[SOLVED] Can't add two drives...menu just loops

Got an issue I haven't encountered before.  Last night I added two new Hitachi 2TB drives to my array and followed standard procedure of stopping array and choosing the two new drives then the clearing ran and now when I go back to the array and select the "check box--I'm sure I want to do this" and then click on "Start" the disks start to mount but hitting "refresh" a minute or two later brings back the same screen with the drives not mounted and the "I'm sure I want to do this" box still there.

 

I've tried rebooting and starting the process over (same result).  Clearing browser cache etc without luck.  Just can't seem to get the array to come back online.  Any thoughts on next steps?  

 

Thanks!

 

Edit--logs are coming.   Version 4.7 final  

syslog.txt

Have you tried restarting?

Have you tried restarting?

 

Yep.  After the reboot it took me back through the "clearing" phase again with the same result.

New syslog from tonight's attempt...noticing some kernal errors...not sure what to make of it.

syslog-2011-03-22.txt

Have you tried restarting?

 

Yep.  After the reboot it took me back through the "clearing" phase again with the same result.

The unRAID array will be off-line while it clears the disks.   This is completely normal.  You will not be able to access your files until the clearing is complete.  (This is one of the two major reasons people pre-clear their disks before adding them to the array, to eliminate the extended down-time)

 

The clearing process will take quite a few hours with 2TB drives.  

However... the spin-down timer seems to be active on the disks being cleared, and that is probably a bug.

The clearing started here:

Mar 22 20:21:33 Deed emhttp: ... clearing 1% complete

Mar 22 20:24:52 Deed emhttp: ... clearing 2% complete

Mar 22 20:28:26 Deed emhttp: ... clearing 3% complete

Mar 22 20:32:18 Deed emhttp: ... clearing 4% complete

Mar 22 20:36:21 Deed emhttp: ... clearing 5% complete

Mar 22 20:40:02 Deed emhttp: ... clearing 6% complete

Mar 22 20:43:45 Deed emhttp: ... clearing 7% complete

Mar 22 20:47:39 Deed emhttp: ... clearing 8% complete

Mar 22 20:51:20 Deed emhttp: ... clearing 9% complete

Mar 22 20:54:29 Deed emhttp: ... clearing 10% complete

Mar 22 20:57:52 Deed emhttp: ... clearing 11% complete

then the spin down commands started...  and since the disk is being cleared, it keeps trying again and again to spin the disk down.

Mar 22 20:58:22 Deed kernel: mdcmd (74): spindown 18

Mar 22 20:58:23 Deed kernel: mdcmd (75): spindown 19

Mar 22 20:59:18 Deed kernel: mdcmd (76): spindown 18

Mar 22 20:59:19 Deed kernel: mdcmd (77): spindown 19

Mar 22 20:59:28 Deed kernel: mdcmd (78): spindown 18

Mar 22 20:59:29 Deed kernel: mdcmd (79): spindown 19

Mar 22 20:59:34 Deed kernel: mdcmd (80): spindown 18

Mar 22 20:59:34 Deed kernel: mdcmd (81): spindown 19

 

About 40 minutes into the clearing process.  Looks to be about 12% done.  Looks like a rate of roughly 20% per hour, but this will slow as the disk get closer to the inner cylinders.   Probably looking at another 4 or 5 hours of clearing past the time you took this syslog capture..

Mar 22 21:01:55 Deed emhttp: ... clearing 12% complete

Thanks Joe--I let the clearing run all night and tried to bring it online this AM with the same failed "looping" result.  I then proceeded to remove the two new drives from the array by unassigning those slots.  Went back to main page, refreshed and it recognized that there were no new drives so I "started the array" and got the same result.  Array would not start but menu looped back to the start array page.

 

syslog attached and it looks like there is a lock_rdev error.  What should I do now?

syslog-2011-03-23.txt

At this point I would try an initconfig to reset the array. Otherwise, you're going to need help from Limetech because I don't think anyone here will know what that lock_rdev is.

 

Peter

Thanks Peter--just tried that and got same result with the lock_rdev error.  Guess I'm stuck at this point.

At this point I would try an initconfig to reset the array.

 

Peter

I do not think an initconfig will help.  It will just invalidate parity and you'll be in the same situation.

you're going to need help from Limetech because I don't think anyone here will know what that lock_rdev is.

We do know what lock_rdev is:  It is part of the "md" source code supplied by lime-tech with unRAID.

/*
* prevent the device from being mounted, repartitioned or
* otherwise reused by a RAID array (or any other kernel
* subsystem), by bd_claiming the device.
*/
static int lock_rdev(mdk_rdev_t *rdev, dev_t dev)
{
       struct block_device *bdev;
       int err = 0;

       bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE);
       if (IS_ERR(bdev))
               return PTR_ERR(bdev);

       err = bd_claim(bdev, rdev);
       if (err) {
               blkdev_put(bdev, FMODE_READ|FMODE_WRITE);
               return err;
       }

       rdev->bdev = bdev;

       return err;
}

 

I think it is trying to lock the disk being affiliated with the "md" device so it has exclusive use.

 

Do you have any applications accessing your disks being started in the "go" script?  If one of them creates a directory before a disk is mounted by unRAID it might be the problem.

 

revert back to a stock "go" script for now.

 

Thanks Joe!

 

The only application I've loaded is unmenu and screen per the configuration tutorial here:

 

http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial

 

Not sure how to modify the go script but I'm off to the wiki to find out.

Thanks Joe!

 

The only application I've loaded is unmenu and screen per the configuration tutorial here:

 

http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial

 

Not sure how to modify the go script but I'm off to the wiki to find out.

Put a "#" character at the beginning of any added lines at the bottom of the file.

 

You can edit the config/go file in any editor, even a windows editor.

 

However, what you described is not likely to be the cause of your problem... not unless you used unMENU to install other utilities.

 

Thanks Joe!

 

The only application I've loaded is unmenu and screen per the configuration tutorial here:

 

http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial

 

Not sure how to modify the go script but I'm off to the wiki to find out.

Put a "#" character at the beginning of any added lines at the bottom of the file.

 

You can edit the config/go file in any editor, even a windows editor.

 

However, what you described is not likely to be the cause of your problem... not unless you used unMENU to install other utilities.

 

 

I'm certain I used unMenu to install the other required apps.  Thanks again!

I am not sure is this is related to anything but you have the following warning about your "sds" - Unraid slot 1:

 

Mar 23 07:26:40 Deed kernel: ata18: SATA link up 3.0 Gbps (SStatus 123 SControl 300)

Mar 23 07:26:40 Deed kernel: ata18.00: ATA-8: ST31500341AS, SD17, max UDMA/133

Mar 23 07:26:40 Deed kernel: ata18.00: 2930277168 sectors, multi 0: LBA48 NCQ (not used)

Mar 23 07:26:40 Deed kernel: ata18.00: WARNING: device requires firmware update to be fully functional.

Mar 23 07:26:40 Deed kernel: ata18.00:          contact the vendor or visit http://ata.wiki.kernel.org.

Mar 23 07:26:40 Deed kernel: ata18.00: configured for UDMA/133

Mar 23 07:26:40 Deed kernel: scsi 4:0:0:0: Direct-Access    ATA      ST31500341AS    SD17 PQ: 0 ANSI: 5

Mar 23 07:26:40 Deed kernel: [894]: scst_suspend_activity:599:suspend_count 0

Mar 23 07:26:40 Deed kernel: [894]: scst_susp_wait:578:

Mar 23 07:26:40 Deed kernel: sd 4:0:0:0: [sds] 2930277168 512-byte logical blocks: (1.50 TB/1.36 TiB)

Mar 23 07:26:40 Deed kernel: sd 4:0:0:0: [sds] Write Protect is off

Mar 23 07:26:40 Deed kernel: sd 4:0:0:0: [sds] Mode Sense: 00 3a 00 00

Mar 23 07:26:40 Deed kernel: sd 4:0:0:0: [sds] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

Mar 23 07:26:40 Deed kernel:  sds:wait_event() returned 0

 

It is from your latest syslog

Thanks for the reminder---its the dreaded 1.5TB Seagate Baracuda 7200.11 drive that I'm sure needs a firmware update.  I saw that message but completely forgot about it in trying to fix my original problem.  Thanks!

Well I just rem'd out the part in my go file about software packages, rebooted and still same result.  Wondering at this point if I should try to remove "sds" from the array and assign one of the hitachi's in its place.  Trying to get back to square one at this point.

Joe - could that be a lock stored somewhere, like on the flash or a bit in the MBR? I was just thinking maybe there is a lock or flag stored somewhere while the drives are being cleared.

 

Peter

Thanks Joe!

 

The only application I've loaded is unmenu and screen per the configuration tutorial here:

 

http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial

 

Not sure how to modify the go script but I'm off to the wiki to find out.

Put a "#" character at the beginning of any added lines at the bottom of the file.

 

You can edit the config/go file in any editor, even a windows editor.

 

However, what you described is not likely to be the cause of your problem... not unless you used unMENU to install other utilities.

 

 

I'm certain I used unMenu to install the other required apps.  Thanks again!

 

What do you mean by "other required apps"?

Thanks Joe!

 

The only application I've loaded is unmenu and screen per the configuration tutorial here:

 

http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial

 

Not sure how to modify the go script but I'm off to the wiki to find out.

Put a "#" character at the beginning of any added lines at the bottom of the file.

 

You can edit the config/go file in any editor, even a windows editor.

 

However, what you described is not likely to be the cause of your problem... not unless you used unMENU to install other utilities.

 

 

I'm certain I used unMenu to install the other required apps.  Thanks again!

 

What do you mean by "other required apps"?

 

Poor wording on my part.  Specifically for screen there was one or two apps I installed via unmenu.  I've commented out the entire go file and nothing seems to work.  I can't get my array back online at this point.

Latest syslog from my reboot etc. 

syslog-2011-03-23.txt

You can't comment out this line.

 

/usr/local/sbin/emhttp &

 

Peter

 

You can't comment out this line.

 

/usr/local/sbin/emhttp &

 

Peter

 

 

;D

 

Yeah I figured that out the hard way!

 

As an FYI here's how it currently looks:

 

#!/bin/bash
# Start the Management Utility
/usr/local/sbin/emhttp &
/boot/unmenu/uu
cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c

 

I've only commented out the last line and that didn't change the result.

Please comment the last 2 lines and then post a syslog.

Please comment the last 2 lines and then post a syslog.

 

Done--thanks for your help!

 

syslog.txt

I think the clues needed are here:

Mar 23 22:38:21 Deed emhttp: get_fstype: open /dev/sdh1: No such file or directory

 

Mar 23 22:38:22 Deed emhttp: mdcmd: write: No such device or address

Mar 23 22:38:22 Deed kernel: mdcmd (40): start STOPPED

Mar 23 22:38:22 Deed kernel: md: do_run: lock_rdev error: -6

 

I'm guessing that the disk partition in the MBR of this disk no longer exists and the partition cannot be associated with the /dev/md device.

 

To confirm, please run the following command and post the output. 

It will print the current structure of the master boot record of /dev/sdh.

dd status=noxfer count=1 if=/dev/sdh | od -Ad -t x1

 

It should be possible to then fix the MBR and get everything back to normal.

 

 

 

Thanks Joe--here's the output of the command given:

 

Deed login: root

Password:

Linux 2.6.32.9-unRAID.

root@Deed:~# dd status=noxfer count=1 if=/dev/sdh | od -Ad -t x1

0000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

1+0 records in

1+0 records out

*

0000512

root@Deed:~#