SSD death ☠️💀

I am busy rebuilding the data from an SSD which died all of a sudden from one moment to the next without giving any warnings in any manner in any way.

I always monitor SMART output of SSD drives & mechanical spinners (HDD)

I've not seen any smart output indicating imminante dead

This drive has acted like your girlfriend when she's just not in a good mood and without explanation says nothing to you in the morning, for hours.

This hard crash, means that the S.M.A.R.T. monitoring hardware didn't have proper Communications with the integrated circuits on the SSD.

The drive was always powered many times a month, never left without power for more than a week or so, thus that has not been a contributing factor to the SSD catastrophic failure

The drive itself is fairly small.

The data on this dead SSD I've backed up on remote drives connected in JBOD format to machines which I have running remotely.

No ZFS on my backup machines

• I want to run ZFS native
• that means running a BSD OS on those machines
• that also means I will need to backup / restore all data on those drives meaning
• I will need 200% of the used HDD / SSD space on those machines
• I need patience for that backup
• I need expensive extra HDD's for that project
• I won't pay USD 300 for a USD 120 HDD

NO ZFS under these global SSD / HDD market prices

Sources:

• Moi
• man ls(1)
• man lsd(1)
• man cp(1)
• man smartctl(8)
• man zfs(8)

#HDD #SSD #crash #no #warning #on #TV #filesystems #remote #backup #network #JBOD #SMART #programming #bacula

Год в проде с Ceph: как мы пришли к новой референсной архитектуре

Привет, Хабр! Меня зовут Игорь Шишкин, я руковожу отделом разработки облачной платформы и архитектором SDS в Рунити. Ранее я уже рассказывал про то, как мы выбирали SDS (Software Defined Storage), почему остановились на Ceph, а также о наших процессах в R&D. В этой статье , поделюсь, что мы поймали за год в продакшене, какие решения в дизайне кластеров оказались ошибочными, как это изменило нашу референсную архитектуру и к чему мы пришли в итоге.

https://habr.com/ru/companies/runity/articles/1021222/

#регоблако #ceph #s3 #hsdc #конфигурация #exhausted #jbod #hba #архитектура #кластер

Someone here with more troubleshooting steps/ideas?

I have a refurbished LSI 9361-8i card and it doesn‘t detect any drives at all.
I checked 8643-8643 to a backplane - nothing.
I checked a 8643-4x8482 cable directly attached to the drives - which spin up on startup, but aren‘t recognized at all by the controller. Regardless of whether these are 2013 SATA or 2020 SATA drives.
Firmware is latest, and neither the #RAID nor #JBOD personality make a difference.

#HomeLab #NAS #Proxmox #Storage

I bought a 16 TB #Seagate Ironwolf Pro drive and just got done installing it into my Sabrent 5 bay #JBOD.

When I turn it on it makes a strange sound and neither Disk Management nor DiskPart at the command prompt can see the drive to initialize or format it.

I don't have to do anything special to a NAS drive running as a regular non-raid HD do I?

My other four drives contain 8 TB barracudas and they have all worked perfectly fine.

Is this perhaps a bad drive?

#AskFedi #AskMastodon

#OpenZFS_2025UserAndDeveloperSummit - Just a Bunch Of Disks? Really? 202510,
by Garrett W,
https://www.youtube.com/watch?v=QmiOzIcfGdM

#OpenZFS #JBOD #NAS #storage #fileSystem

Well, added a storage tier this evening.

A couple 4 Tb drives which I had laying about, as #JBOD (#MergerFS), for those media files which are "replaceable" and don't really need to be on #ZFS mirrors (~70 & ~16 Tb).

I now feed into #Plex from the different tiers using #OverlayFS.

Pretty neat. I can move things around without Plex noticing, select the tier in #Sonarr / #Radarr, and if one of those 4 Tb drives goes down, NBD and I can see what is missing.

More iops don't hurt either.