July 12, 201312 yr Today I upgraded?? to RC16c from RC15a. After that, disk 14 which is on a SATA 4-port Promise card would spindown weirdly. If I let the disk spindown timer spin it down, all would work properly. If I manually spundown all disks, then all disks would appear to spin down but disk 14 would appear to be spun down but if I refreshed the main screen in UnRaid. it would appear to be spun up..and it would not spindown automatically afterwards. I have been using UnRaid versions since 2008 and never had this problem that I know of. I backed up to RC15a and problem disappeared so it seems related to the OS version??? I use stock UnRaid, no addons. It isn't a big deal to me since I almost never manually spin down disks but since I upgraded?? the OS, I just wanted to get past the 30 min timer to see what would be steady state. I looked for similar problems in the threads but could not find anything for stock UnRaid. Syslog attached syslog_7_11_13-2036HRS.txt
July 12, 201312 yr I have the same issue. Drive 6 and 7 are both in a Supermicro AOC-SAS2LP-MV8 card. If I manually spin down drive 7 appears to be spun up still, but if I switch screen from UnRaid main to Main, the drive appears to be spun down.
July 12, 201312 yr Must admit i noticed something similar this morning with my server disk 5 always seems to be span up for some reason even when nothing is using the server, the rest of the disks seem to spin down without problems. Even when i spin all the drives down the rest go down including disk 5 but after a minute or so disk 5 comes up again When using rc15 all the disks span down with out issues after 15min. lsof /dev/md* would show only files being accessed on disk one but the other disks would seem to be span up from time to time also will have to keep an eye on it. Also noticed that unmenu would say a drives span down and the stock gui would say the opposite. The only thing that has changed it the server is I added an faster cache 250GB drive and will move the old cache drive 500GB to disk 6 slot once its finished pre-clearing, both these drives are on my M1015 controller and the rest of the drives are on the on board controller.
July 12, 201312 yr Update: After a little playing i decided to remove SF and reboot my server and low and behold all the drives span down including the troublesome disk 5. Stock UI and unMenu seem to match now with there display of the drive spinning status. Hopping this fixes the problem. Will miss SF though, oh well, now to figure out how to get unMenu to send me status emails
July 12, 201312 yr Author I don't use simple features or unmenu, just Tom's stock UI; the only non-stock code on the flash drive is preclear_disk.sh. My rig is a Tom-built 15 disk unit with p5vmdo MB, 4G, dual-core. I have one other disk (#13) on the same Promise SATA card and it behaves properly?? The drive that acts weird is a WDC_WD2002FAEX-007BA0_WD-WMAY03558572 (sdg) 1953514552, drive #14; the drive that acts properly is a Samsung 1TB drive. I've rebooted and the anomaly persists. Like I say, I rarely manually spin down disks so it isn't an issue of concern but remains confounding. I did notice that other posters are also using WD drives that may be acting strange...don't have a clue whether this similarity is important.
July 12, 201312 yr To all testing if a drive is truly spun up or down here. This command is used to obtain spun up/down status: hdparm -C /dev/sdX e.g. hdparm -C /dev/sdb If drive is spun up you should see this: /dev/sdb: drive state is: active/idle If drive is spun down, you should see this: /dev/sdb: drive state is: standby Based on the output from hdparm you will know if the gui you are using (unRAID Main, unMenu, SF) is or is not displaying properly for some reason. The following command (lsof "list open files") can be used to see if anything is open on the suspect disk as well: lsof | grep mnt
July 12, 201312 yr Author Thanks Madbug for the tool tutorial. I ran hdparm -C and sure enough, the disk 14 is spinning after I hit the spindown all disks button. lsof command produced this: root@Tower:~# lsof | grep mnt smbd 1521 root cwd DIR 9,14 1600 2 /mnt/dis k14 smbd 1521 root 30r DIR 9,14 1600 2 /mnt/dis k14 smbd 1522 root 30r DIR 9,14 1600 2 /mnt/dis k14 smbd 1522 root 32r DIR 9,1 2584 2 /mnt/dis k1 smbd 1522 root 33r DIR 9,2 2408 2 /mnt/dis k2 smbd 1522 root 34r DIR 9,12 2808 2 /mnt/dis k12 smbd 1522 root 35r DIR 9,13 568 2 /mnt/dis k13 I have no clue what this means; please assist. Thanks!
July 17, 201312 yr To all testing if a drive is truly spun up or down here. This command is used to obtain spun up/down status: hdparm -C /dev/sdX e.g. hdparm -C /dev/sdb If drive is spun up you should see this: /dev/sdb: drive state is: active/idle If drive is spun down, you should see this: /dev/sdb: drive state is: standby Based on the output from hdparm you will know if the gui you are using (unRAID Main, unMenu, SF) is or is not displaying properly for some reason. The following command (lsof "list open files") can be used to see if anything is open on the suspect disk as well: lsof | grep mnt I did the hdparm -C /dev/sdh, and it states that the drive is active. I did the lsof | grep mnt, and nothing comes up. Reverting back to 5R12a. UnMenu is unresponsive again every time I click on anything on the menu.
July 17, 201312 yr Thanks Madbug for the tool tutorial. I ran hdparm -C and sure enough, the disk 14 is spinning after I hit the spindown all disks button. lsof command produced this: root@Tower:~# lsof | grep mnt smbd 1521 root cwd DIR 9,14 1600 2 /mnt/dis k14 smbd 1521 root 30r DIR 9,14 1600 2 /mnt/dis k14 smbd 1522 root 30r DIR 9,14 1600 2 /mnt/dis k14 smbd 1522 root 32r DIR 9,1 2584 2 /mnt/dis k1 smbd 1522 root 33r DIR 9,2 2408 2 /mnt/dis k2 smbd 1522 root 34r DIR 9,12 2808 2 /mnt/dis k12 smbd 1522 root 35r DIR 9,13 568 2 /mnt/dis k13 I have no clue what this means; please assist. Thanks! Some client on the network via smb is "looking" at disk 14. There are several ways to re-run the test, one would be to temporarily shut down all your clients, less one that will have a browser open to the WebGui and telnet or ssh to unRAID (or via the console at the server) and make sure you don't have any app/windows explorer open to unRAID and re-run the test. Capture syslogs along the way, from fresh reboot with array started, and along the testing. After this testing is complete. Download hdparm and smartmontools updates from this posting : http://lime-technology.com/forum/index.php?topic=28382.msg252874#msg252874 (original poster of these packages pulled them unfortunitly, long story; or I would have had you reference his). Update your system with those two packages, re-try your test's. Let us know how you make out. I have a third test, but will leave it for later, lets start with these two.
July 17, 201312 yr To all testing if a drive is truly spun up or down here. This command is used to obtain spun up/down status: hdparm -C /dev/sdX e.g. hdparm -C /dev/sdb If drive is spun up you should see this: /dev/sdb: drive state is: active/idle If drive is spun down, you should see this: /dev/sdb: drive state is: standby Based on the output from hdparm you will know if the gui you are using (unRAID Main, unMenu, SF) is or is not displaying properly for some reason. The following command (lsof "list open files") can be used to see if anything is open on the suspect disk as well: lsof | grep mnt I did the hdparm -C /dev/sdh, and it states that the drive is active. I did the lsof | grep mnt, and nothing comes up. Reverting back to 5R12a. UnMenu is unresponsive again every time I click on anything on the menu. There has to be something outputting, highly doubt it was completely null. Please run RC16c, boot up, capture syslog after array starts, perform the test above (my previous post to the other individual) capture syslog, only then access unMenu, capture syslog once more. It needs to be understood, if we can't get enough data to prove what's wrong, RC16c is slated to be final, so going back to another release will not help you guys in the long run...
July 17, 201312 yr Author Since the issue only happens when I manually spin down all drives, and since I almost never do that under normal circumstances, and since most others are not experiencing this problem, I'm just gonna accept it as some sort of quirky anomaly that I can tolerate and avoid by letting the spindown timer run to completion. So I'm just moving on. Thanks again for the help!
July 18, 201312 yr I am on rc11 and I have also run into this issue a few times. Rebooting my server has always corrected the issue. I have also run into an issue where the drives would spin down at a certain time everyday and would spin back up at another specific time. This issue was also corrected by reboot.
July 18, 201312 yr Since the issue only happens when I manually spin down all drives, and since I almost never do that under normal circumstances, and since most others are not experiencing this problem, I'm just gonna accept it as some sort of quirky anomaly that I can tolerate and avoid by letting the spindown timer run to completion. So I'm just moving on. Thanks again for the help! I also noticed the same. I only have this issue when I manually spin down all drives. If I let the system spin down the drives by itself, it works fine.
Archived
This topic is now archived and is closed to further replies.