Bug 710368

Summary:	Bind mount on nfsexports and mount of boot partition
Product:	[Fedora] Fedora	Reporter:	Thomas Bartschies <tom>
Component:	systemd	Assignee:	systemd-maint
Status:	CLOSED INSUFFICIENT_DATA	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	15	CC:	harald, johannbg, johannbg, lpoetter, metherid, mschmidt, notting, plautrba
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:	2012-01-25 16:24:55 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Thomas Bartschies 2011-06-03 07:58:16 UTC

With Fedora 15 Kernel 2.6.38.6-27 some new problems arised.

My /boot partition could not be mounted any more in the boot process. Later
this was entirely possible. With the previous rc1 kernel, there was no such
problem. Because of it, the boot process goes into emergency mode every time.

The boot partition is reachable through a multipath dm device. So it's device is /dev/mapper/mpathap1. May be the mp device isn't ready, when systemd tries to
mount it. Adding the _netdev mount option to fstab cures the problem, because
it postpones the mount.

Similar thing with a bind mount, but on shutdown. systemd hangs until it is
removed, or the _netdev and nofail options are added to the mount options.
Here are the non-working fstab entries:

/home/tom   /nfsexports/tom     none  bind,rw  0 0

and

UUID=xxxxx-xxxxxx-xxxxxx...  /boot     ext4  defaults  1 2

Is it now necessary to use the _netdev option for such mounts?

Comment 1 Thomas Bartschies 2011-06-06 10:09:38 UTC

In both cases, adding the _netdev option, causes the netfs.service to fail.
Specifically the parts for checking the "network attached filesystems" and
"Mounting other filesystems". The root cause is always, that one of the affected
filesystems is already mounted.

This also seems to include the devpts, shm and sysfs mounts. Apparently systemd
mounts these by itself, very early in the boot process. Therefore all subsequent 
mounts by the fedora scripts must fail.

Comment 2 Michal Schmidt 2011-07-04 11:27:20 UTC

Please add "log_buf_len=1M systemd.log_level=debug systemd.log_target=kmsg" to your kernel command line and when you get to the emergency shell, save the output of the "dmesg" command and then attach it here. Thanks.

Comment 3 Fedora Admin XMLRPC Client 2011-10-20 16:27:57 UTC

This package has changed ownership in the Fedora Package Database.  Reassigning to the new owner of this component.