[SOLVED] Array doesn't automatically start at system boot

[boy914]

Over the weekend I had to rebuild my system after the flash drive died (I now know to back up the flash drive along with everything else).

 

After getting everything up and running again, I installed a new 3TB drive for parity, and then after the parity rebuild, deployed the old 1TB parity drive as disk2. I then precleared the former disk2 and unplugged it from power and the motherboard. Disk1 wasn't touched. In the end, I have:

 

• Parity: 3TB Seagate (ST3000DM001)
  
• disk1: 1TB WD (WD1001FALS)
  
• disk2: 1TB WD (WD1001FALS - former parity)
  

 

During the various drive swapping, I was made sure to always have the parity drive on SATA0, disk1 on SATA1 and disk2 on SATA2. Any drive that was not part of the array (ie being precleared) was on SATA3. My assumption was that SATA0 = sda, SATA1 = sdb and so on.

 

Here's the strange behavior: Now that I have everything the way I want it, whenever I boot the system, the array does not start. The syslog fills with a repeating set of messages (see below and attached).

 

I'm also noticing that the "sdX" assignment for each drive is changing on each reboot. One time it will be sda (parity), sdb (disk1), sdc (disk2), sdd (flash), which is what I would expect, but the next will be sda (parity), sdb (flash), sdc (disk1), sdd (disk2).

 

I can start the array on the main page with no issues, and the syslog entries stop and everything works great. No parity problems, no strange errors.

 

Other system info:

 

• Version 5b14
  
• unMENU, with a few basic packages
  
• The current SimpleFeatures release, with all of the sub-packages
  
• Custom "go" script to wait for disk1 to spin up and then start sabnzbd & sickbeard. Since array isn't starting at boot, these packages are not running
  

 

Any ideas? Early in the boot process I see these two lines, but I don't know what they mean:

 

Apr  4 07:48:09 tower logger: Warning: simplexml_load_file(): I/O warning : failed to load external entity "/boot/config/plugins/*.plg" in /usr/local/sbin/installplg on line 13 (Minor Issues)

Apr  4 07:48:09 tower logger:  xml parse error (Errors)

 

These are the repeating lines that keep appearing until I manually start they array. Then they stop, and all is well:

 

Apr  4 08:34:22 tower emhttp: shcmd (111): rmmod md-mod |$stuff$ logger (Other emhttp)

Apr  4 08:34:22 tower emhttp: shcmd (112): udevadm settle (Other emhttp)

Apr  4 08:34:22 tower kernel: md: unRAID driver removed (System)

Apr  4 08:34:22 tower emhttp: Device inventory: (Drive related)

Apr  4 08:34:22 tower emhttp: ST3000DM001-9YN166_Z1F0AAWT (sda) 2930266584 (Drive related)

Apr  4 08:34:22 tower emhttp: WDC_WD1001FALS-00J7B1_WD-WMATV2061219 (sdd) 976762584 (Drive related)

Apr  4 08:34:22 tower emhttp: WDC_WD1001FALS-00J7B1_WD-WMATV2199858 (sdc) 976762584 (Drive related)

Apr  4 08:34:22 tower emhttp: shcmd (113): modprobe md-mod super=/boot/config/super.dat slots=3 |$stuff$ logger (unRAID engine)

Apr  4 08:34:22 tower kernel: md: unRAID driver 2.1.3 installed (System)

Apr  4 08:34:22 tower kernel: mdcmd (1): import 0 8,0 2930266532 ST3000DM001-9YN166_Z1F0AAWT (unRAID engine)

Apr  4 08:34:22 tower kernel: md: import disk0: [8,0] (sda) ST3000DM001-9YN166_Z1F0AAWT size: 2930266532 (Drive related)

Apr  4 08:34:22 tower kernel: mdcmd (2): import 1 8,32 976762552 WDC_WD1001FALS-00J7B1_WD-WMATV2199858 (unRAID engine)

Apr  4 08:34:22 tower kernel: md: import disk1: [8,32] (sdc) WDC_WD1001FALS-00J7B1_WD-WMATV2199858 size: 976762552 (Drive related)

Apr  4 08:34:22 tower emhttp: shcmd (114): /usr/local/sbin/emhttp_event driver_loaded (System)

Apr  4 08:34:22 tower emhttp_event: driver_loaded (System)

Apr  4 08:34:22 tower kernel: mdcmd (3): import 2 8,48 976762552 WDC_WD1001FALS-00J7B1_WD-WMATV2061219 (unRAID engine)

Apr  4 08:34:22 tower kernel: md: import disk2: [8,48] (sdd) WDC_WD1001FALS-00J7B1_WD-WMATV2061219 size: 976762552 (Drive related)

Apr  4 08:34:23 tower emhttp: shcmd (115): rmmod md-mod |$stuff$ logger (Other emhttp)

Apr  4 08:34:23 tower emhttp: shcmd (116): udevadm settle (Other emhttp)

Apr  4 08:34:23 tower emhttp: Device inventory: (Drive related)

Apr  4 08:34:23 tower emhttp: ST3000DM001-9YN166_Z1F0AAWT (sda) 2930266584 (Drive related)

Apr  4 08:34:23 tower emhttp: WDC_WD1001FALS-00J7B1_WD-WMATV2061219 (sdd) 976762584 (Drive related)

Apr  4 08:34:23 tower emhttp: WDC_WD1001FALS-00J7B1_WD-WMATV2199858 (sdc) 976762584 (Drive related)

Apr  4 08:34:23 tower emhttp: shcmd (117): modprobe md-mod super=/boot/config/super.dat slots=3 |$stuff$ logger (unRAID engine)

Apr  4 08:34:23 tower emhttp: shcmd (118): /usr/local/sbin/emhttp_event driver_loaded (System)

Apr  4 08:34:23 tower kernel: md: unRAID driver removed (System)

Apr  4 08:34:23 tower kernel: md: unRAID driver 2.1.3 installed (System)

syslog-2012-04-04.txt

[dikkiedirk]

Is auto-start enabled in Disk Settings?

[boy914]

Yup. I had to check just to be sure.

[prostuff1]

sdX assignments do not matter, stop trying to make sense of them

 

as for the array not starting:

 

set it to no, hit apply, hit done, reboot, set to yes, hit apply, hit done, reboot.  The array should start.

[boy914]

Huh. Sorry to doubt you, prostuff1, but I didn't think that would solve it. I figured the repeating syslog entries pointed to some other issue.

 

But I was wrong: Array started up right away after the 2nd reboot, the Python servers started up, and the syslog looks good.

 

Thanks for the help, glad it was something easy!