February 15, 20251 yr Hello guys, I've been on a journey these past few months.....See my previous post So over the past few months i have done a lot. I recently replaced 6x4TB drives with 3x14TB. My thought was that i had just far far too many drives connected. (8 port HBA, 4 port SATA pci card, and 2 into the mobo. Now I am running with 2 cache plugged directly into the mobo, and 8 drives connected via the HBA: 16 parity ------------- Mobo 16 parity ------------- Mobo 14 ---------- HBA ---- Mobo 14 ---------- HBA ---- Mobo 14 ---------- HBA ---- Mobo 6 ---------- HBA ---- Mobo < --- Current errors 6 ---------- HBA ---- Mobo < --- Previous errors before drives swap 16 ---------- HBA ---- Mobo 16 ---------- HBA ---- Mobo Cache ---------m.2----- Mobo Cache2 --------m.2----- Mobo Download ----- HBA ---- Mobo This morning i woke up with yet another disabled device. But this specific drive has NEVER given me errors. Overnight, the NAS was not doing anything special, nothing heavy, just its usual BTRFS background, and maybe download an episode or 2. As per usual, by simply explaing the problem, my 'desk ducky' has given me an answer I think its time to replace the HBA? or maybe just start with the cables. Any help understanding my issues is greatly appreciated hal-9000-diagnostics-20250215-0951.zip
February 15, 20251 yr Community Expert Not logged as a disk problem, could be power/connection, or since it happened just after spin up, it could also be this issue: https://forums.unraid.net/topic/103938-69x-lsi-controllers-ironwolf-disks-disabling-summary-fix/
February 15, 20251 yr Author This is the first time i'm seeing this. I will go through and try these tweaks tomor. Fingers crossed this will fix it
February 21, 20251 yr Author So, no dice. I followed those instructions and applied the fix to all my drives. I tried swapping the cables around....but the SAME drive disabled. I bought new cable......but the SAME drive disabled. I ran SMART.......passed. I ran BTRFS scrub....passed. So I now have had 3 more instances of the SAME drive disabling. In my anger/stress, i forgot to get diagnostics the 1st and 2nd time. But i have diags from this morning. I have now taken the cables, and the HBA out of the equation. I have found the drive to be in working order.....according to the self tests. At this point Im not buying a new drive, i've spent too much on this recently. I am currently emptying the disk, and going to just remove it from the array. Unless anyone has any ideas? My theory is with the cascade of issues, i've been dealing with recently, i think the drive is just buggered. It doesn't like coming back up from sleep, fails, gets disconnected, with an almost predictable routine. Its not even a year old! hal-9000-diagnostics-20250221-0925_disabled-drive.zip
February 21, 20251 yr Community Expert You never mention taking the steps to re-enable the disk? If you did, how did you do it?
February 21, 20251 yr Community Expert That looks correct. i notice that disk4 started getting read errors just after spinning back up. It might be worth trying disabling spin down for that drive to see if it helps.
February 21, 20251 yr Community Expert Normally these issues are caused by either the SATA cabling or power to the drive, but the entries in the syslog do not look quite like I would expect for those.
February 21, 20251 yr Author As I was typing my reply (below). I heard it.....not the clicking....the short repeated 'duh duh DUH....duh duh DUH' little whine. I think we have it....the answer! I have just re-enabled the drive......again....and I have set Spin Down Delay == Never. From my perspective its neither of those. I just dont get it. Maybe the motor is sluggish or something? Either way, this is my last attempt. I've had more than a few system hangs, read errors, cache filling up, just too many problems recently. I've managed to get things sorted, but my last issue is this one drive. It hates me. Maybe it got damaged during one of the colossal If i get any more errors, i think im just going to drop the drive.
February 21, 20251 yr Author So now, i need to wait for the parity sync to finish, empty the drive, and im taking it out. The whine was not there when i started the array without Disk4 (to register the disk missing) Stopped the array Re-added Disk4, and Started the array with parity-sync.....whine.
February 21, 20251 yr Author And now....in the middle of the parity rebuild.....the entire system just locked up. Can't even ssh in
February 23, 20251 yr Author OK, I REALLY need some help. So, I cleared out offending 6TB drive (WVY043GB). I removed the drive from the array. Made a new config, and set off a parity rebuild. Woke up this morning to ANOTHER bloody drive gave me 24,235 read errors! This is a different drive. Another 6TB, but now showing as Disk4 in the new config. My only choice now is the HBA. It cant be anything else. Does anyone have any recommendations for a new one? hal-9000-diagnostics-20250223-0913.zip
February 23, 20251 yr Author Am i right thinking that "I/O error, dev sdd, sector 7888116504" is outside the bounds of the 6TB? Isn't 7888116504 > 6TB?
February 23, 20251 yr Community Expert 55 minutes ago, 7hr08ik said: is outside the bounds of the 6TB? No, they are 512B sectors, I would try disabling spin down.
February 23, 20251 yr Author This is the same issue i was having with the other drive i just removed from the array. Now im getting it on this drive. Ive bought a new HBA, got a 9300-8i turning up, and a mini Noctua to give it some extra airflow. Im sick of all these errors. It goes from one drive, to the next, to the next. Changed cables, PSU isnt the issue. Once the error starts on a drive, it stays on that drive, until i change a cable, then it starts all over again on a new drive.
February 23, 20251 yr Community Expert Any power splitters? Ideally no more than 4 drives per PSU cable.
February 23, 20251 yr Author Nope. No power splitters. Just PSU cables that came with it. 3 per chain.
February 23, 20251 yr Author All my drives are iron wolf. I have tried applying the firmware fix suggested above, but that did nothing. First 6TB started getting disabled. Swapped cable around, same drive again. New cable, same drive again. Removed the drive from the array. New failures on a different drive, in the same way, on a different SFF cable to the other drive. (2 x SFF 8087 into my HBA) So is it just something with iron wolfs cant spin up properly any more? X doubt. The drives pass SMART, they pass BTRFS scrub. They only break, early morning, several hours after any overnight things finish (cold drives) At this point, I have a couple of options; I have borken multiple drives, that are less than a year old in such a way that they only error out, one after another, in succession, in repeatable ways. My HBA is broken/overheating. (Bought a new 9300-8i, and some airflow) Unraid has changed something that is causing my drives to fail, when coming out of sleep. Edited February 23, 20251 yr by 7hr08ik
February 23, 20251 yr Author Oh FML, i'm going to find a dark corner and cry. On a mere whim, I decided to double check SeaChest...just in case.... Guess what?? Im a moron, both 6TB have NOT changed settings Low Current Spinup, and EPC are still enabled. 😐
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.