July 6, 201511 yr I've installed a PCI-E sata card (IOCrest w/ Marvell9215 chipset) to get more ports and I've plugged an SSD (Samsung 840 Pro) into it. On boot, I get the following messages in the syslog (full syslog attached) Jul 1 22:20:54 Filebox kernel: ata12: SATA link up 6.0 Gbps (SStatus 133 SControl 300) Jul 1 22:20:54 Filebox kernel: ata12.00: failed to get NCQ Send/Recv Log Emask 0x1 Jul 1 22:20:54 Filebox kernel: ata12.00: ATA-9: Samsung SSD 840 PRO Series, S12SNEACC09230D, DXM06B0Q, max UDMA/133 Jul 1 22:20:54 Filebox kernel: ata12.00: 1000215216 sectors, multi 16: LBA48 NCQ (depth 31/32), AA Jul 1 22:20:54 Filebox kernel: ata12.00: failed to get NCQ Send/Recv Log Emask 0x1 Jul 1 22:20:54 Filebox kernel: ata12.00: configured for UDMA/133 As far as I can tell, the drive works fine. I've set it up in the cache pool and am running a VM off of it. Is this message something I should be concerned about, or can I just ignore it? EDIT: Added more details and changed title to be more relevant. filebox-syslog-20150705-1452.zip
July 6, 201511 yr Author I think I found some more information. It appears this might be a linux kernel bug affecting Samsung SSDs (840, maybe others?). The SSD I'm using is an 840 Pro. https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1338706 https://bugzilla.kernel.org/show_bug.cgi?id=72341 I'll try connecting the SDD directly to the motherboard to see if the issue remains. If it really is just a kernel bug, I'm not sure what to do other than wait and hope the unraid kernel catches up with a fix eventually.
July 6, 201511 yr Author I connected the 840 Pro to a sata port on the motherboard and get the same messages in the syslog. Jul 6 17:38:11 Filebox kernel: ata2.00: failed to get NCQ Send/Recv Log Emask 0x1 Jul 6 17:38:11 Filebox kernel: ata2.00: ATA-9: Samsung SSD 840 PRO Series, S12SNEACC09230D, DXM06B0Q, max UDMA/133 Jul 6 17:38:11 Filebox kernel: ata2.00: 1000215216 sectors, multi 16: LBA48 NCQ (depth 31/32), AA Jul 6 17:38:11 Filebox kernel: ata1.00: supports DRM functions and may not be fully accessible Jul 6 17:38:11 Filebox kernel: ata3.00: ATA-8: ST3000DM001-1CH166, Z1F318DZ, CC26, max UDMA/133 Jul 6 17:38:11 Filebox kernel: ata3.00: 5860533168 sectors, multi 16: LBA48 NCQ (depth 31/32), AA Jul 6 17:38:11 Filebox kernel: ata4.00: ATA-8: ST3000DM001-1CH166, W1F1QBYC, CC24, max UDMA/133 Jul 6 17:38:11 Filebox kernel: ata4.00: 5860533168 sectors, multi 16: LBA48 NCQ (depth 31/32), AA Jul 6 17:38:11 Filebox kernel: ata2.00: failed to get NCQ Send/Recv Log Emask 0x1 Jul 6 17:38:11 Filebox kernel: ata2.00: configured for UDMA/133 If anyone has any suggestions, I'd appreciate it. The bug reports mentioned in the previous post seem to indicate that this prevents TRIM from working, but I'm not entirely sure.
July 6, 201511 yr If this matters to you, you may want to enable NCQ under settings, disk settings.. By default it is forced to off. This may cause more issues than its worth, just thought I'd suggest.
July 7, 201511 yr I have several Samsung drives 840 Pro (240G) EVO (750G) and an 850 (128G). All working default out of the box. Is there a setting I could look at for you? I've also added a Sata III drive caddy that works pretty well for the SSD's - http://www.amazon.com/Syba-2-5-Inch-Caddy-PCI-Express-SD-PEX40044/dp/B005MN2GQQ/ref=sr_1_3?s=electronics&ie=UTF8&qid=1436233529&sr=1-3&keywords=syba+ssd+caddy in case you're interested...
July 7, 201511 yr Author If this matters to you, you may want to enable NCQ under settings, disk settings.. By default it is forced to off. This may cause more issues than its worth, just thought I'd suggest. But is the NCQ disable setting the cause of this issue?
July 7, 201511 yr Author Is there a setting I could look at for you? Thanks! What is your NCQ disable setting (as referenced by bungee91 above) set to? I don't think it's the cause here, but I'm out of ideas so I'll check anything.
July 7, 201511 yr I have a Samsung 840 EVO as my cache drive. Force NCQ disabled has always been set to 'yes' on my systems. I, too, am seeing the same error in my syslog. I only mention this for diagnostic purposes. I haven't seen any issues with performance/stability. From what I can tell, trim is working fine. Since I've got NCQ disabled, I'm not terribly concerned by this error. FYI, the drive is connected via a m1015 flashed to IT mode. -A
July 7, 201511 yr Author I have a Samsung 840 EVO as my cache drive. Force NCQ disabled has always been set to 'yes' on my systems. I, too, am seeing the same error in my syslog. I only mention this for diagnostic purposes. I haven't seen any issues with performance/stability. From what I can tell, trim is working fine. Since I've got NCQ disabled, I'm not terribly concerned by this error. FYI, the drive is connected via a m1015 flashed to IT mode. -A It's good to know that you haven't noticed any negative impacts from it. With the only information I can find on this being the previously reference bug reports for the kernel used in Ubuntu, I'm inclined to think it's a kernel bug. Unless someone who knows more than me can point me in a different direction, I'll try posting it as a defect and see what the mods/devs say.
Archived
This topic is now archived and is closed to further replies.