Bug 492049

Summary: graphical partitioner gives None for most filesystem Type
Product: [Fedora] Fedora Reporter: John Reiser <jreiser>
Component: anacondaAssignee: Anaconda Maintenance Team <anaconda-maint-list>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: rawhideCC: anaconda-maint-list, dlehman, pjones, rmaximo, tmraz, vanmeeuwen+fedora
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-03-27 18:19:32 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
anaconda.log
none
program.log
none
storage.log
none
syslog
none
updates with longer udev settle timeout none

Description John Reiser 2009-03-25 03:41:15 UTC
Description of problem: The graphical disk partitioner lists None as the type for most partitions, where the actual types are ext3, ntfs, and gpt_data.


Version-Release number of selected component (if applicable):
anaconda-11.5.0.36-1.i586
pyparted-2.0.9-1.fc11.i586


How reproducible: always


Steps to Reproduce:
1. Compose DVD from rawhide of Tues.Mar.24 using pungi.
2. Boot DVD in install mode, choose "Create custom layout" for storage configuration.
3.
  
Actual results: List of partitions says Type is None instead of ext3, etc.  Extended partion on two drives with DOS label is identified correctly.


Expected results: ext3, ntfs, ext4, and gpt data partitions that are present should be identified properly.


Additional info: Disk labels are DOS (6 partitions), GPT (9 partitions), DOS (15 partitions.)

Comment 1 John Reiser 2009-03-25 17:23:29 UTC
Some of the problem reproduces today with i386 boot.iso containing anaconda-11.5.0.37-1.  Boot in install mode, choose "Create custom layout".  Several partitions appear with Type as "None".  Most of them are listed on VT1 as "udevadm settle - timeout of 2 seconds reached, the event queue contains:".  Will attach anaconda.log, program.log, storage.log, syslog.

Comment 2 John Reiser 2009-03-25 17:24:11 UTC
Created attachment 336680 [details]
anaconda.log

Comment 3 John Reiser 2009-03-25 17:24:35 UTC
Created attachment 336681 [details]
program.log

Comment 4 John Reiser 2009-03-25 17:24:56 UTC
Created attachment 336682 [details]
storage.log

Comment 5 John Reiser 2009-03-25 17:25:17 UTC
Created attachment 336683 [details]
syslog

Comment 6 David Lehman 2009-03-25 18:35:41 UTC
Created attachment 336688 [details]
updates with longer udev settle timeout

Try this updates.img and see if it helps. Let us know what happens.

Comment 7 John Reiser 2009-03-25 19:24:08 UTC
There was still one filesystem listed on VT1, and its Type was None.

However, the timeout was still 2, so I'm not sure that the updates.img actually changed anything.  Since this is my first time using anaconda updates, I'll be very specific about what I did.

Download the attachment in Comment #5.  Copy it with name "updates.img" onto a USB2.0 flash memory device (VFAT), where the file appears along with several others in the root directory:
-----
$ ls  ## total 718MB on 1GB flash storage
090325   f10utest-root  LiveOS        syslinux
090325b  f10utest-tmp   m2a-vm_h.bin  updates.img
$ ls -l updates.img
-rwxr-xr-x. 1 jreiser jreiser 140290 2009-03-25 12:00 updates.img
$ md5sum updates.img
75461d5ba8bb9815465431f7b7c194bc  updates.img
-----
Boot today's i386 boot.iso, type <Tab> with cursor on first line (Install), append " updates vga=ask" to kernel command line, type <Enter> to boot.  Select vga mode 0x324 (1280x1024x32).  Anaconda asks to choose between /dev/sr0 (CD/DVD) and /dev/sdd (USB 2.0 flash memory device.)  I choose /dev/sdd.  Anaconda asks to verify partition /dev/sdd4.  That's the correct one; I say OK.  Anaconda spends half a minute accessing the USB storage (LED activity light), then proceeds.  [This turns out to be copying all files on the flash memory into RAM.]  Eventually /dev/sdc5 still has Type None, and is still listed as late in the udevadm settle; but the timeout still is 2 seconds.

Comment 8 John Reiser 2009-03-25 19:39:40 UTC
Retrying with only the single file updates.img on the USB2.0 flash device: all partitions have correct Type.  However, the timeout is still 2 in "cd /tmp; grep settle *".  So where would any change have appeared?  What did the updates.img supposedly set the timeout?

Comment 9 David Lehman 2009-03-25 19:53:59 UTC
First: http://fedoraproject.org/wiki/Anaconda/Updates

To verify that the updates are in effect, go to tty2 and list /tmp/updates/storage/ -- if there are several files there, the timeout should no longer be 2 seconds. If there is no /tmp/updates/storage directory the updates were specified incorrectly.

Comment 10 David Lehman 2009-03-25 19:56:07 UTC
The updates.img I gave you is a compressed cpio archive, which may not work when copied onto a device via dd.

Comment 11 John Reiser 2009-03-25 20:18:14 UTC
Retrying after:
   gzip -d  < updates-492049.img  |  (cd USBdevice-root-directory;  cpio --extract --make-directories)
now gives a 'storage' directory that appears as a subdirectory in /tmp of running anaconda installer.

All partitions have correct Type, and "cd /tmp; grep settle *" shows timeout was 10.  So that looks like everything worked.

Comment 12 Chris Lumens 2009-03-27 18:19:32 UTC
Closing on the basis of comment #11.  Thanks for the testing.

Comment 13 David Lehman 2009-04-02 19:09:16 UTC
*** Bug 493270 has been marked as a duplicate of this bug. ***