July 24, 201510 yr Posted this in the v5 forum originally. Re-posting here as it's more appropriate. unRAID OS Version: 6.0.1 Description: After doing a TimeMachine backup, the disk with the time machine share spins up every 30 minutes even if the OS X device is powered off/wifi turned off. How to reproduce: With unraid server up, verify that all disks are spun down and have been spun down for hours. Connect an OS X (Yosemite) machine and perform a time machine backup. Disconnect OS X device (disable network, power down). Watch as the disk with the backup spins up every ~30 minutes. Expected results: Disk doesn't spin up after it is done being used. Actual results: Disk spins up every 30 minutes even when not in use after the backup. Other information: Been trying to work out how to prevent this. I have a time machine share set up and confined to a single disk. Nothing else is on that disk. If I do not connect my laptop to the server, it spins down as normal. However, once I connect my laptop the disk will spin up every ~30 minutes or so, even after the laptop is powered off/wifi turned off. inotify says the file being accessed is: /mnt/disk9/TimeCapsule/.AppleDB/ MODIFY __db.002 lsof lists the files open on that drive as all being open under shfs: shfs 2139 root 4u REG 9,9 0 11 /mnt/disk9/TimeCapsule/.AppleDB/lock shfs 2139 root 5w REG 9,9 2379 12 /mnt/disk9/TimeCapsule/.AppleDB/db_errlog shfs 2139 root 7w REG 9,9 2379 12 /mnt/disk9/TimeCapsule/.AppleDB/db_errlog shfs 2139 root 8u REG 9,9 3653632 14 /mnt/disk9/TimeCapsule/.AppleDB/__db.002 shfs 2139 root 9u REG 9,9 24576 13 /mnt/disk9/TimeCapsule/.AppleDB/__db.001 shfs 2139 root 10u REG 9,9 10493952 15 /mnt/disk9/TimeCapsule/.AppleDB/__db.003 shfs 2139 root 11u REG 9,9 163840 16 /mnt/disk9/TimeCapsule/.AppleDB/__db.004 shfs 2139 root 12u REG 9,9 12337152 17 /mnt/disk9/TimeCapsule/.AppleDB/__db.005 shfs 2139 root 13u REG 9,9 49152 18 /mnt/disk9/TimeCapsule/.AppleDB/__db.006 shfs 2139 root 14u REG 9,9 10485760 76101 /mnt/disk9/TimeCapsule/.AppleDB/log.0000000012 shfs 2139 root 16u REG 9,9 8863744 20 /mnt/disk9/TimeCapsule/.AppleDB/cnid2.db My laptop is powered off with the wifi turned off, so I'm fairly sure it's not the laptop doing it, plus checking TimeMachine itself indicates that it is not making backups during this time. This seems similar to the issues seen here: http://lime-technology.com/forum/index.php?topic=21479.msg206906#msg206906 http://lime-technology.com/forum/index.php?topic=23132.0 I haven't been able to catch anything interested in lsof other than the above. I tried to do an inotifywait -> lsof to catch what is accessing the file, but didn't get anything different than normal when doing: inotifywait -r -e modify /mnt/disk9/TimeCapsule ; lsof | grep disk9 I'm open to suggestions on how to debug this or if it's just a quirk of the shfs program or something. Needless to say I'd be really happy if it didn't do this. Attached diagnostics for good measure. Any ideas? So far the only solution I have is to reboot the server after I do my backups but that doesn't seem like the right process. tower-diagnostics-20150718-2003.zip
July 26, 201510 yr Author Interestingly, after about 20 hours the syslog reports that the connection has timed out and the disks stop spinning up. Jul 25 20:49:23 tower kernel: mdcmd (125): spindown 0 Jul 25 20:49:24 tower kernel: mdcmd (126): spindown 9 Jul 25 21:19:23 tower kernel: mdcmd (127): spindown 0 Jul 25 21:19:24 tower kernel: mdcmd (128): spindown 9 Jul 25 21:49:23 tower kernel: mdcmd (129): spindown 0 Jul 25 21:49:24 tower kernel: mdcmd (130): spindown 9 Jul 25 22:19:23 tower kernel: mdcmd (131): spindown 0 Jul 25 22:19:24 tower kernel: mdcmd (132): spindown 9 Jul 25 22:49:23 tower kernel: mdcmd (133): spindown 0 Jul 25 22:49:24 tower kernel: mdcmd (134): spindown 9 Jul 25 23:19:23 tower kernel: mdcmd (135): spindown 0 Jul 25 23:19:23 tower kernel: mdcmd (136): spindown 9 Jul 25 23:38:17 tower afpd[11948]: afp_alarm: reconnect timer expired, goodbye Jul 25 23:38:17 tower afpd[12714]: afp_alarm: reconnect timer expired, goodbye Jul 25 23:38:17 tower afpd[13333]: afp_alarm: reconnect timer expired, goodbye Jul 25 23:38:17 tower afpd[13981]: afp_alarm: reconnect timer expired, goodbye Jul 26 00:03:23 tower kernel: mdcmd (137): spindown 0 Jul 26 00:03:24 tower kernel: mdcmd (138): spindown 9
Archived
This topic is now archived and is closed to further replies.