IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
There have been two file locks used to protect lvm
"global state": "ORPHANS" and "GLOBAL".
Commands that used the ORPHAN flock in exclusive mode:
pvcreate, pvremove, vgcreate, vgextend, vgremove,
vgcfgrestore
Commands that used the ORPHAN flock in shared mode:
vgimportclone, pvs, pvscan, pvresize, pvmove,
pvdisplay, pvchange, fullreport
Commands that used the GLOBAL flock in exclusive mode:
pvchange, pvscan, vgimportclone, vgscan
Commands that used the GLOBAL flock in shared mode:
pvscan --cache, pvs
The ORPHAN lock covers the important cases of serializing
the use of orphan PVs. It also partially covers the
reporting of orphan PVs (although not correctly as
explained below.)
The GLOBAL lock doesn't seem to have a clear purpose
(it may have eroded over time.)
Neither lock correctly protects the VG namespace, or
orphan PV properties.
To simplify and correct these issues, the two separate
flocks are combined into the one GLOBAL flock, and this flock
is used from the locking sites that are in place for the
lvmlockd global lock.
The logic behind the lvmlockd (distributed) global lock is
that any command that changes "global state" needs to take
the global lock in ex mode. Global state in lvm is: the list
of VG names, the set of orphan PVs, and any properties of
orphan PVs. Reading this global state can use the global lock
in sh mode to ensure it doesn't change while being reported.
The locking of global state now looks like:
lockd_global()
previously named lockd_gl(), acquires the distributed
global lock through lvmlockd. This is unchanged.
It serializes distributed lvm commands that are changing
global state. This is a no-op when lvmlockd is not in use.
lockf_global()
acquires an flock on a local file. It serializes local lvm
commands that are changing global state.
lock_global()
first calls lockf_global() to acquire the local flock for
global state, and if this succeeds, it calls lockd_global()
to acquire the distributed lock for global state.
Replace instances of lockd_gl() with lock_global(), so that the
existing sites for lvmlockd global state locking are now also
used for local file locking of global state. Remove the previous
file locking calls lock_vol(GLOBAL) and lock_vol(ORPHAN).
The following commands which change global state are now
serialized with the exclusive global flock:
pvchange (of orphan), pvresize (of orphan), pvcreate, pvremove,
vgcreate, vgextend, vgremove, vgreduce, vgrename,
vgcfgrestore, vgimportclone, vgmerge, vgsplit
Commands that use a shared flock to read global state (and will
be serialized against the prior list) are those that use
process_each functions that are based on processing a list of
all VG names, or all PVs. The list of all VGs or all PVs is
global state and the shared lock prevents those lists from
changing while the command is processing them.
The ORPHAN lock previously attempted to produce an accurate
listing of orphan PVs, but it was only acquired at the end of
the command during the fake vg_read of the fake orphan vg.
This is not when orphan PVs were determined; they were
determined by elimination beforehand by processing all real
VGs, and subtracting the PVs in the real VGs from the list
of all PVs that had been identified during the initial scan.
This is fixed by holding the single global lock in shared mode
while processing all VGs to determine the list of orphan PVs.
When a single copy of metadata gets within 1MB of the
current io_memory_size value, begin printing a warning
that the io_memory_size should be increased.
Save the list of PVs in /run/lvm/hints. These hints
are used to reduce scanning in a number of commands
to only the PVs on the system, or only the PVs in a
requested VG (rather than all devices on the system.)
Native disk scanning is now both reduced and
async/parallel, which makes it comparable in
performance (and often faster) when compared
to lvm using lvmetad.
Autoactivation now uses local temp files to record
online PVs, and no longer requires lvmetad.
There should be no apparent command-level change
in behavior.
When lvmetad is not used, use temporary files to record
which PVs have appeared. Use these temp files to determine
when a VG is complete, to trigger autoactivation.
This change allows us to remove lvmetad while keeping the
same autoactivation behavior that lvmetad provides.
The temp files are created in /run/lvm/pvs_online/ and are
named for the PVID of the PV. The files contain the
major:minor of the device the PV was read from.
e.g. if VG foo has dev1 and dev2, then:
. pvscan --cache -aay dev1
reads vg metadata from dev1
creates /run/lvm/pvs_online/<pvid-of-dev1>
checks if all vg->pvs are online: no
. pvscan --cache -aay dev2
reads vg metadata from dev2
creates /run/lvm/pvs_online/<pvid-of-dev2>
checks if all vg->pvs are online: yes
autoactivates vg
A 'pvscan --cache dev' (without -aay) still records that
dev is online.
A 'pvscan --cache --major X --minor Y' after a device is
gone will remove the temp file for it.
A 'pvscan --cache [-aay]' (no devs) resets the state of
temp files by removing them all, then scanning all devs
and creating temp files for PVs that are found.
If no online files exist, the first pvscan --cache scans
all devs and creates temp files for any PVs found.
The scope of the temp files is only pvscan, and they are only
used for pvscan-based autoactivation. No other commands are
concerned with or aware of these temp files. When lvm creates
or removes PVs, no attempt is made to update the temp files.
The md filter can operate in two native modes:
- normal: reads only the start of each device
- full: reads both the start and end of each device
md 1.0 devices place the superblock at the end of the device,
so components of this version will only be identified and
excluded when lvm uses the full md filter.
Previously, the full md filter was only used in commands
that could write to the device. Now, the full md filter
is also applied when there is an md 1.0 device present
on the system. This means the 'pvs' command can avoid
displaying md 1.0 components (at the cost of doubling
the i/o to every device on the system.)
(The md filter can operate in a third mode, using udev,
but this is disabled by default because there have been
problems with reliability of the info returned from udev.)
A few places were calling a function to check if a
VG lock was held. The only place it was actually
needed is for pvcreate which wants to do its own
locking (and scanning) around process_each_pv.
The locking/scanning exceptions for pvcreate in
process_each_pv/vg_read can be enabled by just passing
a couple of flags instead of checking if the VG is
already locked. This also means that these special
cases won't be enabled unknowingly in other places
where they shouldn't be used.
Four commands lock two VGs at a time:
- vgsplit and vgmerge already have their own logic to
acquire the locks in the correct order.
- vgimportclone and vgrename disable this ordering check.
We have been warning about duplicate devices (and disabling lvmetad)
immediately when the dup was detected (during label_scan). Move the
warnings (and the disabling) to happen later, after label_scan is
finished.
This lets us avoid an unwanted warning message about duplicates
in the special case were md components are eliminated during the
duplicate device resolution.
As we start refactoring the code to break dependencies (see doc/refactoring.txt),
I want us to use full paths in the includes (eg, #include "base/data-struct/list.h").
This makes it more obvious when we're breaking abstraction boundaries, eg, including a file in
metadata/ from base/
md devices using an older superblock version have
superblocks at the end of the md device. For commands
that skip reading the end of devices during filtering,
the md component devs will be scanned, and will appear
as duplicate PVs to the original md device. Remove
these md components from the list of unused duplicate
devices, so they are treated as if they had been
ignored during filtering. This avoids the restrictions
that are placed on using PVs with duplicates.
Filters are still applied before any device reading or
the label scan, but any filter checks that want to read
the device are skipped and the device is flagged.
After bcache is populated, but before lvm looks for
devices (i.e. before label scan), the filters are
reapplied to the devices that were flagged above.
The filters will then find the data they need in
bcache.
The clvmd saved_vg data is independent from the normal lvm
lvmcache vginfo data, so separate saved_vg from vginfo.
Normal lvm doesn't need to use save_vg at all, and in clvmd,
lvmcache changes on vginfo can be made without worrying
about unwanted effects on saved_vg.
To avoid the chance of freeing a saved vg while another
code path is using it, defer freeing saved vgs until
all the lvmcache content is dropped for the vg.
There are likely more bits of code that can be removed,
e.g. lvm1/pool-specific bits of code that were identified
using FMT flags.
The vgconvert command can likely be reduced further.
The lvm1-specific config settings should probably have
some other fields set for proper deprecation.
In some pvmove tests, clvmd uses the new (precommitted)
saved_vg, but then requests the old saved_vg, and
expects that the new saved_vg be returned instead of
the old. So, when returning the new saved_vg, forget
the old one so we don't return it again.
The filters save information about devices that should
be ignored, so if we need to repeat a scan (unusual,
but happens in clvmd), we need to update the filters.
After reading a VG, stash it in lvmcache as "saved_vg".
Before reading the VG again, try to use the saved_vg.
The saved_vg is dropped on VG lock operations.
The copy of the VG which clvmd stashes in lvmcache should
not only be used between suspend and resume, but between
sequential LV operations in clvmd, so that clvmd does not
need to reread the VG for each one. Prepare for that by
renaming the stashed VG as "saved_vg".
For reporting commands (pvs,vgs,lvs,pvdisplay,vgdisplay,lvdisplay)
we do not need to repeat the label scan of devices in vg_read if
they all had matching metadata in the initial label scan. The
data read by label scan can just be reused for the vg_read.
This cuts the amount of device i/o in half, from two reads of
each device to one. We have to be careful to avoid repairing
the VG if we've skipped rescanning. (The VG repair code is very
poor, and will be redone soon.)
Recent changes allow some major simplification of the way
lvmcache works and is used. lvmcache_label_scan is now
called in a controlled fashion at the start of commands,
and not via various unpredictable side effects. Remove
various calls to it from other places. lvmcache_label_scan
should not be called from anywhere during a command, because
it produces an incorrect representation of PVs with no MDAs,
and misclassifies them as orphans. This has been a long
standing problem. The invalid flag and rescanning based on
that is no longer used and removed. The 'force' variation is
no longer needed and removed.
Create a new dev->bcache_fd that the scanning code owns
and is in charge of opening/closing. This prevents other
parts of lvm code (which do various open/close) from
interfering with the bcache fd. A number of dev_open
and dev_close are removed from the reading path since
the read path now uses the bcache.
With that in place, open(O_EXCL) for pvcreate/pvremove
can then be fixed. That wouldn't work previously because
of other open fds.
The copy of VG metadata stored in lvmcache was not being used
in general. It pretended to be a generic VG metadata cache,
but was not being used except for clvmd activation. There
it was used to avoid reading from disk while devices were
suspended, i.e. in resume.
This removes the code that attempted to make this look
like a generic metadata cache, and replaces with with
something narrowly targetted to what it's actually used for.
This is a way of passing the VG from suspend to resume in
clvmd. Since in the case of clvmd one caller can't simply
pass the same VG to both suspend and resume, suspend needs
to stash the VG somewhere that resume can grab it from.
(resume doesn't want to read it from disk since devices
are suspended.) The lvmcache vginfo struct is used as a
convenient place to stash the VG to pass it from suspend
to resume, even though it isn't related to the lvmcache
or vginfo. These suspended_vg* vginfo fields should
not be used or touched anywhere else, they are only to
be used for passing the VG data from suspend to resume
in clvmd. The VG data being passed between suspend and
resume is never modified, and will only exist in the
brief period between suspend and resume in clvmd.
suspend has both old (current) and new (precommitted)
copies of the VG metadata. It stashes both of these in
the vginfo prior to suspending devices. When vg_commit
is successful, it sets a flag in vginfo as before,
signaling the transition from old to new metadata.
resume grabs the VG stashed by suspend. If the vg_commit
happened, it grabs the new VG, and if the vg_commit didn't
happen it grabs the old VG. The VG is then used to resume
LVs.
This isolates clvmd-specific code and usage from the
normal lvm vg_read code, making the code simpler and
the behavior easier to verify.
Sequence of operations:
- lv_suspend() has both vg_old and vg_new
and stashes a copy of each onto the vginfo:
lvmcache_save_suspended_vg(vg_old);
lvmcache_save_suspended_vg(vg_new);
- vg_commit() happens, which causes all clvmd
instances to call lvmcache_commit_metadata(vg).
A flag is set in the vginfo indicating the
transition from the old to new VG:
vginfo->suspended_vg_committed = 1;
- lv_resume() needs either vg_old or vg_new
to use in resuming LVs. It doesn't want to
read the VG from disk since devices are
suspended, so it gets the VG stashed by
lv_suspend:
vg = lvmcache_get_suspended_vg(vgid);
If the vg_commit did not happen, suspended_vg_committed
will not be set, and in this case, lvmcache_get_suspended_vg()
will return the old VG instead of the new VG, and it will
resume LVs based on the old metadata.
When process_each_pv() calls vg_read() on the orphan VG, the
internal implementation was doing an unnecessary
lvmcache_label_scan() and two unnecessary label_read() calls
on each orphan. Some of those unnecessary label scans/reads
would sometimes be skipped due to caching, but the code was
always doing at least one unnecessary read on each orphan.
The common format_text case was also unecessarily calling into
the format-specific pv_read() function which actually did nothing.
By analyzing each case in which vg_read() was being called on
the orphan VG, we can say that all of the label scans/reads
in vg_read_orphans are unnecessary:
1. reporting commands: the information saved in lvmcache by
the original label scan can be reported. There is no advantage
to repeating the label scan on the orphans a second time before
reporting it.
2. pvcreate/vgcreate/vgextend: these all share a common
implementation in pvcreate_each_device(). That function
already rescans labels after acquiring the orphan VG lock,
which ensures that the command is using valid lvmcache
information.
When lvmlockd indicates that the lvmetad cache is out of
date because of changes by another node, lvmetad_pvscan_vg()
rescans the devices in the VG to update lvmetad. Use the
new label_scan in this function to use the common code and
take advantage of the new aio and reduced reads.
Move the location of scans to make it clearer and avoid
unnecessary repeated scanning. There should be one scan
at the start of a command which is then used through the
rest of command processing.
Previously, the initial label scan was called as a side effect
from various utility functions. This would lead to it being called
unnecessarily. It is an expensive operation, and should only be
called when necessary. Also, this is a primary step in the
function of the command, and as such it should be called prominently
at the top level of command processing, not as a hidden side effect
of a utility function. lvm knows exactly where and when the
label scan needs to be done. Because of this, move the label scan
calls from the internal functions to the top level of processing.
Other specific instances of lvmcache_label_scan() are still called
unnecessarily or unclearly by specific commands that do not use
the common process_each functions. These will be improved in
future commits.
During the processing phase, rescanning labels for devices in a VG
needs to be done after the VG lock is acquired in case things have
changed since the initial label scan. This was being done by way
of rescanning devices that had the INVALID flag set in lvmcache.
This usually approximated the right set of devices, but it was not
exact, and obfuscated the real requirement. Correct this by using
a new function that rescans the devices in the VG:
lvmcache_label_rescan_vg().
Apart from being inexact, the rescanning was extremely well hidden.
_vg_read() would call ->create_instance(), _text_create_text_instance(),
_create_vg_text_instance() which would call lvmcache_label_scan()
which would call _scan_invalid() which repeats the label scan on
devices flagged INVALID. lvmcache_label_rescan_vg() is now called
prominently by _vg_read() directly.
To do label scanning, lvm code calls lvmcache_label_scan().
Change lvmcache_label_scan() to use the new label_scan()
based on bcache.
Also add lvmcache_label_rescan_vg() which calls the new
label_scan_devs() which does label scanning on only the
specified devices. This is for a subsequent commit and
is not yet used.
No longer use the external 'result' pointer internally to set up the
cached label. The callback _set_label_read_result() is now given the
internal label pointer directly
Callers that don't need the result are no longer required to pass a
label pointer into label_read().
- Use 'lvmcache' consistently instead of 'metadata cache'
- Always use 5 characters for source line number
- Remember to convert uuids into printable form
- Use <no name> rather than (null) when VG has no name.
Switch from warn to log_error since this generated
failing return code for command so printing log_error()
is mandatory.
Happens with i.e. pvscan --cache meets crashing lvmetad.
Repairing missing devices does not work reliably
with lvmetad, so disable lvmetad before repair.
A standard lvmetad refresh (pvscan --cache) will
enable lvmetad again.
Sending %d as format argument in lvmetad_vg_remove_pending() will cause
segfaults in config_make_nodes_v() when va_arg() casts to int64_t. Also, it is
clearly advertised in the lvm source code that using plain %d is prohibited, so
let's switch to FMTd64.
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
When rescanning a VG from disk, the metadata read from
each PV was compared as a sanity check. The comparison
is done by exporting the vg metadata from each dev to
a config tree, and then comparing the config trees.
The function to create the config tree inserts
extraneous information along with the actual VG metadata.
This extra info includes creation_time. The config
trees for two devs can easily be created one second
apart in which case the different creation_times would
cause the metadata comparison to fail. The fix is to
exclude the extraneous info from the metadata comparison.
It's possible (mainly during boot) that udev has not finished
processing the device and hence the udev database record for that
device is still marked as uninitialized when we're trying to look
at it as part of multipath component check in pvscan --cache code.
So check several times with a short delay to wait for the udev db
record to be initialized before giving up completely.
When scanning devs to populate lvmetad during system startup,
filter-mpath with native sysfs multipath component detection
may not detect that a dev is multipath component. This is
because the multipath devices may not be set up yet.
Because of this, pvscan will scan multipath components during
startup, will see them as duplicate PVs, and will disable
lvmetad. This will leave lvmetad disabled on systems using
multipath, unless something or someone runs pvscan --cache
to rescan.
To avoid this problem, the code that is scanning devices to
populate lvmetad will now check the udev db to see if a
dev is a multipath component that should be skipped.
(This may not be perfect due to inherent udev races, but will
cover most cases and will be at least as good as it's ever
been.)
In some cases, the command will update VG metadata
in lvmetad without writing it. In these cases there
is no vg->vg_committed and it should use 'vg' directly.
This happens when the command finds that the lvmetad
VG has been invalidated, rereads the metadata from disk,
then updates lvmetad with that metadata. This happens
often with lvmlockd or foreign VGs, and can happen without
lvmlockd if a previous command fails after invalidating
the VG in lvmetad.
This fixes a regression from commit a7c45ddc5, which moved
the lvmetad VG update from vg_commit() to unlock_vg().
The lvmetad VG update needs to send the version of metadata
that was committed rather than sending the state of struct 'vg'.
The 'vg' may have been partially modified since vg_commit(),
and contain non-committed metadata that shouldn't be sent
to lvmetad.
Apply the same idea as vg_update.
Before doing the VG remove on disk, invalidate
the VG in lvmetad. After the VG is removed,
remove the VG in lvmetad. If the command fails
after removing the VG on disk, but before removing
the VG metadata from lvmetad, then a subsequent
command will see the INVALID flag and not use the
stale metadata from lvmetad.
Previously, a command sent lvmetad new VG metadata in vg_commit().
In vg_commit(), devices are suspended, so any memory allocation
done by the command while sending to lvmetad, or by lvmetad while
updating its cache could deadlock if memory reclaim was triggered.
Now lvmetad is updated in unlock_vg(), after devices are resumed.
The new method for updating VG metadata in lvmetad is in two phases:
1. In vg_write(), before devices are suspended, the command sends
lvmetad a short message ("set_vg_info") telling it what the new
VG seqno will be. lvmetad sees that the seqno is newer than
the seqno of its cached VG, so it sets the INVALID flag for the
cached VG. If sending the message to lvmetad fails, the command
fails before the metadata is committed and the change is not made.
If sending the message succeeds, vg_commit() is called.
2. In unlock_vg(), after devices are resumed, the command sends
lvmetad the standard vg_update message with the new metadata.
lvmetad sees that the seqno in the new metadata matches the
seqno it saved from set_vg_info, and knows it has the latest
copy, so it clears the INVALID flag for the cached VG.
If a command fails between 1 and 2 (after committing the VG on disk,
but before sending lvmetad the new metadata), the cached VG retains
the INVALID flag in lvmetad. A subsequent command will read the
cached VG from lvmetad, see the INVALID flag, ignore the cached
copy, read the VG from disk instead, update the lvmetad copy
with the latest copy from disk, (this clears the INVALID flag
in lvmetad), and use the correct VG metadata for the command.
(This INVALID mechanism already existed for use by lvmlockd.)
