| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
| |
When monitoring, skip exported VGs without causing a command
failure.
The lvm2-monitor service runs 'vgchange --monitor y', so
any exported VG on the system would cause the service to
fail.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Do this at two levels, although one would be enough to
fix the problem seen recently:
- Ignore any reported sector size other than 512 of 4096.
If either sector size (physical or logical) is reported
as 512, then use 512. If neither are reported as 512,
and one or the other is reported as 4096, then use 4096.
If neither is reported as either 512 or 4096, then use 512.
- When rounding up a limited write in bcache to be a multiple
of the sector size, check that the resulting write size is
not larger than the bcache block itself. (This shouldn't
happen if the sector size is 512 or 4096.)
|
|
|
|
|
| |
Should fix a reported problem of pvs seeing bad orphan
info while vgrename was running.
|
|
|
|
|
|
|
|
|
|
|
| |
This allows the creation of a striped mirror leg(s) during upconvert
by adding lvconvert command line options --stripes/--stripesize
for 'mirror' to tools/command-lines.in.
In case multiple mirror legs are being added, all will have the
same requested striped layout.
Resolves: rhbz1720705
|
|
|
|
|
|
|
|
|
|
| |
Some older BB with older cryptsetup tool do not 'retry' on remove
and when remove is issued right after 'fsck' - it might be
rejected with:
Device @PREFIX@-tcrypt2 is busy.
Try to use udevadm settle.
|
|
|
|
| |
mkfs.xfs was rejecting previously working value
|
|
|
|
| |
Here we want args to be splited into individual strings.
|
|
|
|
|
|
| |
Since we do not need anywhere luks2 - pick older format
which does not require password for resize to keep
the rest of test unmodified.
|
|
|
|
| |
mkfs.xfs now needs at least ~1600...
|
|
|
|
|
| |
Include: toolcontext.h -> dev-type.h -> label.h -> toolcontext.h
Replace with struct predeclaration.
|
|
|
|
| |
Another missed release on error path.
|
|
|
|
|
| |
Some older gcc versions shows this (FP) warning:
label/label.c:360: warning: â€sector’ may be used uninitialized in this function
|
|
|
|
|
| |
misc/lvm-globals.c:76: warning: declaration of â€use_aio’ shadows a global declaration
misc/lvm-globals.h:63: warning: shadowed declaration is here
|
|
|
|
|
|
|
| |
Kernels <2.6.27 don't have /sys/dev dir - add code for looking
out device name via longre seach in /sys/block
This makes commands like 'dmsetup dep -o blkdevname' working.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We've been assigning this in 69-dm-lvm-metad.rules:
ENV{ID_MODEL}="LVM PV $env{ID_FS_UUID_ENC} on /dev/$name"
This was for the description to appear for each systemd device
unit representing this device, for example:
$systemctl -a | grep "LVM PV"
dev-block-252:2.device loaded active plugged LVM PV JhxC7B-YTgk-3jIU-5GVo-c4gV-W8t3-UUz06p on /dev/vda2 2
dev-disk-by\x2did-lvm\x2dpv\x2duuid\x2dJhxC7B\x2dYTgk\x2d3jIU\x2d5GVo\x2dc4gV\x2dW8t3\x2dUUz06p.device loaded active plugged LVM PV JhxC7B-YTgk-3jIU-5GVo-c4gV-W8t3-UUz06p on /dev/vda2 2
...
However, there could be an actual ID_MODEL that people are interested in
more than the fact that this is an LVM PV and so we shouldn't overwrite
the value.
Also, we already have a symlink /dev/disk/by-id/lvm-pv-uuid-<PV_UUID>
created which is then reflected as device unit (all device's symlinks
have systemd device unit representation) so we can still reach this
information in systemd unit listings even without setting the ID_MODEL.
Reported here: https://github.com/lvmteam/lvm2/issues/21
|
|
|
|
|
|
| |
These commands were looking for the requested device alias
before dev_cache_scan had created the list of devs on the
system, so they would fail and report the dev wasn't found.
|
| |
|
|
|
|
|
| |
Just make it clear to analyzers cname can't be NULL.
TODO: maybe exclude NULL at front of the function...
|
|
|
|
| |
As _init_free_list() cannot accept negative numbers
|
|
|
|
| |
Make the code more obvious.
|
|
|
|
| |
Add missing stack reports on error paths.
|
|
|
|
|
| |
As we check for existince on entering path
let's ensure it's there also on exit path.
|
|
|
|
| |
Use lv_info results only when valid.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Add missing unlock call on theretical error path where
we would be missing our configured command.
|
|
|
|
| |
So to check the reason of failure is not influenced by missing cmirrord.
|
|
|
|
| |
Older kernels provide fixed value 512 instead of 0.
|
|
|
|
|
| |
Since with smq policy we can still pick format v2 - enforce
format v1 explicitely.
|
| |
|
|
|
|
|
| |
Cluster test would fail without exlusively activated mirror.
FIXME: test is still broken, since LV2 is not active after split.
|
|
|
|
|
| |
If the /usr/bin/time command is not installed, skip test
(do not want to use shells' built-in).
|
|
|
|
|
| |
Since we now properly extend also _pmspare - there was not enough free
space to add 8extents to both volumes.
|
|
|
|
| |
scan_lvs now automatically comes with extend_filter_LVMTEST
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For older bashes (i.e. rhel5) code can't use internal bash arrays
and uses old-school greps and sorts to create configuration file.
But the split of value was mismatching names with same prefix
so i.e.
devices/scan
devices/scan_lvs
was matching always the later one - fix this by expecting
whitespace or '=' after the name.
|
|
|
|
|
|
|
|
| |
This reverts commit 8623e336513c6ac6fcb24aa7e2ef10b8a3a36c59.
The problem this was avoiding now seems to be fixed by
commit d6bce036155ae973c869bdce3ca5f824f933f599
mirror: fix monitoring change
|
| |
|
|
|
|
|
| |
Since with some installed package like 'tuned' the value of 'RA' on
PV origin device can be different, adapting tests to count with this.
|
|
|
|
|
|
|
|
|
|
|
| |
Since we use 'set -euE -o pipefail' for shell execution,
any failure of any command in the 'piped' shell can result
in failure of whole executed chain - resulting in typically
unsually test skip, that was left unnoticed.
Since checked command have usually short output, the simplest
fix seems to be to let grep parse whole output instead
of quiting after first match.
|
|
|
|
|
|
|
|
|
|
| |
Commit a8921be641afe865c177e11b8859f4b937f76995 was supposedly a fix
for unwanted table reload - however before final commit, the tiny
change has been made that was believed to be an enhancment
of original prepared patch. Unfortunatelly the function
locking_is_clustered is not meant to be an equivalent test
of original function. Drop this change and using
original patch.
|
|
|
|
|
|
|
| |
Avoid bug 1711427 in which an exclusively active linear LV in a
clustered VG, when upconverted to a mirror, will mistakenly use
a userspace/cmirror log in _add_log() beause laopts.exclusive is
not set.
|
|
|
|
|
|
|
|
|
| |
Save the previous duplicate PVs in a global list instead
of a list on the cmd struct. dmeventd reuses the cmd struct
for multiple commands, and the list entries between commands
were being freed (apparently), causing a segfault in dmeventd
when it tried to use items in cmd->unused_duplicate_devs
that had been saved there by the previous command.
|
|
|
|
| |
Check reporting works
|
|
|
|
| |
When using 'aux extend_filter' we always want to use LV as PV.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Recent kernel version from kernel commit:
de7180ff908b2bc0342e832dbdaa9a5f1ecaa33a
started to report in cache status line new flag:
no_discard_passdown
Whenever lvm spots unknown status it reports:
Unknown feature in status:
So add reconginzing this feature flag and also report this with
'lvs -o+kernel_discards'
When no_discard_passdown is found in status 'nopassdown' gets reported
for this field (roughly matching what we report for thin-pools).
|
|
|
|
|
|
|
| |
Since migration_threshold is now protected to not be smaller
then 8*chunk_size - update tests to count with this modification.
(cherry picked from commit 2cac4bfc150ea1e5459580e3bac4907446ea7f3a)
|