Bug 1068942 - FormatCreateError: ('invalid device specification', '/dev/sdb')
Summary: FormatCreateError: ('invalid device specification', '/dev/sdb')
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: anaconda
Version: rawhide
Hardware: x86_64
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Anaconda Maintenance Team
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard: abrt_hash:eab46a1b9e4ef86391353234a56...
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-02-23 12:58 UTC by satellitgo
Modified: 2014-03-10 13:45 UTC (History)
6 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2014-03-10 13:45:55 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
File: anaconda-tb (731.39 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: anaconda.log (65.41 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: environ (427 bytes, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: lsblk_output (1.77 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: nmcli_dev_list (1.70 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: os_info (381 bytes, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: program.log (35.15 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: storage.log (135.74 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: syslog (117.77 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: ifcfg.log (5.02 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details
File: packaging.log (334.50 KB, text/plain)
2014-02-23 12:58 UTC, satellitgo
no flags Details

Description satellitgo 2014-02-23 12:58:18 UTC
Description of problem:
custom /boot 500 MB  / 20GB format of MATE

Version-Release number of selected component:
anaconda-21.24-1

The following was filed automatically by anaconda:
anaconda 21.24-1 exception report
Traceback (most recent call first):
  File "/usr/lib/python2.7/site-packages/blivet/formats/__init__.py", line 391, in create
    raise FormatCreateError("invalid device specification", self.device)
  File "/usr/lib/python2.7/site-packages/blivet/formats/disklabel.py", line 244, in create
    DeviceFormat.create(self, *args, **kwargs)
  File "/usr/lib/python2.7/site-packages/blivet/deviceaction.py", line 470, in execute
    options=self.device.formatArgs)
  File "/usr/lib/python2.7/site-packages/blivet/devicetree.py", line 265, in processActions
    action.execute()
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 314, in doIt
    self.devicetree.processActions()
  File "/usr/lib/python2.7/site-packages/blivet/__init__.py", line 172, in turnOnFilesystems
    storage.doIt()
  File "/usr/lib64/python2.7/site-packages/pyanaconda/install.py", line 142, in doInstall
    turnOnFilesystems(storage, mountOnly=flags.flags.dirInstall)
  File "/usr/lib64/python2.7/threading.py", line 766, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/lib64/python2.7/site-packages/pyanaconda/threads.py", line 227, in run
    threading.Thread.run(self, *args, **kwargs)
FormatCreateError: ('invalid device specification', '/dev/sdb')

Additional info:
cmdline:        /usr/bin/python  /sbin/anaconda
cmdline_file:   initrd=initrd.img inst.stage2=hd:LABEL=Fedora\x20rawhide\x20x86_64 quiet BOOT_IMAGE=vmlinuz 
executable:     /sbin/anaconda
hashmarkername: anaconda
kernel:         3.14.0-0.rc3.git5.1.fc21.x86_64
product:        Fedora
release:        Cannot get release name.
type:           anaconda
version:        rawhide

Comment 1 satellitgo 2014-02-23 12:58:25 UTC
Created attachment 866650 [details]
File: anaconda-tb

Comment 2 satellitgo 2014-02-23 12:58:29 UTC
Created attachment 866651 [details]
File: anaconda.log

Comment 3 satellitgo 2014-02-23 12:58:31 UTC
Created attachment 866652 [details]
File: environ

Comment 4 satellitgo 2014-02-23 12:58:32 UTC
Created attachment 866653 [details]
File: lsblk_output

Comment 5 satellitgo 2014-02-23 12:58:33 UTC
Created attachment 866654 [details]
File: nmcli_dev_list

Comment 6 satellitgo 2014-02-23 12:58:35 UTC
Created attachment 866655 [details]
File: os_info

Comment 7 satellitgo 2014-02-23 12:58:36 UTC
Created attachment 866656 [details]
File: program.log

Comment 8 satellitgo 2014-02-23 12:58:38 UTC
Created attachment 866657 [details]
File: storage.log

Comment 9 satellitgo 2014-02-23 12:58:39 UTC
Created attachment 866658 [details]
File: syslog

Comment 10 satellitgo 2014-02-23 12:58:41 UTC
Created attachment 866659 [details]
File: ifcfg.log

Comment 11 satellitgo 2014-02-23 12:58:44 UTC
Created attachment 866660 [details]
File: packaging.log

Comment 12 David Lehman 2014-02-24 19:12:56 UTC
I can't tell if your drives are failing or if you are removing them, but you can check the syslog starting at 11:55:42,530 to see why the installer cannot complete installation on this system.

Comment 13 satellitgo 2014-02-28 01:56:54 UTC
21-24-1 boot.iso feb 27 2014 produced the same bug  works fine in virtualbox. will test to see if gparted format on disk to fat first works. I suspect this will work as virtualbox install works. Disk is not formatted here.

Comment 14 satellitgo 2014-02-28 01:58:20 UTC
in virtualbox

Comment 15 David Shea 2014-02-28 13:29:27 UTC
That doesn't change the fact that you're having hardware problems. I mean look at this:

11:55:43,271 INFO kernel:[ 1717.244348] sd 7:0:0:0: [sdb] Unhandled error code
11:55:43,271 INFO kernel:[ 1717.244358] sd 7:0:0:0: [sdb]  
11:55:43,271 WARNING kernel:[ 1717.244364] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
11:55:43,271 INFO kernel:[ 1717.244373] sd 7:0:0:0: [sdb] CDB: 
11:55:43,271 WARNING kernel:[ 1717.244380] Write(10): 2a 00 74 70 6d a7 00 00 01 00
11:55:43,272 ERR kernel:[ 1717.244549] end_request: I/O error, dev sdb, sector 1953525159
11:55:43,272 ERR kernel:[ 1717.244556] Buffer I/O error on device sdb, logical block 1953525159
11:55:43,272 WARNING kernel:[ 1717.244558] lost page write due to I/O error on sdb
11:55:43,272 ERR kernel:[ 1717.244593] Buffer I/O error on device sdb, logical block 1953525160
11:55:43,272 WARNING kernel:[ 1717.244594] lost page write due to I/O error on sdb
11:55:43,272 ERR kernel:[ 1717.244598] Buffer I/O error on device sdb, logical block 1953525161
11:55:43,272 WARNING kernel:[ 1717.244599] lost page write due to I/O error on sdb
11:55:43,272 ERR kernel:[ 1717.244602] Buffer I/O error on device sdb, logical block 1953525162
11:55:43,272 WARNING kernel:[ 1717.244603] lost page write due to I/O error on sdb
11:55:43,283 INFO kernel:[ 1717.255796] usb 4-1: USB disconnect, device number 3



>  works fine in virtualbox.

So it works differently in a different envrionment with virtualized hardware instead of the hardware that is failing? Congratulations.

Comment 16 satellitgo 2014-03-01 23:09:22 UTC
this does not occur if I use gparted on the same  external USB disk (externally) to create a new partition table and format it fat32 /dev/sdb. then in 21.24-1 boot iso 2/28/2104 it relclaims the disk and installs gnome3 normally on the external USB HD. Std LVM (5 partitions?)
gparted sees the previous live install on the disk as unallocated when it first sees it. There is something wrong with the detection of the disk if it has been previously used as a live f21 (rawhide) MATE install. VirtualBox has no previous install to format so it works on a new disk. 

This looks like a bug to me...

Is "reset MBR" missing in anaconda in this situation?

Comment 17 David Shea 2014-03-02 13:57:02 UTC
You are missing the point. This is not a problem with what anaconda is partitioning, or what it is zeroing, or what it is writing to the disk in any way. The problem is that anaconda CANNOT write to the disk, because from anaconda's point of view, there are problems with the disk. Maybe certain situations will dodge bad sectors, or maybe sometimes you just get lucky. But anaconda is attempting to write to the disk, and it is getting an I/O error back and the kernel is reporting this:

11:55:43,272 WARNING kernel:[ 1717.244599] lost page write due to I/O error on sdb
11:55:43,272 ERR kernel:[ 1717.244602] Buffer I/O error on device sdb, logical block 1953525162


That's really bad!

If you continue to insist that this is a software problem, report a bug against the kernel. There is nothing that anaconda can change to make it possible to write to your disk.

Comment 18 satellitgo 2014-03-09 11:43:15 UTC
FYI: I am finding that to successfully reuse a lvm install on an external USB HD of f21(rawhide); I have to first use gparted to make a new MBR and /dev/sdb1 in fat 32. 
I then use custom partitioning as ext4 std partitioning with 21.24-1 anaconda live CD. I delete the fat32 partition and use the std partitioning selection to
layout the disk. 
if I try to use anaconda's delete function to reclaim disk I get a kernel error on /dev/sdb USB HD. and install fails.

This is the same USB HD that fails if I use reclaim disk on a previous LVM f21(rawhide) install. Is this related to a USB recognition bug? 

This method of installation works every time, while the "reclaim disk" method fails?

Comment 19 David Shea 2014-03-10 13:45:55 UTC
Please stop reopening this bug. Please stop reopening bugs as MODIFIED.

It is possible that the errors writing to the drive are due to a bug in the kernel USB drivers. If you believe that is the case, open a bug against the kernel.


Note You need to log in before you can comment on or make changes to this bug.