|
|
|
#!/bin/sh
|
|
|
|
|
|
|
|
# run lvm scan if udev has settled
|
|
|
|
|
|
|
|
extraargs="$@"
|
|
|
|
type getarg >/dev/null 2>&1 || . /lib/dracut-lib.sh
|
|
|
|
|
|
|
|
VGS=$(getargs rd.lvm.vg -d rd_LVM_VG=)
|
|
|
|
LVS=$(getargs rd.lvm.lv -d rd_LVM_LV=)
|
|
|
|
SNAPSHOT=$(getargs rd.lvm.snapshot -d rd_LVM_SNAPSHOT=)
|
|
|
|
SNAPSIZE=$(getargs rd.lvm.snapsize -d rd_LVM_SNAPSIZE=)
|
|
|
|
|
|
|
|
[ -d /etc/lvm ] || mkdir -m 0755 -p /etc/lvm
|
|
|
|
# build a list of devices to scan
|
|
|
|
lvmdevs=$(
|
|
|
|
for f in /tmp/.lvm_scan-*; do
|
|
|
|
[ -e "$f" ] || continue
|
|
|
|
printf '%s' "${f##/tmp/.lvm_scan-} "
|
|
|
|
done
|
|
|
|
)
|
|
|
|
|
|
|
|
if [ ! -e /etc/lvm/lvm.conf ]; then
|
|
|
|
{
|
|
|
|
echo 'devices {';
|
|
|
|
printf ' filter = [ '
|
|
|
|
for dev in $lvmdevs; do
|
|
|
|
printf '"a|^/dev/%s$|", ' $dev;
|
|
|
|
done;
|
|
|
|
echo '"r/.*/" ]';
|
|
|
|
echo '}';
|
|
|
|
|
|
|
|
# establish LVM locking
|
|
|
|
if [ -n $SNAPSHOT ]; then
|
|
|
|
echo 'global {';
|
|
|
|
echo ' locking_type = 1';
|
|
|
|
echo ' use_lvmetad = 0';
|
|
|
|
echo '}';
|
|
|
|
else
|
|
|
|
echo 'global {';
|
|
|
|
echo ' locking_type = 4';
|
|
|
|
echo ' use_lvmetad = 0';
|
|
|
|
echo '}';
|
|
|
|
fi
|
|
|
|
} > /etc/lvm/lvm.conf
|
|
|
|
lvmwritten=1
|
|
|
|
fi
|
|
|
|
|
|
|
|
check_lvm_ver() {
|
|
|
|
maj=$1
|
|
|
|
min=$2
|
|
|
|
ver=$3
|
|
|
|
# --poll is supported since 2.2.57
|
|
|
|
[ $4 -lt $maj ] && return 1
|
|
|
|
[ $4 -gt $maj ] && return 0
|
|
|
|
[ $5 -lt $min ] && return 1
|
|
|
|
[ $5 -gt $min ] && return 0
|
|
|
|
[ $6 -ge $ver ] && return 0
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
|
|
|
# hopefully this output format will never change, e.g.:
|
|
|
|
# LVM version: 2.02.53(1) (2009-09-25)
|
|
|
|
OLDIFS=$IFS
|
|
|
|
IFS=.
|
|
|
|
set $(lvm version 2>/dev/null)
|
|
|
|
IFS=$OLDIFS
|
|
|
|
maj=${1##*:}
|
|
|
|
min=$2
|
|
|
|
sub=${3%% *}
|
|
|
|
sub=${sub%%\(*};
|
|
|
|
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm_ignorelockingfailure="--ignorelockingfailure"
|
|
|
|
lvm_quirk_args="--ignorelockingfailure --ignoremonitoring"
|
|
|
|
|
|
|
|
check_lvm_ver 2 2 57 $maj $min $sub && \
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm_quirk_args="$lvm_quirk_args --poll n"
|
|
|
|
|
|
|
|
if check_lvm_ver 2 2 65 $maj $min $sub; then
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm_quirk_args=" --sysinit $extraargs"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if check_lvm_ver 2 2 221 $maj $min $sub; then
|
|
|
|
lvm_quirk_args=" $extraargs"
|
|
|
|
unset lvm_ignorelockingfailure
|
|
|
|
fi
|
|
|
|
|
|
|
|
unset extraargs
|
|
|
|
|
|
|
|
export LVM_SUPPRESS_LOCKING_FAILURE_MESSAGES=1
|
|
|
|
|
|
|
|
if [ -n "$SNAPSHOT" ] ; then
|
|
|
|
# HACK - this should probably be done elsewhere or turned into a function
|
|
|
|
# Enable read-write LVM locking
|
|
|
|
sed -i -e 's/\(^[[:space:]]*\)locking_type[[:space:]]*=[[:space:]]*[[:digit:]]/\1locking_type = 1/' ${initdir}/etc/lvm/lvm.conf
|
|
|
|
|
|
|
|
# Expected SNAPSHOT format "<orig lv name>:<snap lv name>"
|
|
|
|
ORIG_LV=${SNAPSHOT%%:*}
|
|
|
|
SNAP_LV=${SNAPSHOT##*:}
|
|
|
|
|
|
|
|
info "Removing existing LVM snapshot $SNAP_LV"
|
|
|
|
lvm lvremove --force $SNAP_LV 2>&1| vinfo
|
|
|
|
|
|
|
|
# Determine snapshot size
|
|
|
|
if [ -z "$SNAPSIZE" ] ; then
|
|
|
|
SNAPSIZE=$(lvm lvs --noheadings --units m --options lv_size $ORIG_LV)
|
|
|
|
info "No LVM snapshot size provided, using size of $ORIG_LV ($SNAPSIZE)"
|
|
|
|
fi
|
|
|
|
|
|
|
|
info "Creating LVM snapshot $SNAP_LV ($SNAPSIZE)"
|
|
|
|
lvm lvcreate -s -n $SNAP_LV -L $SNAPSIZE $ORIG_LV 2>&1| vinfo
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -n "$LVS" ] ; then
|
|
|
|
info "Scanning devices $lvmdevs for LVM logical volumes $LVS"
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm lvscan $lvm_ignorelockingfailure 2>&1 | vinfo
|
|
|
|
for LV in $LVS; do
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm lvchange --yes -ay $lvm_quirk_args $LV 2>&1 | vinfo
|
|
|
|
done
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$LVS" -o -n "$VGS" ]; then
|
|
|
|
info "Scanning devices $lvmdevs for LVM volume groups $VGS"
|
lvm: remove all quirk arguments
https://bugzilla.redhat.com/show_bug.cgi?id=742564
Once lvm2 starts using /run (bug 742554), it should be no longer
necessary to disable file-based locking in the vgchange call in
fedora-storage-init.
Removing '--sysinit' will make it safe to call LVM operations
concurrently from other units.
The --sysinit is a compound option consisting of:
-> --ignorelockingfailure - not needed anymore, the /run/lock/lvm is
available rw soon in boot process
-> --ignoremonitoring - not needed since /run is available, this would
require the dm-event.service to be run before
fedora-storage-init.service (and new lvm2-activation.service when
deployed). But that's a one line change - I'll have a look whether it
plays well with other services and if yes, I'll commit the change. N.B.:
This has a consequence that all volumes activated on vgchange -ay will
be monitored at the same time they're activated (which is a plus I
think). The lvm2-monitor will just grab all the other volumes not
activated at the time of the boot's vgchange -ay call. But that's not an
issue (for already monitored volumes, calling vgchange --monitor y will
just be a NOOP).
-> --poll n - not needed, we can run the polldaemon as the /run is
available and rw (in case there's unfinished merge or mirror sync from
previous system run, the poll-daemon will be triggered at boot now).
Actually, the polldaemon should be triggered as a service like dmeventd
is, not forked off from the LVM command itself, like from vgchange in
this case - we still need to change this - there's a bug open for this
request already (bug #814857). However, we don't have this feature ready
yet so I need to check whether this is OK with the early boot process
with the current state.
9 years ago
|
|
|
lvm vgscan $lvm_ignorelockingfailure 2>&1 | vinfo
|
|
|
|
lvm vgchange -ay $lvm_quirk_args $VGS 2>&1 | vinfo
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$lvmwritten" ]; then
|
|
|
|
rm -f -- /etc/lvm/lvm.conf
|
|
|
|
fi
|
|
|
|
unset lvmwritten
|
|
|
|
|
|
|
|
udevadm settle
|
|
|
|
|
|
|
|
need_shutdown
|