Hi everyone,

I have a 10 drive array (1P+9D) and I have noticed that all xfs drives tend to spin ups and remain spun up. Meanwhile, similar ReiserFS drives do not spin up unexpectedly.
 

 

All disks belong to the same share and have the same settings: 

Spin down delay: use default

 

Tools -> Disk Settings:

Default spin down delay: 30 minutes

Enable spinup groups: No

 

Is there a reasonable explanation and a fix? I searched at the forums and found nothing....

towerp-diagnostics-20180504-0801.zip

Edited May 4, 20188 yr by papnikol

2 hours ago, papnikol said:

Is there a reasonable explanation and a fix? I searched at the forums and found nothing....

 

I'm sure there is but I looked for your diagnostics and found nothing too

35 minutes ago, John_M said:

 

I'm sure there is but I looked for your diagnostics and found nothing too

 

This is the nicest way I have ever been reminded that I can be stupid sometimes 

I have now attached the diagnostics file at the original post. Unfortunately, this time the drives happen to be spun down so that might be no helpful.

 

Hmmm... I have to say that I'm not really seeing what you're describing. Here's a section from your syslog covering a period of a little over a full day showing each disk as it spins down, with the irrelevant stuff snipped out:

May  2 14:35:33 towerP kernel: mdcmd (62): spindown 1
May  2 14:35:35 towerP kernel: mdcmd (63): spindown 3
May  2 14:35:36 towerP kernel: mdcmd (64): spindown 5
May  2 14:37:12 towerP kernel: mdcmd (65): spindown 2
May  2 14:37:13 towerP kernel: mdcmd (66): spindown 4
May  2 14:37:14 towerP kernel: mdcmd (67): spindown 6
May  2 14:37:15 towerP kernel: mdcmd (68): spindown 0
May  2 14:38:26 towerP kernel: mdcmd (69): spindown 7
May  2 14:40:57 towerP kernel: mdcmd (70): spindown 8
May  2 14:41:02 towerP kernel: mdcmd (71): spindown 9
May  2 21:09:39 towerP kernel: mdcmd (72): spindown 2
May  2 21:09:49 towerP kernel: mdcmd (73): spindown 9
May  2 21:09:59 towerP kernel: mdcmd (74): spindown 8
May  2 22:43:59 towerP kernel: mdcmd (75): spindown 1
May  2 22:44:09 towerP kernel: mdcmd (76): spindown 2
May  2 22:44:10 towerP kernel: mdcmd (77): spindown 3
May  2 22:44:16 towerP kernel: mdcmd (78): spindown 6
May  2 22:44:26 towerP kernel: mdcmd (79): spindown 8
May  2 22:44:46 towerP kernel: mdcmd (80): spindown 9
May  3 03:44:28 towerP kernel: mdcmd (81): spindown 5
May  3 03:44:45 towerP kernel: mdcmd (82): spindown 7
May  3 03:44:53 towerP kernel: mdcmd (83): spindown 4
May  3 03:45:15 towerP kernel: mdcmd (84): spindown 8
May  3 03:45:17 towerP kernel: mdcmd (85): spindown 1
May  3 03:45:21 towerP kernel: mdcmd (86): spindown 6
May  3 03:45:26 towerP kernel: mdcmd (87): spindown 2
May  3 03:45:27 towerP kernel: mdcmd (88): spindown 3
May  3 03:45:28 towerP kernel: mdcmd (89): spindown 9
May  3 04:24:50 towerP kernel: mdcmd (90): spindown 4
May  3 04:34:41 towerP kernel: mdcmd (91): spindown 0
May  3 04:49:39 towerP kernel: mdcmd (92): spindown 6
May  3 05:08:23 towerP kernel: mdcmd (93): spindown 5
May  3 05:09:11 towerP kernel: mdcmd (94): spindown 4
May  3 05:09:21 towerP kernel: mdcmd (95): spindown 9
May  3 05:10:04 towerP kernel: mdcmd (96): spindown 8
May  3 05:11:30 towerP kernel: mdcmd (97): spindown 1
May  3 05:12:28 towerP kernel: mdcmd (98): spindown 2
May  3 05:14:23 towerP kernel: mdcmd (99): spindown 3
May  3 08:22:17 towerP kernel: mdcmd (100): spindown 2
May  3 08:22:57 towerP kernel: mdcmd (101): spindown 4
May  3 08:24:52 towerP kernel: mdcmd (102): spindown 9
May  3 12:08:27 towerP kernel: mdcmd (103): spindown 0
May  3 12:08:28 towerP kernel: mdcmd (104): spindown 3
May  3 12:08:41 towerP kernel: mdcmd (105): spindown 4
May  3 12:08:43 towerP kernel: mdcmd (106): spindown 2
May  3 12:08:43 towerP kernel: mdcmd (107): spindown 9
May  3 12:08:51 towerP kernel: mdcmd (108): spindown 1
May  3 20:08:17 towerP kernel: mdcmd (109): spindown 4

Disks 2, 4 and 9 (the XFS-formatted ones) are all present and correct - several times. I can see no obvious difference between their pattern and that of the other disks. In fact I think the spin down delay you've chosen is too short because your disks are spinning up and down all the time. It's better for them to spin for longer and then stop for longer. I have my global default spindown delay set to 2 hours and all the disks are set to the global default. You have all except your parity disk set to the global default. You have the global default set to 30 minutes and the parity disk also set independently to 30 minutes, over-riding the global default. May I suggest you try a 2 hour delay for a few days and see how it goes. Review the situation and perhaps settle on a setting of 1 hour as a compromise?

Thanks for taking the time to check the diagnostics. I actually have normally set the default at 1hr, I had just changed it to 30min in order to check the problem I mentioned. I must admit that since I rebooted I haven't seen the spinups I were describing. Maybe I am obsessing but after the reboot, without me accessing the disc in order to write to it, I saw a few writes only to the 3 XFS disks. Still, this might be a quirk of XFS...