Bug 593096 - initrd is wrong with lvm.conf containing 'command_names = 1' or 'prefix='something''
Summary: initrd is wrong with lvm.conf containing 'command_names = 1' or 'prefix='some...
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 5
Classification: Red Hat
Component: mkinitrd
Version: 5.5
Hardware: All
OS: Linux
urgent
high
Target Milestone: rc
: ---
Assignee: Brian Lane
QA Contact: Release Test Team
URL:
Whiteboard:
: 748733 (view as bug list)
Depends On:
Blocks: 726828
TreeView+ depends on / blocked
 
Reported: 2010-05-17 19:48 UTC by Flavio Leitner
Modified: 2018-11-14 17:40 UTC (History)
5 users (show)

Fixed In Version: mkinitrd-5.1.19.6-75
Doc Type: Bug Fix
Doc Text:
Clone Of:
: 760946 (view as bug list)
Environment:
Last Closed: 2012-02-21 05:33:37 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
suggested patch (500 bytes, patch)
2010-05-17 19:48 UTC, Flavio Leitner
no flags Details | Diff


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2012:0157 0 normal SHIPPED_LIVE mkinitrd bug fix update 2012-02-20 14:54:17 UTC

Description Flavio Leitner 2010-05-17 19:48:13 UTC
Created attachment 414648 [details]
suggested patch

Description of problem:
The mkinitrd script is not taking into account command_names or prefix 
from LVM's configuration. These 2 parameters can modify the output of 
an LVM command, and therefore the parsing can go wrong.

Major example is on the latest version : copying the LVM config file 
has been replace by dumping it (lvm dumpconfig).
The result is a corrupted config file (each line starts with dumpconfig) 
... and the machine won't manage to boot.

I believe the easiest would be to force the removal of such parameter with
the --config option in order to force the default behavior.

Example, line 1599 :
/sbin/lvm.static dumpconfig --config 'log{command_names=0 prefix="  "}' > $MNTIMAGE/etc/lvm/lvm.conf

Version-Release number of selected component (if applicable):


How reproducible:
mkinitrd-5.1.19.6-61.el5_5.1.i386

Steps to Reproduce:
1. run the command manually and check its output for command_names and prefix
/sbin/lvm.static dumpconfig --config 'log{command_names=1 prefix="lala"}'
  
Actual results:
The initrd image created my mkinitrd contains a lvm.conf with garbage
because the original lvm.conf has changed command_names and prefix defaults.

Expected results:
To work either with defaults or not.

Additional info:
This bug is a consequence of the 
https://bugzilla.redhat.com/show_bug.cgi?id=517868

Comment 5 Alexander Todorov 2010-10-06 15:30:08 UTC
Flavio,
does command_names or prefix affect orher invocations of lvm.static that are made from mkinitrd? I can see that they change the output of vgdisplay and lvs which are used in mkinitrd but I'm not sure if parsing will go wrong in all cases.

Comment 6 Flavio Leitner 2010-10-14 15:08:06 UTC
Hi Alexander,

Yes, it does.

Comment 10 RHEL Program Management 2011-08-05 12:35:35 UTC
This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux maintenance release.  Product Management has requested
further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed
products.  This request is not yet committed for inclusion in an Update
release.

Comment 15 Brian Lane 2011-10-25 22:43:30 UTC
*** Bug 748733 has been marked as a duplicate of this bug. ***

Comment 16 Alexander Todorov 2011-11-01 09:31:23 UTC
(In reply to comment #15)
> *** Bug 748733 has been marked as a duplicate of this bug. ***

With the Beta 1.0 tree the duplicate bug is not happening anymore. Systems can boot fine.

Comment 17 Alexander Todorov 2011-12-07 11:58:21 UTC
# rpm -q mkinitrd
mkinitrd-5.1.19.6-75.el5

# grep lvm.static /sbin/mkinitrd 
                local vg=$(lvm.static lvs --config 'log{command_names=0 prefix="  "}' --ignorelockingfailure --noheadings -o vg_name /dev/$slavedev 2>/dev/null)
    local vg=$(lvm.static lvs --config 'log{command_names=0 prefix="  "}' --ignorelockingfailure --noheadings -o vg_name $1 2>/dev/null)
            for device in $(lvm.static vgdisplay --config 'log{command_names=0 prefix="  "}' --ignorelockingfailure -v $vg 2>/dev/null | sed -n 's/PV Name//p'); do
                vg=$(lvm.static lvs --config 'log{command_names=0 prefix="  "}' --ignorelockingfailure --noheadings -o vg_name $rootdev 2>/dev/null)
                    for pv in $(lvm.static vgdisplay --config 'log{command_names=0 prefix="  "}' --ignorelockingfailure -v $vg 2>/dev/null | sed -n 's/PV Name//p'); do
    inst /sbin/lvm.static "$MNTIMAGE/bin/lvm"
        /sbin/lvm.static dumpconfig --config 'log{command_names=0 prefix="  "}' > $MNTIMAGE/etc/lvm/lvm.conf


Update /etc/lvm/lvm.conf and set command_names=1, prefix="---"
# lvs
lvs---LV       VG         Attr   LSize  Origin Snap%  Move Log Copy%  Convert
lvs---LogVol00 VolGroup00 -wi-ao 37,91G                                      
lvs---LogVol01 VolGroup00 -wi-ao  1,97G                                      

# pvs
pvs---PV         VG         Fmt  Attr PSize  PFree
pvs---/dev/vda2  VolGroup00 lvm2 a--  39,88G    0 

# vgs
vgs---VG         #PV #LV #SN Attr   VSize  VFree
vgs---VolGroup00   1   2   0 wz--n- 39,88G    0 


Rebuild the initrd.img:
# mkinitrd -f /boot/initrd-2.6.18-300.el5.img $(uname -

lvm.conf from the newly generated initrd.img contains:
  log {
  	verbose=0
  	syslog=1
  	overwrite=0
  	level=0
  	indent=1
  	command_names=1
  	prefix="---"
  }


Reboot:
... skip ...
Scanning logical volumes
vgscan---Reading all physical volumes.  This may take a while...
vgscan---Found volume group "VolGroup00" using metadata type lvm2
Activating logical volumes
vgchange---2 logical volume(s) in volume group "VolGroup00" now active
Trying to resume from /dev/VolGroup00/LogVol01
No suspend signature on swap, not resuming.
Creating root device.
Mounting root filesystem.
kjournald starting.  Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
Setting up other filesystems.
Setting up new root fs
no fstab.sys, mounting internal defaults
Switching to new root and running init.
unmounting old /dev
unmounting old /proc
unmounting old /sys
type=1404 audit(1323258909.670:2): enforcing=1 old_enforcing=0 auid=4294967295 ses=4294967295
type=1403 audit(1323258909.866:3): policy loaded auid=4294967295 ses=4294967295
INIT: version 2.86 booting
		Welcome to Red Hat Enterprise Linux Server
		Press 'I' to enter interactive startup.
Setting clock  (localtime): Wed Dec  7 11:55:10 EST 2011 [  OK  ]
Starting udev: [  OK  ]
Loading default keymap (us): [  OK  ]
Setting hostname dhcp70-168.rhts.eng.bos.redhat.com:  [  OK  ]
Setting up Logical Volume Management: vgchange---2 logical volume(s) in volume group "VolGroup00" now active
[  OK  ]
Checking filesystems
Checking all file systems.
[/sbin/fsck.ext3 (1) -- /] fsck.ext3 -a /dev/VolGroup00/LogVol00 
/dev/VolGroup00/LogVol00: clean, 34450/9942016 files, 588135/9936896 blocks
[/sbin/fsck.ext3 (1) -- /boot] fsck.ext3 -a /dev/vda1 
/boot: clean, 34/26104 files, 16145/104388 blocks
[  OK  ]
Remounting root filesystem in read-write mode:  [  OK  ]
Mounting local filesystems:  [  OK  ]
Enabling local filesystem quotas:  [  OK  ]
Enabling /etc/fstab swaps:  [  OK  ]
INIT: Entering runlevel: 3


I'm able to login into the system. Moving to VERIFIED.

Comment 18 errata-xmlrpc 2012-02-21 05:33:37 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2012-0157.html


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