March 18, 201115 yr Hey All, Just added a Supermicro AOC-SASLP-MV8 to my Biostar A760GM2+ based unraid box. The cable I used is one from a recommended build page (can't find the exact post right now). It's a House Brand Molex 79576-3007 Mini SAS/SATA cable from surpluscomputers.com. I believe this is the one. Also just added two brand-new WD20EARS hard drives (they're reason for the new controller card). I previously had 6 drives taking up all 6 of the onboard sata ports, so I needed more sata ports in order to add the new drives. Here's the issue: Drives don't appear as available to add in unraid and they also don't show when doing a preclear_disk.sh -l. Here are the steps I've taken up to this point: 1: Shut down the box and added the card, then plugged the sas/sata cable into the to bottom slot on the card (ports 4-7). I then plugged two of the four sata ends into the two new drives and booted the box. Drives didn't appear as available in unraid or preclear -l. Check the syslog and didn't see them detected in there either. 2: Shut down the box, re-seated the card, plugged the sas/sata cable into the top slot on the card (ports 0-3) and made sure cables labeled ports 0 and 1 were the ones plugged into the two new drives (just in case the order matters to this card?) and booted the box. Again, drives didn't appear in unraid or preclear -l. Checked syslog and still didn't see them detected. This time, I looked for signs of the new card and I did see what looked like a detection of it, listing ports 0-7. 3: Again, I shut the box down, and this time just unplugged the sata connector from the parity drive and plugged it into one of the new drives and booted - just as a sanity check to make sure the drives didn't suffer some sort of fatal shipping accident. Sure enough, sage saw the new drive in the parity slot this time and wanted to upgrade the parity drive. After this, I did some searching around on the forums here and didn't really find anything that seemed related to my issue. Looking for advice as to what to try next. syslog from after boot from Step 2 is attached. Thanks, -Gary edit: updated title to SOLVED syslog_20110318-000100.txt
March 18, 201115 yr Hi there, Is your server headless or can you see the boot up screen. That would be the first thing. when the server boots/posts you should see it hook into the bios for the supermicro card where you can hit something like CTRL + M to get into the bios for the supermicro card. If you don't see that, your motherboard isn't picking up the supermicro card. Josh
March 18, 201115 yr Make sure to push the SAS connector into the controller until it clicks into place.
March 18, 201115 yr Do you see the bios of the card when booting? This is after the motherboard's bios boots and right before unraid boots. I know when I first got my card, unraid didn't see any drives either. When I rebooted, I noticed that the Supermicro bios wasn't picking up any drives. So I rebooted again, used cntrl M to get into the Supermicro bios. I don't think I even changed anything, but I could see my drives hooked up. After that, I rebooted, so the card bios pick the drives up and haven't had any problems with drive detection in unraid either. Hope it helps. Good luck.
March 18, 201115 yr Hey All, Just added a Supermicro AOC-SASLP-MV8 to my Biostar A760GM2+ based unraid box. The cable I used is one from a recommended build page (can't find the exact post right now). It's a House Brand Molex 79576-3007 Mini SAS/SATA cable from surpluscomputers.com. I believe this is the one. After this, I did some searching around on the forums here and didn't really find anything that seemed related to my issue. Gary, I am not sure how you came into this Molex cable but if you have searched here before - http://lime-technology.com/forum/index.php?topic=10547.0 you will know that this is a "reverse breakout" cable and as such not usable to the vast majority of people using the Supermicro AOC-SASLP-MV8 Only users with Norco 4220/4224 cases (or similar designs with SEF8087 backplanes) can use these cables (and usually only one or two at the max) If someone put this Molex part number in the recommended build - this is a mistake and it should be removed ASAP
March 18, 201115 yr Author @joshpond and icon123: Thanks, I'll check that when I get home (currently at work) @bjp999: Good advice, it is not immediately obvious how far the cable should go in, but i did notice it 'click' when i pushed it just a bit further than it seemed was enough, just to be sure it was fully seated. Thanks! @bcbgboy13: Hmmm... reading that post you linked, it looks like this could well be the problem. Funny thing is, i searched for the card and motherboard combo, but didn't think to include the cable... I obviously should have! I could also swear that I found that cable recommended (and linked) in reference to a non-Norco build, but of course I can't find anything but a Norco-related build reference now. One question: what did you mean when you said: ...not usable to the vast majority of people using Supermicro AOC-SASLP-MV8... Does that mean you know of someone who's gotten it to work?
March 18, 201115 yr Oops, mine mistake - this particular cable (Molex 79576-3007) is of no use to anyone with AOC-SASLP-MV8 controller. This cable is a reverse breakout and its intended purpose is to connect a backplane with SFF8087 connector to the motherboard SATA ports or controller(s) with regular SATA connectors. Only users with Norco 4220/4224 or better industrial design cases can use this particular cable and they will use only one (to connect a single backplane to four of the motherboard SATA ports). They can use second cable to connect another backplane to more of the motherboard SATA ports (a typical MB has 6 SATA connectors) and the remaining can be connected to a cheap 2-ports PCI-e 1x card (or any other controllers with regular SATA connectors such as the older Supermicro AOC-SAT2-MV8 (picture here - http://www.newegg.com/Product/ImageGallery.aspx?CurImage=15-121-009-S02&SCList=15-121-009-S01%2c15-121-009-S02%2c15-121-009-S03%2c15-121-009-S04%2c15-121-009-S05%2c15-121-009-S06%2c15-121-009-S07%2c15-121-009-S08%2c15-121-009-S09%2c15-121-009-S10%2c15-121-009-S11&S7ImageFlag=2&Item=N82E16815121009&Depa=0&WaterMark=1&Description=SUPERMICRO%20AOC-SAT2-MV8%2064-bit%20PCI-X133MHz%20SATA%20Controller%20Card) But no one can make this work with AOC-SASLP-MV8 However you can use Molex 79576-3003 (it is 0.5 m long) or 79576-3004 (it is 0.6 m long) as these are the regular "forward breakout" cables Check your PM
March 19, 201115 yr Author Thanks for the additional info. I'll have to order the correct cables and then post an update after I have a chance to try them out.
March 20, 201115 yr Wait. Are you sure you don't already have the right cable? I think you would know because it wouldn't plug into either the card or the hard drive correctly.
March 20, 201115 yr When this exact same symptom happened to me with this card, I found that I had the wrong cables too. They fit perfectly into the card and drives, but they were reverse breakouts. Fortunately NCIX.com replaced the cables free of charge because their original description of the reverse breakouts (wrong cable for this card) made them sound like they were really forward breakout cables. Once i had the right cables all was well. The clue was in the card's bios as already mentioned; the card clearly wasn't recognising any of the 6 drives attached to it.
March 25, 201115 yr Author Update: I now have the correct cables and the drives are recognized when I connect them to the SASLP-MV8 (Yay!) : read to bottom of post - it looks like there may be a syslog problem now... One minor new concern: I haven't added them to the array yet, but they are precleared. One strange thing I noticed is when I do a preclear_disk.sh -l just to see if they show up, they look like this: /dev/sdb = ata-WDC_WD20EARS-22MVWB0_WD-WCAZA3958401 /dev/sdc = scsi-SATA_WDC_WD20EARS-22_WD-WCAZA3952545 looks kinda odd to me... they are being detected as different types of drives? They are both WD20EARS SATA drives. Is this normal? Anything else I should look at to make sure all is well with these drives and the SASLP-MV8 card? - just looked at latest syslog and it's not looking too clean (to me), especially at the end... Can someone have a look at it as well - thanks!! Edit: <syslog attached> Thanks, -Gary syslog-20110325-010000.zip
March 25, 201115 yr I'm interested in seeing what an expert tells you. I have the same situation with the same setup as you. I can't get the card to preclear cleanly and I don't know if I even trust it at this point Edit: Sorry about the cables. I had know idea that they would fit but still be wrong.
March 25, 201115 yr Few recommendations: 1. Flash the motherboard BIOS to the latest version, then load the "default values", change the SATA to AHCI in the both locations and then disable anything that you wont use - serial and parallel ports, audio, firewire, even the IDE port if you do not plan to use older IDE drives - in this way you will free some resources. 2. Make sure you disable INT13 in the SM BIOS. You did not say what case do you have and now many drives you think you will get but the 6 onboard SATA ports plus the 8 from the SM card provide ample solutions for many people with smaller cases. You do not need the SIL3124 card (and I believe this is an older PCI card - fine for a single drive or two but anything more and it will slow down your parity checks) and it can be removed if you are not going to use it now. If you still want to use this SIL3124 card I will suggest to upgrade to the latest BIOS for it too - it is found here: http://www.siliconimage.com/support/searchresults.aspx?pid=27&cat=15&os=0 If the crash continues move the card to another PCI slot. And do not forget to perform a nice MEMTEST to be sure that your memory is 100% reliable.
March 25, 201115 yr Time for me to come clean. gary's original purchase of the wrong cable is very likely my fault. There was a short while (one or two weeks) in which I had the house brand reverse breakout cable listed in many of my builds that should have been using a forward breakout cable. This mistake was copied to the wiki as well. I have since fixed the issue, but unfortunately gary (and probably a handful of others as well) fell victim to it. I will be buying the reverse breakout cables from gary and using them in future client builds. If anyone else has this same issue, PM me and I'll see what I can do about it. gary, your syslog looks pretty normal to me. As bcbgboy13 said, the drives are showing up differently because your motherboard SATA ports must still be set to IDE mode, whereas they should be set to AHCI mode. You'll also want to disable anything called 'IDE combined mode'. Basically just disable anything that says 'IDE' at all.
March 25, 201115 yr Rajahal, I saw a thread once before from you in regards to the biostar mother board recommended bios settings, but when I went to that link it wasn't for the bios settings and I could never located it. If you know what I am talking about, can you see if you can find the link to the recommended bios settings for this board? Thanks.
March 25, 201115 yr As bcbgboy13 said, the drives are showing up differently because your motherboard SATA ports must still be set to IDE mode, whereas they should be set to AHCI mode. You'll also want to disable anything called 'IDE combined mode'. Basically just disable anything that says 'IDE' at all. I have not said this^ The two drives in question Gary is concerned are attached to the SM controller: Mar 25 00:20:44 NAS1 emhttp: pci-0000:02:00.0-sas-phy2:1-0x0200000000000000:2-lun0 host1 (sdb) WDC_WD20EARS-22MVWB0_WD-WCAZA3958401 Mar 25 00:20:44 NAS1 emhttp: pci-0000:02:00.0-sas-phy3:1-0x0300000000000000:3-lun0 host1 (sdc) WDC_WD20EARS-22_WD-WCAZA3952545 You should keep in mind that WD20EARS is the model but the manufacturer can have running change at any time (change in the numbers of the platters, change in the logic board and related firmware, change in the manufacturing location) and all this can result in change in the way the drive will be recognized (and WDC keeping the tracks of it) With the tragedy in Japan (they manufacture them self 40% of the flash memory, 15% of the DRAM and who knows what else - magnetic heads,etc..) this can be expected and not a cause for concern - especially if the drives were purchased from different places and at different times.
March 25, 201115 yr Rajahal, I saw a thread once before from you in regards to the biostar mother board recommended bios settings, but when I went to that link it wasn't for the bios settings and I could never located it. If you know what I am talking about, can you see if you can find the link to the recommended bios settings for this board? Thanks. Look in the wiki under hardware compatibility. Sent from my DROID2 using Tapatalk
March 26, 201115 yr Author Time for me to come clean. gary's original purchase of the wrong cable is very likely my fault. There was a short while (one or two weeks) in which I had the house brand reverse breakout cable listed in many of my builds that should have been using a forward breakout cable. This mistake was copied to the wiki as well. I have since fixed the issue, but unfortunately gary (and probably a handful of others as well) fell victim to it. No worries, Raj. For all you do to help people on these forums, you're entitled to a mistake once in awhile! --- Few recommendations: 1. Flash the motherboard BIOS to the latest version I decided not to do a BIOS upgrade just yet - rather exhaust other options since I have a (mostly) working system and I have been burned by unnecessary firmware upgrades in the past - not to say I don't trust your advice, just being extra cautious in this case. ... load the "default values", change the SATA to AHCI in the both locations and then disable anything that you wont use - serial and parallel ports, audio, firewire, even the IDE port if you do not plan to use older IDE drives - in this way you will free some resources. Booted into bios and verified all settings - they are default, except for changes recommended in the unRAID wiki specifically for my MB here. Namely, set everything to AHCI, disabled sata/ide combined mode, changed graphics to use smallest amount of memory (32M), changed USB to emulate HD and disabled all boot drives except USB. 2. Make sure you disable INT13 in the SM BIOS. Done. You did not say what case do you have and now many drives you think you will get but the 6 onboard SATA ports plus the 8 from the SM card provide ample solutions for many people with smaller cases. You do not need the SIL3124 card (and I believe this is an older PCI card - fine for a single drive or two but anything more and it will slow down your parity checks) and it can be removed if you are not going to use it now. If you still want to use this SIL3124 card I will suggest to upgrade to the latest BIOS for it too - it is found here: http://www.siliconimage.com/support/searchresults.aspx?pid=27&cat=15&os=0 The SIL3124 card was something I had laying around from my old unraid build. I installed it temporarily to preclear the two new wd20ears drives while I was waiting for the correct cables to arrive (multitasking! ) I simply forgot to remove it when I unhooked the drives from it and hooked up the sm card instead. I agree, I won't be needing it, so it's now removed. If the crash continues move the card to another PCI slot. My MB only has one suitable (>=4x) pcie slot, so changing slots won't be possible. And do not forget to perform a nice MEMTEST to be sure that your memory is 100% reliable. Ran 8 hours of memtest when i first built the box about 2-3 weeks ago. No problems. So, thanks for the tips! >> see next post for results of new tests done while making above changes.
March 26, 201115 yr Author Ok, new round of changes and testing based on posted suggestions (thanks!): 1) Shut down the box and pulled the temp 4-port SIL card I had in there for preclearing the new drives while waiting for the correct cable for the SM card. 1a) Booted into bios and verified settings were correct (see above post), then booted into SM bios and disabled int13. Rebooted. - results: same odd drive types when issuing a preclear -l. /dev/sdb = ata-WDC_WD20EARS-22MVWB0_WD-WCAZA3958401 /dev/sdc = scsi-SATA_WDC_WD20EARS-22_WD-WCAZA3952545 2) Shut down and swapped sata cables 0 and 1 on the drive end of the new forward break-out cable. So, basically I switched which drives the cables were plugged into. - result: same exact ids (in the same order) only serial #s reversed. This means the mis-identification seems to be with the port on the card and not with the drives themselves. /dev/sdb = ata-WDC_WD20EARS-22MVWB0_WD-WCAZA3952545 /dev/sdc = scsi-SATA_WDC_WD20EARS-22_WD-WCAZA3958401 3) Shut down and swapped out cable #1 for cable #2. So, I changed out the first forward break-out cable for the second one (just to verify it's not the cable) - result: same exact behavior as above. 4) Shut down and changed break-out cable from top port on the card to the bottom port on the card. - result: same exact behavior as above. 5) Shut down and unplugged SATA 0 and 1 on drive end of break-out cable and plugged in SATA 2 and 3 instead. - result: same exact behavior as above. So, nothing I've tried so far has changed the fact that it's seeing the drive types strangely. I also re-seated the SM card (again) during one of the shutdowns. I also took copies of the syslog each time after booting up. One piece of good news (maybe?) is that the nasty looking code-dump stuff at the end of the first syslog I posted in this thread never reared its head again. Here's it is, for reference (maybe someone can give an idea of what caused it): Mar 25 00:30:48 NAS1 kernel: WARNING: at drivers/ata/libata-core.c:5186 ata_qc_issue+0x10b/0x308() (Minor Issues) Mar 25 00:30:48 NAS1 kernel: Pid: 3426, comm: smartctl Not tainted 2.6.32.9-unRAID #8 (Errors) Mar 25 00:30:48 NAS1 kernel: Call Trace: (Errors) Mar 25 00:30:48 NAS1 kernel: [<c102449e>] warn_slowpath_common+0x60/0x77 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c10244c2>] warn_slowpath_null+0xd/0x10 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11b624d>] ata_qc_issue+0x10b/0x308 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11ac0ba>] ? scsi_init_sgtable+0x5b/0x76 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11ba260>] ata_scsi_translate+0xd1/0xff (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d (Errors) Mar 25 00:30:48 NAS1 kernel: [<f843369a>] sas_queuecommand+0x65/0x20d [libsas] (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11ace4d>] scsi_request_fn+0x351/0x376 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1126798>] __blk_run_queue+0x78/0x10c (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1124446>] elv_insert+0x67/0x153 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11245b8>] __elv_add_request+0x86/0x8b (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1129343>] blk_execute_rq_nowait+0x4f/0x73 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11293dc>] blk_execute_rq+0x75/0x91 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11292cc>] ? blk_end_sync_rq+0x0/0x28 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11295fa>] ? blk_recount_segments+0x16/0x24 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11250c6>] ? blk_rq_bio_prep+0x47/0x78 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1128edb>] ? blk_rq_append_bio+0x14/0x3b (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1129121>] ? blk_rq_map_user+0x12c/0x1be (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112c2bf>] sg_io+0x22d/0x30a (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112471f>] ? elv_queue_empty+0x20/0x22 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc (Errors) Mar 25 00:30:48 NAS1 kernel: [<c104cbfe>] ? get_page_from_freelist+0x2e4/0x344 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11b3257>] sd_ioctl+0x6a/0x8c (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112a420>] __blkdev_driver_ioctl+0x50/0x62 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c112b0f8>] ? get_disk+0x4a/0x61 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1192208>] ? get_device+0x11/0x18 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf (Errors) Mar 25 00:30:48 NAS1 kernel: [<c104a179>] ? filemap_fault+0xb8/0x305 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1048c43>] ? unlock_page+0x18/0x1b (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1057c63>] ? __do_fault+0x3a7/0x3da (Errors) Mar 25 00:30:48 NAS1 kernel: [<c10757d4>] ? do_filp_open+0x3d8/0x6d1 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c105985f>] ? handle_mm_fault+0x42d/0x8f1 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c108b6c6>] block_ioctl+0x2a/0x32 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c108b69c>] ? block_ioctl+0x0/0x32 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c10769d5>] vfs_ioctl+0x22/0x67 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1076f33>] do_vfs_ioctl+0x478/0x4ac (Errors) Mar 25 00:30:48 NAS1 kernel: [<c106afbf>] ? fd_install+0x1e/0x43 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1075b02>] ? set_close_on_exec+0x1e/0x38 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1076030>] ? do_fcntl+0x139/0x390 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1076f93>] sys_ioctl+0x2c/0x45 (Errors) Mar 25 00:30:48 NAS1 kernel: [<c1002935>] syscall_call+0x7/0xb (Errors) Mar 25 00:31:30 NAS1 ata_id[3784]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Here are just the Errors and Minor Issues that I get every time I boot (even with SM card removed): Mar 25 21:55:32 NAS1 kernel: ACPI Error (psargs-0359): [ECEN] Namespace lookup failure, AE_NOT_FOUND (Minor Issues) Mar 25 21:55:32 NAS1 kernel: ACPI Error (psparse-0537): Method parse/execution failed [\] (Node c14760c8), AE_NOT_FOUND (Minor Issues) Mar 25 21:55:32 NAS1 kernel: ACPI Warning: Incorrect checksum in table [OEMB] - 95, should be 88 (20090903/tbutils-314) (Minor Issues) Mar 25 21:55:32 NAS1 kernel: ACPI: I/O resource it87 [0xe85-0xe86] conflicts with ACPI region SENP [0xe85-0xe86] (Minor Issues) Mar 25 21:55:35 NAS1 emhttp: shcmd (29): killall -HUP smbd (Minor Issues) Mar 25 21:55:44 NAS1 apcupsd[1679]: apcupsd FATAL ERROR in linux-usb.c at line 609 Cannot find UPS device -- For a link to detailed USB trouble shooting information, please see <http://www.apcupsd.com/support.html>. (Errors) Mar 25 21:55:44 NAS1 apcupsd[1679]: apcupsd error shutdown completed (Errors) Mar 25 21:57:02 NAS1 sshd[2765]: error: Could not get shadow information for root (Errors) So, here are some of the questions I'm left with: 1) Can I reasonably conclude (probably not, but...) that the nasty syslog messages were caused by a conflict w/ the SIL card I removed before the first reboot on this last round of testing? If not, what other testing can I do to determine the cause? 2) Is it normal to have the first two drives recognized this way on the SM card (does anybody else w/ this card see this?!) ? 3) Are the remaining, consistent syslog issues/errors anything to worry about? 4) Assuming the answers to 1,2,3 are yes,yes & no respectively, should I take the big leap and add these drives to the array? (they passed preclear w/ flying colors, but that was on the SIL card, before I got the new cables) Not sure where to go from here... Thanks in advance for advice/suggestions on this. If needed, I can provide syslogs for any/all of the steps I listed above. I am including the syslog from after step 5 above, for reference. <syslog attached> -Gary edit: added step 1a syslog-20110325-213200.zip
March 26, 201115 yr 1) Can I reasonably conclude (probably not, but...) that the nasty syslog messages were caused by a conflict w/ the SIL card I removed before the first reboot on this last round of testing? If not, what other testing can I do to determine the cause? Sounds reasonable to me. If you removed the card and the syslog messages went away then that seems pretty straight-forward. 2) Is it normal to have the first two drives recognized this way on the SM card (does anybody else w/ this card see this?!) ? At first I didn't think so, but looking now at the syslog of a client build I'm working on (using the same SASLP card and A760G M2+ motherboard) it appears to be normal. I've attached that syslog for reference. Both syslogs also show the drives running at full speed. 3) Are the remaining, consistent syslog issues/errors anything to worry about? Your syslog looks clean. Compare it to the attached one, they are very similar. 4) Assuming the answers to 1,2,3 are yes,yes & no respectively, should I take the big leap and add these drives to the array? (they passed preclear w/ flying colors, but that was on the SIL card, before I got the new cables) Yes, I would feel confident in doing so. If you are still trepidatious, consider running another pass of preclear on the drives while connected to the SASLP card. You can use the -n option in preclear to skip the preread and it will finish much faster (generally less than 10 hours for 2 TB drives). syslog.txt
March 26, 201115 yr I agree with Raj. If your not seeing a performance issue then there is nothing to worry about. unRAID is very good at identifying correctness issues.
March 26, 201115 yr Author Yes, I would feel confident in doing so. If you are still trepidatious, consider running another pass of preclear on the drives while connected to the SASLP card. You can use the -n option in preclear to skip the preread and it will finish much faster (generally less than 10 hours for 2 TB drives). Thanks, Raj. Good to see that another system is having similar drive detection on the SALSP card. I'm not in a hurry to add the drives to the array, so I'll try another preclear on them and see how that comes out. I'll post back the results.
March 27, 201115 yr I have the same motherboard and Super Micro card. I haven't added any drives to my raid yet though. I have the same results with the preclear -l switch, so that must be normal at least with this motherboard. I had some issues preclearing drives with it though. Theses were also 20EARS. It would just lock up. Haven't got down to the bottow of it yet.
March 27, 201115 yr Author On last boot, the following popped into the syslog about 3 minutes after boot completed. Is this something to be concerned about? Mar 26 20:29:30 NAS1 kernel: sdb: sdb1 (Drive related) Mar 26 20:29:30 NAS1 kernel: ------------[ cut here ]------------ Mar 26 20:29:30 NAS1 kernel: WARNING: at drivers/ata/libata-core.c:5186 ata_qc_issue+0x10b/0x308() (Minor Issues) Mar 26 20:29:30 NAS1 kernel: Hardware name: A760G M2+ Mar 26 20:29:30 NAS1 kernel: Modules linked in: md_mod xor atiixp ahci r8169 mvsas libsas scst scsi_transport_sas (Drive related) Mar 26 20:29:30 NAS1 kernel: Pid: 3864, comm: smartctl Not tainted 2.6.32.9-unRAID #8 (Errors) Mar 26 20:29:30 NAS1 kernel: Call Trace: (Errors) Mar 26 20:29:30 NAS1 kernel: [<c102449e>] warn_slowpath_common+0x60/0x77 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c10244c2>] warn_slowpath_null+0xd/0x10 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11b624d>] ata_qc_issue+0x10b/0x308 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11ac0ba>] ? scsi_init_sgtable+0x5b/0x76 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11ba260>] ata_scsi_translate+0xd1/0xff (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d (Errors) Mar 26 20:29:30 NAS1 kernel: [<f843369a>] sas_queuecommand+0x65/0x20d [libsas] (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11a816c>] ? scsi_done+0x0/0xd (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11ace4d>] scsi_request_fn+0x351/0x376 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1126798>] __blk_run_queue+0x78/0x10c (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1124446>] elv_insert+0x67/0x153 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11245b8>] __elv_add_request+0x86/0x8b (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1129343>] blk_execute_rq_nowait+0x4f/0x73 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11293dc>] blk_execute_rq+0x75/0x91 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11292cc>] ? blk_end_sync_rq+0x0/0x28 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11295fa>] ? blk_recount_segments+0x16/0x24 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11250c6>] ? blk_rq_bio_prep+0x47/0x78 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1128edb>] ? blk_rq_append_bio+0x14/0x3b (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1129121>] ? blk_rq_map_user+0x12c/0x1be (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112c2bf>] sg_io+0x22d/0x30a (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112471f>] ? elv_queue_empty+0x20/0x22 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc (Errors) Mar 26 20:29:30 NAS1 kernel: [<c104cbfe>] ? get_page_from_freelist+0x2e4/0x344 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11b3257>] sd_ioctl+0x6a/0x8c (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112a420>] __blkdev_driver_ioctl+0x50/0x62 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c112b0f8>] ? get_disk+0x4a/0x61 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1192208>] ? get_device+0x11/0x18 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf (Errors) Mar 26 20:29:30 NAS1 kernel: [<c104a179>] ? filemap_fault+0xb8/0x305 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1048c43>] ? unlock_page+0x18/0x1b (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1057c63>] ? __do_fault+0x3a7/0x3da (Errors) Mar 26 20:29:30 NAS1 kernel: [<c10757d4>] ? do_filp_open+0x3d8/0x6d1 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c105985f>] ? handle_mm_fault+0x42d/0x8f1 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c108b6c6>] block_ioctl+0x2a/0x32 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c108b69c>] ? block_ioctl+0x0/0x32 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c10769d5>] vfs_ioctl+0x22/0x67 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1076f33>] do_vfs_ioctl+0x478/0x4ac (Errors) Mar 26 20:29:30 NAS1 kernel: [<c106afbf>] ? fd_install+0x1e/0x43 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1075b02>] ? set_close_on_exec+0x1e/0x38 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1076030>] ? do_fcntl+0x139/0x390 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1076f93>] sys_ioctl+0x2c/0x45 (Errors) Mar 26 20:29:30 NAS1 kernel: [<c1002935>] syscall_call+0x7/0xb (Errors) Mar 26 20:29:30 NAS1 kernel: ---[ end trace 048e91acf2138a64 ]--- Mar 26 20:29:47 NAS1 kernel: sdb: sdb1 (Drive related) Mar 26 20:30:13 NAS1 sshd[4128]: error: Could not get shadow information for root (Errors) Mar 26 20:30:20 NAS1 sshd[4169]: lastlog_filetype: Couldn't stat /var/log/lastlog: No such file or directory Mar 26 20:30:20 NAS1 sshd[4169]: lastlog_openseek: /var/log/lastlog is not a file or directory! Mar 26 20:30:20 NAS1 sshd[4169]: lastlog_filetype: Couldn't stat /var/log/lastlog: No such file or directory Mar 26 20:30:20 NAS1 sshd[4169]: lastlog_openseek: /var/log/lastlog is not a file or directory! Mar 26 20:30:40 NAS1 ata_id[4458]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 26 20:30:40 NAS1 kernel: sdc: sdc1 (Drive related) Mar 26 20:33:22 NAS1 ntpd[1453]: synchronized to 204.62.14.98, stratum 2 Mar 26 21:26:00 NAS1 kernel: mdcmd (18): spindown 0 (Routine) Mar 26 21:26:00 NAS1 kernel: mdcmd (19): spindown 1 (Routine) Mar 26 21:26:00 NAS1 kernel: mdcmd (20): spindown 2 (Routine) Mar 26 21:26:00 NAS1 kernel: mdcmd (21): spindown 3 (Routine) Mar 26 21:26:01 NAS1 kernel: mdcmd (22): spindown 4 (Routine) Mar 26 21:26:01 NAS1 kernel: mdcmd (23): spindown 5 (Routine) Full syslog also attached. Thanks, -Gary syslog.txt
March 28, 201115 yr Author New stuff in syslog since last post: Mar 27 03:01:05 NAS1 ata_id[29269]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 08:25:54 NAS1 dhcpcd[1445]: sending DHCP_REQUEST for 192.168.1.92 to 192.168.1.1 (Routine) Mar 27 08:25:54 NAS1 dhcpcd[1445]: dhcpIPaddrLeaseTime=86400 in DHCP server response. (Routine) Mar 27 08:25:54 NAS1 dhcpcd[1445]: DHCP_ACK received from (192.168.1.1) (Routine) Mar 27 09:01:13 NAS1 ata_id[6036]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:01:35 NAS1 ata_id[6109]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:01:45 NAS1 ata_id[6160]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:01:52 NAS1 ata_id[6198]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:02:03 NAS1 ata_id[6246]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:02:08 NAS1 ata_id[6273]: HDIO_GET_IDENTITY failed for '/dev/block/8:32' (Minor Issues) Mar 27 09:02:08 NAS1 kernel: sdc: sdc1 (Drive related) Mar 27 09:02:18 NAS1 kernel: udev: starting version 141 (System) Mar 27 09:30:01 NAS1 kernel: sdb: sdb1 (Drive related) Mar 27 09:30:12 NAS1 kernel: udev: starting version 141 (System) These occurred while doing a full preclear on the two new drives that are attatched to the new supermicro card. Anybody got any clues on these (and the ones in the previous post)? Thanks -Gary
Archived
This topic is now archived and is closed to further replies.