Bug 1369713 - RHEL installation failed when with a ntfs formatted datadisk
Summary: RHEL installation failed when with a ntfs formatted datadisk
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: python-blivet
Version: 7.3
Hardware: x86_64
OS: Linux
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Blivet Maintenance Team
QA Contact: Release Test Team
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-08-24 08:31 UTC by yduan
Modified: 2021-09-03 14:09 UTC (History)
7 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-10-18 15:26:12 UTC
Target Upstream Version:


Attachments (Terms of Use)
Installation failed (75.58 KB, image/png)
2016-08-24 08:31 UTC, yduan
no flags Details
Delete partition in /dev/sdb (30.90 KB, image/png)
2016-08-24 08:34 UTC, yduan
no flags Details
Full CMD (2.24 KB, text/plain)
2016-08-24 08:38 UTC, yduan
no flags Details
anaconda.log (11.44 KB, text/plain)
2016-08-24 08:39 UTC, yduan
no flags Details
anaconda-yum.conf (373 bytes, text/plain)
2016-08-24 08:40 UTC, yduan
no flags Details
ifcfg.log (1.67 KB, text/plain)
2016-08-24 08:41 UTC, yduan
no flags Details
packaging.log (42.71 KB, text/plain)
2016-08-24 08:41 UTC, yduan
no flags Details
program.log (15.84 KB, text/plain)
2016-08-24 08:42 UTC, yduan
no flags Details
storage.log (33.48 KB, text/plain)
2016-08-24 08:45 UTC, yduan
no flags Details
syslog (75.02 KB, text/plain)
2016-08-24 08:45 UTC, yduan
no flags Details
X.log (16.80 KB, text/plain)
2016-08-24 08:46 UTC, yduan
no flags Details

Description yduan 2016-08-24 08:31:21 UTC
Created attachment 1193519 [details]
Installation failed

Description of problem:
RHEL installation failed when with a ntfs formatted datadisk.

Version-Release number of selected component (if applicable):
Host:
  kernel: 3.10.0-492.el7.x86_64
  qemu-kvm-rhev-2.6.0-22.el7.x86_64
Guest:
  RHEL 7.2 x86_64

How reproducible:
100%

Steps to Reproduce:
1.Create a new empty disk in host:
#qemu-img create -f qcow2 sysdisk 30G

2.Prepare a NTFS formatted disk file "datadisk5G.qcow2".

3.Start VM installation with following commands:
...
 -device virtio-scsi-pci,id=scsi_pci_bus0 \
 -drive file=/home/sysdisk,format=qcow2,id=drive_sysdisk,if=none,cache=none,aio=native,werror=stop,rerror=stop \
 -device scsi-hd,drive=drive_sysdisk,bus=scsi_pci_bus0.0,id=device_sysdisk,bootindex=0 \
 -drive file=/home/backup/RHEL-7.2-20151030.0-Server-x86_64-dvd1.iso,if=none,media=cdrom,id=drive_syscd,readonly=on,format=raw \
 -device scsi-disk,bus=scsi_pci_bus0.0,drive=drive_syscd,id=device_syscd,bootindex=1 \
 -device virtio-scsi-pci,id=scsi \
 -drive file=/home/datadisk5G.qcow2,format=qcow2,id=drive_datadisk5G,if=none,cache=none,aio=native,werror=stop,rerror=stop \
 -device scsi-hd,drive=drive_datadisk5G,bus=scsi.0,id=device_datadisk5G \
....

Actual results:
Installation failed as attachment "Installation failed".

Additional info:
Installation can go on after deleting the partition in ntfs formatted datadisk:
1.Enter shell in anaconda through send key "Ctrl+Alt+F2".
2.Delete partition in /dev/sdb as attachment "Delete partition in /dev/sdb".
3.Restart through send key "Ctrl+Alt+Del" and continue the installation.

Comment 1 yduan 2016-08-24 08:33:46 UTC
It is reproducible with qemu-kvm-rhev-2.3.0-31.el7_2.21.x86_64

Comment 2 yduan 2016-08-24 08:34:52 UTC
Created attachment 1193521 [details]
Delete partition in /dev/sdb

Comment 3 yduan 2016-08-24 08:36:32 UTC
Full command line is as attachment "Full CMD".
Available installation log files are as attachments.

Comment 4 yduan 2016-08-24 08:38:28 UTC
Created attachment 1193536 [details]
Full CMD

Comment 5 yduan 2016-08-24 08:39:34 UTC
Created attachment 1193537 [details]
anaconda.log

Comment 6 yduan 2016-08-24 08:40:40 UTC
Created attachment 1193538 [details]
anaconda-yum.conf

Comment 7 yduan 2016-08-24 08:41:09 UTC
Created attachment 1193539 [details]
ifcfg.log

Comment 8 yduan 2016-08-24 08:41:37 UTC
Created attachment 1193540 [details]
packaging.log

Comment 9 yduan 2016-08-24 08:42:09 UTC
Created attachment 1193541 [details]
program.log

Comment 10 yduan 2016-08-24 08:45:16 UTC
Created attachment 1193542 [details]
storage.log

Comment 11 yduan 2016-08-24 08:45:45 UTC
Created attachment 1193543 [details]
syslog

Comment 12 yduan 2016-08-24 08:46:13 UTC
Created attachment 1193545 [details]
X.log

Comment 14 Ademar Reis 2016-08-31 20:56:56 UTC
(In reply to yduan from comment #0)
> Steps to Reproduce:
> 1.Create a new empty disk in host:
> #qemu-img create -f qcow2 sysdisk 30G
> 
> 2.Prepare a NTFS formatted disk file "datadisk5G.qcow2".
> 

Please explain this step, as the error appears to be related to the layout/contents of the disk partition table. I can't see how QEMU would be at fault here.

Comment 15 yduan 2016-09-01 01:24:01 UTC
(In reply to Ademar Reis from comment #14)
> (In reply to yduan from comment #0)
> > Steps to Reproduce:
> > 1.Create a new empty disk in host:
> > #qemu-img create -f qcow2 sysdisk 30G
> > 
> > 2.Prepare a NTFS formatted disk file "datadisk5G.qcow2".
> > 
> 
> Please explain this step, as the error appears to be related to the
> layout/contents of the disk partition table. I can't see how QEMU would be
> at fault here.

File "datadisk5G.qcow2" was created as datadisk in former windows10 guest test, and it was formatted to NTFS in windows10 guest.
In this test case, i accidentally used this file "datadisk5G.qcow2" as datadisk too and then met this issue.
Iā€˜m not sure whether it is related to QEMU. Please feel free to reassign it to corresponding component if it is not QEMU's fault.

Thanks!

Comment 16 Laszlo Ersek 2016-10-18 15:01:51 UTC
I looked at the attached logs a bit. I think it's a problem with blivet, or maybe even intentional behavior from blivet. I'm unsure, but I think we can ask the blivet owners to take a look. So, reassigning just for the purpose of triaging. Thanks!

Comment 17 David Lehman 2016-10-18 15:26:12 UTC
The attached logs don't seem to match the reported failure. However, you will not see this failure in RHEL-7.3 due to an improvement in the handling of corrupt/unsupported disklabels. Generally, if you want to change the storage layout on the console while the installer is running, first wait until it has finished scanning the storage. Then make your changes and, finally, tell the installer to refresh its view of the storage (there's a refresh/rescan button). You could just as easily reboot after making the change if you cannot find the rescan button.

Comment 18 Laszlo Ersek 2016-10-18 15:38:52 UTC
Thank you, David!


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