2 to put into my motherboards single M2 slot, right? Then just boot off that M2 SSD and everything gonna Just Work. E: As a second question, what are the thoughts on the Toshiba Enterprise MG drives? They are by far the cheapest in my region -- I'm in S.E.A. -- coming at $15.8/TB for the 16tb drives when all other competitors are $17.5+/TB. Shrimp or Shrimps fucked around with this message at 02:18 on Feb 9, 2022 |
|
#
?
Feb 9, 2022 02:05
|
|
|
Shrimp or Shrimps posted:Is it possible to clone a TrueNAS OS drive to a new drive and just keep plugging along as usual? Maybe but I'm not sure you want to solve this issue in that way. I've had to recover from a lovely USB boot drive dying on a TrueNAS reboot that didn't work. I put in a new USB drive with a stock TrueNAS on it and restored my config backup. Worked perfectly. Just no drama at all.
|
|
#
?
Feb 9, 2022 04:13
|
|
|
Motronic posted:Maybe but I'm not sure you want to solve this issue in that way. Thanks for the reply. So what you're saying is I'll just be able to make a config backup and then when it comes time to switch boot drives, I just do a restore? Basically I just want the option to switch boot drives in the future without losing any data on the actual data disks.
|
|
#
?
Feb 9, 2022 04:32
|
|
|
Shrimp or Shrimps posted:Thanks for the reply. So what you're saying is I'll just be able to make a config backup and then when it comes time to switch boot drives, I just do a restore? Basically I just want the option to switch boot drives in the future without losing any data on the actual data disks. That's been my experience, but please do shop for others. I only use TrueNAS at home, I'm used to (often very much shittier) "enterprise solutions". My boot drive(s) have no data, just the OS and config. So when I made a new boot USB and put the config on it all data drives/pools just magiced right on back up for me. I've had to deal with this exactly one time. So that's your sample size unless you get other opinions.
|
|
#
?
Feb 9, 2022 04:47
|
|
|
So my big 10 disk RaidZ2 array of 8TB Ironwolf drives is finally starting to age out. 1 disk failed outright and 3 more are on the SMART 'you....should think about replacing me soon' pre-fail list. Time to slowly Ship of Theseus the array over the next 2 years as more and more of these drives slowly die off. To be entirely fair, they're like 8 years old now, so they had a hell of a run. I do have to say though, OmniOS and Napp-it makes doing a DIY SAN out of whitebox crap really easy to set up and monitor, and with almost 400 days of uptime before I had to reboot it, it basically just doesn't ever quit.
|
|
#
?
Feb 9, 2022 04:55
|
|
|
Since zfs raidz expansion is ready to be tested broadly, I think everyone in vicinity of this post with spare equipment should throw everything including the software equivalent of the kitchen-sink at it.
|
|
#
?
Feb 9, 2022 09:37
|
|
|
BlankSystemDaemon posted:Since zfs raidz expansion is ready to be tested broadly, I think everyone in vicinity of this post with spare equipment should throw everything including the software equivalent of the kitchen-sink at it. Oh, that's neat. Shame I don't currently have hardware spare to mess with that.
|
|
#
?
Feb 9, 2022 14:52
|
|
|
I think it's time to built a NAS. To start with, I'm just going to buy two 8TB drives, and then install them onto my Ubuntu system using RAID1 (redundancy). Eventually, I'm going to expand it out to more drives and it's own enclosure, but I'll deal with that when the time comes. It'll take me a while to fill up 8TB. When that day comes, will I be able to move two RAID1 drives from one system and place them onto another system and have all the data remain intact? The drive I'm probably going get are these: https://www.newegg.com/seagate-barracuda-st8000dm004-8tb/p/N82E16822183793?quicklink=true Also, what filesystem should I use for these? I assume EXT3? Would NTFS be better? I already have Samba installed, I'm just going to symlink the mountpoint of the RAID array into my Samba share folder.
|
|
#
?
Feb 10, 2022 03:05
|
|
|
I agree with your first step. If you are into this stuff as a hobby at all you'll find the setup insufficient and bad. When that happens, or when you fill your 8GB you can simply buy a pair or more of bigger drives and a dedicated NAS/better PC and copy the existing system onto it. You can archive your data onto the two resulting 8GB drives any way you like and keep them in a box.
|
|
#
?
Feb 10, 2022 07:29
|
|
|
School of How posted:I think it's time to built a NAS. To start with, I'm just going to buy two 8TB drives, and then install them onto my Ubuntu system using RAID1 (redundancy). Reminder: raid is not a backup. School of How posted:When that day comes, will I be able to move two RAID1 drives from one system and place them onto another system and have all the data remain intact? Move them into a commercial NAS box? No. School of How posted:Also, what filesystem should I use for these? I assume EXT3? Would NTFS be better? I already have Samba installed, I'm just going to symlink the mountpoint of the RAID array into my Samba share folder.
|
|
#
?
Feb 10, 2022 14:54
|
|
|
Isn't XFS nowadays' love child for Linux-based NAS applications (at least when you don't want to use OpenZFS)?
|
|
#
?
Feb 10, 2022 15:50
|
|
|
XFS doesn't really bring that much to the table compared to ext4, but Red Hat pushes it heavily. Neither is especially good for the NAS use case.
|
|
#
?
Feb 10, 2022 16:20
|
|
|
|
| # ? Jun 10, 2024 15:01 |
|
|
Klyith posted:are you going to mirror them in the OS, or with RAID on the system motherboard? The better option is via the OS. I had planned on using motherboard RAID. Is it really that bad? My motherboard is a B550 motherboard that I bought in 2020, so the tech on it should be pretty recent. How do I go about setting up OS RAID on ubuntu? edit: just found a guide: https://kifarunix.com/setup-software-raid-on-ubuntu-20-04/, holy crap that looks like too much work, and will likely break at some point in the future. School of How fucked around with this message at 00:38 on Feb 11, 2022 |
|
#
?
Feb 10, 2022 23:04
|
|