Bug 1592783

Summary: Logical volumes cannot be found so system won't boot
Product: [Fedora] Fedora Reporter: Tom Yates <madhatter>
Component: lvm2Assignee: LVM and device-mapper development team <lvm-team>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 28CC: agk, anprice, bmarzins, bmr, cfeist, dracut-maint-list, harald, heinzm, jonathan, kzak, lvm-team, madhatter, mcsontos, msnitzer, prajnoha, prockai, teigland, zbyszek, zkabelac
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-07-09 12:51:27 UTC Type: Bug
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
This is the dracut SOS report file none

Description Tom Yates 2018-06-19 10:02:20 UTC
Created attachment 1452866 [details]
This is the dracut SOS report file

Description of problem: when rebooting either with 4.16.5-300 or 4.16.15-300, the boot gets about 15 seconds in, goes very quiet, then puts up a dracut timeout warning after about a further 140s.  Eventually it drops to the dracut shell with the warnings

Warning: /dev/VolGroup01/LogVol00 does not exist
Warning: /dev/VolGroup01/LogVol01 does not exist
Warning: /dev/mapper/VolGroup01-LogVol01 does not exist

I can boot happily under the remaining old F26 kernel, 4.15.17-200.

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


How reproducible: for me, very.


Steps to Reproduce:
1. reboot
2. don't select old F26 kernel at grub menu
3.

Actual results: system fails to boot, drops to dracut shell


Expected results: system should boot


Additional info: dracut recommended I save a copy of the rdsosreport.txt file, which is attached.

Comment 1 Harald Hoyer 2018-07-05 13:14:00 UTC
Please check, if your MD_UUID / rd.md.uuid is valid. We had some reports about the MD_UUID changing.

What is the output of:

# dracut --print-cmdline

# for i in /dev/md*; do mdadm -D --export $i; done | grep MD_UUID

Comment 2 Tom Yates 2018-07-05 13:18:32 UTC
[root@risby ~]# dracut --print-cmdline
 rd.lvm.lv=VolGroup01/LogVol01 
 rd.lvm.lv=VolGroup01/LogVol00 
 rd.md.uuid=51e28529:17ef962c:8df82e22:eb3716d4  rd.md.uuid=98c9bd70:7521058d:b1062995:9cd227d7  resume=/dev/mapper/VolGroup01-LogVol00 root=/dev/mapper/VolGroup01-LogVol01 rootfstype=ext4 rootflags=rw,relatime,data=ordered
[root@risby ~]# for i in /dev/md*; do mdadm -D --export $i; done | grep MD_UUID
MD_UUID=51e28529:17ef962c:8df82e22:eb3716d4
MD_UUID=98c9bd70:7521058d:b1062995:9cd227d7

Comment 3 Harald Hoyer 2018-07-05 13:23:16 UTC
[   15.640057] risby dracut-initqueue[362]: Scanning devices md127  for LVM logical volumes VolGroup01/LogVol00 VolGroup01/LogVol01
[   15.690730] risby dracut-initqueue[362]: WARNING: PV jmpqN0-1RqJ-3UJq-3gG2-Lz4I-yHJZ-pi0K3o on /dev/sdb2 was already found on /dev/sda2.
[   15.697382] risby dracut-initqueue[362]: WARNING: PV jmpqN0-1RqJ-3UJq-3gG2-Lz4I-yHJZ-pi0K3o on /dev/md127 was already found on /dev/sda2.
[   15.699154] risby dracut-initqueue[362]: WARNING: PV jmpqN0-1RqJ-3UJq-3gG2-Lz4I-yHJZ-pi0K3o prefers device /dev/sda2 because device was seen first.
[   15.700944] risby dracut-initqueue[362]: WARNING: PV jmpqN0-1RqJ-3UJq-3gG2-Lz4I-yHJZ-pi0K3o prefers device /dev/sda2 because device was seen first.

lvm really shouldn't touch the raid parts! reassigning

Comment 4 Marian Csontos 2018-07-09 07:35:43 UTC
What's lvm2 version? This looks like yet another duplicate of Bug 1575762. Do you still see the problem after update of lvm2 to lvm2-2.02.177-5.fc28 and regenerating initramfs using `dracut -f` ?

Comment 5 Tom Yates 2018-07-09 10:24:40 UTC
It was indeed lvm2-2.02.177-4.fc28.  Having updated that package, brought the kernel up-to-patch, and regenerated the appropriate file with `dracut -f --kver 4.17.3-200-fc28.x86_64` my system has now happily booted into 4.17.3-200.fc28, so I guess you are right, and this is a duplicate of the referenced bug, and can be closed as such.  I apologise for not having found the other bug when searching before logging this one.

Comment 6 Marian Csontos 2018-07-09 12:51:27 UTC
Thanks for verifying it

*** This bug has been marked as a duplicate of bug 1575762 ***