Writing to USB disk fills up root partition

nilleso · 05-29-2009, 05:24 PM

Hi..
writing to mounted USB disk fills up the root partition instead of the USB disk. Please help!!

Red Hat Enterprise Linux AS release 4 (Nahant Update 6)

On plugging in USB external drive, it get mounted automatically at /media/USB_BKUP_001:

Code:

May 29 17:51:24 hn kernel: usb 6-7: new high speed USB device using address 2
May 29 17:51:24 hn kernel: Initializing USB Mass Storage driver...
May 29 17:51:24 hn kernel: scsi2 : SCSI emulation for USB Mass Storage devices
May 29 17:51:24 hn kernel:   Vendor: Seagate   Model: FreeAgent XTreme  Rev: 4115
May 29 17:51:24 hn kernel:   Type:   Direct-Access                      ANSI SCSI revision: 02
May 29 17:51:24 hn kernel: SCSI device sdy: 2930277168 512-byte hdwr sectors (1500302 MB)
May 29 17:51:24 hn kernel: sdy: assuming drive cache: write through
May 29 17:51:24 hn kernel: SCSI device sdy: 2930277168 512-byte hdwr sectors (1500302 MB)
May 29 17:51:24 hn kernel: sdy: assuming drive cache: write through
May 29 17:51:24 hn kernel:  sdy: sdy1
May 29 17:51:24 hn kernel: Attached scsi disk sdy at scsi2, channel 0, id 0, lun 0
May 29 17:51:24 hn kernel: Attached scsi generic sg24 at scsi2, channel 0, id 0, lun 0,  type 0
May 29 17:51:24 hn kernel: usbcore: registered new driver usb-storage
May 29 17:51:24 hn kernel: USB Mass Storage support registered.
May 29 17:51:24 hn scsi.agent[6984]: disk at /devices/pci0000:00/0000:00:1d.7/usb6/6-7/6-7:1.0/host2/target2:0:0/2:0:0:0
May 29 17:51:24 hn multipathd: sdy: path checker registered
May 29 17:51:24 tor01cs1 multipathd: 1ST31500341AS____________________________2GER6M8Y: event checker started
May 29 17:51:24 tor01cs1 fstab-sync[7130]: added mount point /media/USB_BKUP_001 for /dev/sdy1
May 29 17:51:25 hn multipathd: 65:128: readsector0 checker reports path is up
May 29 17:51:25 hn multipathd: 65:128: reinstated
May 29 17:51:25 hn multipathd: 1ST31500341AS____________________________2GER6M8Y: remaining active paths: 1

/etc/fstab:

Code:

/dev/sdy1               /media/USB_BKUP_001     ext3    pamconsole,exec,noauto,managed 0 0

more details:

Code:

[root@hn log]# umount /media/USB_BKUP_001                        
umount: /media/USB_BKUP_001: not mounted
[root@hn log]# mount /dev/sdy1 /media/USB_BKUP_001/
mount: /dev/sdy1 already mounted or /media/USB_BKUP_001/ busy
[root@hn log]# umount /dev/sdy1                     
umount: /dev/sdy1: not mounted
[root@hn log]# mkdir /media/test
[root@hn log]# mount /dev/sdy1 /media/test/        
mount: /dev/sdy1 already mounted or /media/test/ busy
[root@hn log]#

Code:

[root@hn~]# [root@hnUSB_BKUP_001]# sdparm -a /dev/sdy
    /dev/sdy: Seagate   FreeAgent XTreme  4115Power condition mode page:
  IDLE        0  [cha: n, def:  0, sav:  0]
  STANDBY     0  [cha: n, def:  1, sav:  0]
  ICT         0  [cha: n, def:  0, sav:  0]
  SCT         0  [cha: n, def:9000, sav:  0]

stress_junkie · 05-29-2009, 06:16 PM

You should do some research before posting a question. The Seagate FreeAgent external disk problem has been around and discussed since the drives were first sold. Try using the LQ search feature and look for freeagent. Here are some references. Some of these were found using Google, others were found using the LQ web site search.

http://www.linuxquestions.org/questi...5/#post3489551

http://alienghic.livejournal.com/382903.html

http://www.linuxquestions.org/questi...3/#post2812441

http://www.theinquirer.net/inquirer/...rkaround-linux

There are many more available.

nilleso · 05-29-2009, 10:40 PM

Thanks for that but what makes you think I didn't do research?
In fact I'm completely familiar with the issue presented in the url's you've provided and they're entirely unrelated.

The disk is not not spinning down or going to sleep... but it isn't auto-mounting correctly.
cheers

nilleso · 05-29-2009, 11:59 PM

figured out the problem. multipath daemon was growing the disk into the / filesystem. To correct I needed to add the following line to /etc/multipath.conf

Code:

devnode_blacklist {
      wwid 1ST31500341AS____________________________2GER6M8Y
}

then `multipath -F` to flush all multipath device maps.

now this one disk is behaving well and working independently from the all the others.

Hope this solution helps someone.
cheers