2017-03-23 18:20 UTC

View Issue Details Jump to Notes ]
IDProjectCategoryView StatusLast Update
0012426CentOS-7grub2public2017-03-23 07:43
Platformx86_64OSCentOS7OS Versionall versions
Product Version 
Target VersionFixed in Version 
Summary0012426: system boot hangs after lvm snapshot from root partition
DescriptionI'm using LVM snapshots for backups on CentOS7. After create a snapshot from root partition, the system hangs on next start (all versions).

To fix this problem it is possible to boot rescue option (grub2 boot menu) and then remove the snapshot from the root partition. After a reboot the system is ok (but without a snapshot based restore option).

The Issuse was also discussed in CentOS-Forum

thanks ...
TagsNo tags attached.
Attached Files




kabe (reporter)

I succeeded in reproducing something similar.

- Install CentOS 7 with default disk allocation.
  Default Volume Group: "cl"
  Root Logical Volume: "cl/root"
  Swap Logical Volume: "cl/swap"

- After reboot, erase the swap to free up the Volume Group.
  # swapoff /dev/mapper/swap
  # lvremove cl/swap

- Delete /dev/mapper/cl-swap entry from /etc/fstab.

- Make a non-thin snapshot of root partition (this seems optional)
  # lvcreate -s cl/root -n root_snap -L 500m

- Reboot.

The initramfs will stop after
    File System Check on /dev/mapper/cl-root
(Actually, it is waiting for /bin/dracut-initqueue to finish)

After 120 seconds ((rd.retry=180)/3*2),
    dracut-initqueue[257]: Warning: dracut-initqueue timeout - starting timeout scripts
will be repeated in (sleep 0.5) interval.

After 180 seconds, my installation resumed normal booting.

There was a kernel commandline option remaining:
so, during boot, dracut-initqueue seems to be waiting for /dev/cl/swap to appear.

But you deleted the swap, so it never appears.
So, dracut-initqueue tries to wait for 180 seconds and gives up.

Deleting "rd.lvm.lv=cl/swap" option from /boot/grub2/grub.cfg
fixed my installation.
(Yours may have "rd.lvm.lv=vgsys/swap")
Boots flulently, even with root partition snapshot existing.

Your installation dropped to rdshell, mine resumed normal booting.
I don't know how to exactly reproduce the symptom described in

The timeout could be adjusted by "rd.retry=180" kernel option.

* What do you have on kernel boot option in /boot/grub2/grub.cfg?
* What is current /etc/fstab content?

-Issue History
Date Modified Username Field Change
2016-12-18 14:44 sz1305 New Issue
2017-03-23 07:43 kabe Note Added: 0028923
+Issue History