The conversion to rhashtable introduced a hash table lock per lockspace,
in place of per bucket locks. To make this more scalable, switch to
using a rwlock for hash table access. The common case fast path uses
it as a read lock.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Currently the scand kthread acts like a garbage collection for expired
rsbs on toss list, to clean them up after a certain timeout. It triggers
every couple of seconds and iterates over the toss list while holding
ls_rsbtbl_lock for the whole hash bucket iteration.
To reduce the amount of time holding ls_rsbtbl_lock, we now handle the
disposal of expired rsbs using a per-lockspace timer that expires for the
earliest tossed rsb on the lockspace toss queue. This toss queue is
ordered according to the rsb res_toss_time with the earliest tossed rsb
as the first entry. The toss timer will only trylock() necessary locks,
since it is low priority garbage collection, and will rearm the timer
if trylock() fails. If the timer function does not find any expired
rsb's, it rearms the timer with the next earliest expired rsb.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
To prepare for using rhashtable, add two rsb lists for iterating
through rsb's in two uncommon cases where this is necesssary:
- when dumping rsb state from debugfs, now using seq_list.
- when looking at all rsb's during recovery.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
There are several places where lock processing can perform two hash table
lookups, first in the "keep" list, and if not found, in the "toss" list.
This patch introduces a new rsb state flag "RSB_TOSS" to represent the
difference between the state of being on keep vs toss list, so that the
two lists can be combined. This avoids cases of two lookups.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Prepare to replace our own hash table with rhashtable by replacing
the per-bucket locks in our own hash table with a single lock.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Use spin_lock_bh for all spinlocks involved in message processing,
in preparation for softirq message processing. DLM lock requests
from user space involve dlm processing in user context, in addition
to the standard kernel context, necessitating bh variants.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Convert ls_recv_active rw_semaphore to an rwlock to avoid
sleeping, in preparation for softirq message processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Add a new struct to save the current position in the rsb masters_list
while sending the rsb names to other nodes. The rsb names are sent in
multiple chunks, and for each new chunk, the new "dlm_dir_dump" struct
saves the last position in the masters_list. The new struct is also
used to save more information to sanity check the recovery process.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Move the rsb root_list from the lockspace to a stack variable since
it is now only used by the ls_recover() function.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Add a new "masters_list" for master rsb structs, with a new
rwlock. The new list is created and used during the recovery
process to send the master rsb names to new nodes. With this
change, the current "root_list" can be used without locking.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Move dlm_create_root_list() and dlm_release_root_list() to
recover.c and declare them static because they are only used
there.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch removes a read of the ls->ls_recover_seq uint64_t number in
_create_rcom(). If the ls->ls_recover_seq is readed the ls_recover_lock
need to held. However this number was always readed before when any rcom
message is received and it's not necessary to read it again from a per
lockspace variable to use it for the replying message. This patch will
pass the sequence number as parameter so another read of ls->ls_recover_seq
and holding the ls->ls_recover_lock is not required.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch removes code parts which was declared deprecated by
commit 6b0afc0cc3 ("fs: dlm: don't use deprecated timeout features by
default"). This contains the following dlm functionality:
- start a cancel of a dlm request did not complete after certain timeout:
The current way how dlm cancellation works and interfering with other
dlm requests triggered by the user can end in an overlapping and
returning in -EBUSY. The most user don't handle this case and are
unaware that DLM can return such errno in such situation. Due the
timeout the user are mostly unaware when this happens.
- start a netlink warning messages for user space if dlm requests did
not complete after certain timeout:
This feature was never being built in the only known dlm user space side.
As we are to remove the timeout cancellation feature we can directly
remove this feature as well.
There might be the possibility to bring the timeout cancellation feature
back. However the current way of handling the -EBUSY case which is only
a software limitation and not a hardware limitation should be changed.
We minimize the current code base in DLM cancellation feature to not have
to deal with those existing features while solving the DLM cancellation
feature in general.
UAPI define DLM_LSFL_TIMEWARN is commented as deprecated and reserved
value. We should avoid at first to give it a new meaning but let
possible users still compile by keeping this define. In far future we
can give this flag a new meaning. The same for the DLM_LKF_TIMEOUT lock
request flag.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch cleans up the handling of recovery results by moving it from
ls_recover() to the caller do_ls_recovery(). This makes the error handling
clearer.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Make dlm_new_lockspace() wait until a full recovery completes
sucessfully or fails. Previously, dlm_new_lockspace() returned
to the caller after dlm_recover_members() finished, which is
only partially through recovery. The result of the previous
behavior is that the new lockspace would not be usable for some
time (especially with overlapping recoveries), and some errors
in the later part of recovery could not be returned to the caller.
Kernel callers gfs2 and cluster-md have their own wait handling to
wait for recovery to complete after calling dlm_new_lockspace().
This continues to work, but will be unnecessary.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Make clear that a particular recovery iteration must not be aborted
before membership changes are applied to the members list (ls_nodes)
and midcomms layer. Interrupting recovery before this can result
in missing node-specific changes in midcomms or through lsops.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch will change to evaluate the dlm_recovery_stopped() in the
condition of the if branch instead fetch it before evaluating the
condition. As this is an atomic test-set operation it should be
evaluated in the condition itself.
Reported-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch will return -EINTR instead of 1 if recovery is stopped. In
case of ping_members() the return value will be checked if the error is
-EINTR for signaling another recovery was triggered and the whole
recovery process will come to a clean end to process the next one.
Returning 1 will abort the recovery process and can leave the recovery
in a broken state.
It was reported with the following kernel log message attached and a gfs2
mount stopped working:
"dlm: bobvirt1: dlm_recover_members error 1"
whereas 1 was returned because of a conversion of "dlm_recovery_stopped()"
to an errno was missing which this patch will introduce. While on it all
other possible missing errno conversions at other places were added as
they are done as in other places.
It might be worth to check the error case at this recovery level,
because some of the functionality also returns -ENOBUFS and check why
recovery ends in a broken state. However this will fix the issue if
another recovery was triggered at some points of recovery handling.
Reported-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Based on 1 normalized pattern(s):
this copyrighted material is made available to anyone wishing to use
modify copy or redistribute it subject to the terms and conditions
of the gnu general public license v 2
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 45 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Richard Fontana <rfontana@redhat.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Alexios Zavras <alexios.zavras@intel.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190528170027.342746075@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Call schedule() here could make the thread miss wake
up from kthread_stop(), so it is better to recheck
kthread_should_stop() before call schedule(), a symptom
happened when I run indefinite test (which mostly created
clustered raid1, assemble it in other nodes, then stop
them) of clustered raid.
$ ps aux|grep md|grep D
root 4211 0.0 0.0 19760 2220 ? Ds 02:58 0:00 mdadm -Ssq
$ cat /proc/4211/stack
kthread_stop+0x4d/0x150
dlm_recoverd_stop+0x15/0x20 [dlm]
dlm_release_lockspace+0x2ab/0x460 [dlm]
leave+0xbf/0x150 [md_cluster]
md_cluster_stop+0x18/0x30 [md_mod]
bitmap_free+0x12e/0x140 [md_mod]
bitmap_destroy+0x7f/0x90 [md_mod]
__md_stop+0x21/0xa0 [md_mod]
do_md_stop+0x15f/0x5c0 [md_mod]
md_ioctl+0xa65/0x18a0 [md_mod]
blkdev_ioctl+0x49e/0x8d0
block_ioctl+0x41/0x50
do_vfs_ioctl+0x96/0x5b0
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x1e/0xad
This maybe not resolve the issue completely since the
KTHREAD_SHOULD_STOP flag could be set between "break"
and "schedule", but at least the chance for the symptom
happen could be reduce a lot (The indefinite test runs
more than 20 hours without problem and it happens easily
without the change).
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: David Teigland <teigland@redhat.com>
When dlm_recoverd_stop() is called between kthread_should_stop() and
set_task_state(TASK_INTERRUPTIBLE), dlm_recoverd will not wake up.
Signed-off-by: Tadashi Miyauchi <miyauchi@toshiba-tops.co.jp>
Signed-off-by: Tsutomu Owa <tsutomu.owa@toshiba.co.jp>
Signed-off-by: David Teigland <teigland@redhat.com>
The log messages relating to the progress of recovery
are minimal and very often useful. Change these to
the KERN_INFO level so they are always available.
Signed-off-by: David Teigland <teigland@redhat.com>
The in_recovery rw_semaphore has always been acquired and
released by different threads by design. To work around
the "BUG: bad unlock balance detected!" messages, adjust
things so the dlm_recoverd thread always does both down_write
and up_write.
Signed-off-by: David Teigland <teigland@redhat.com>
Remove the dir hash table (dirtbl), and use
the rsb hash table (rsbtbl) as the resource
directory. It has always been an unnecessary
duplication of information.
This improves efficiency by using a single rsbtbl
lookup in many cases where both rsbtbl and dirtbl
lookups were needed previously.
This eliminates the need to handle cases of rsbtbl
and dirtbl being out of sync.
In many cases there will be memory savings because
the dir hash table no longer exists.
Signed-off-by: David Teigland <teigland@redhat.com>
The "nodir" mode (statically assign master nodes instead
of using the resource directory) has always been highly
experimental, and never seriously used. This commit
fixes a number of problems, making nodir much more usable.
- Major change to recovery: recover all locks and restart
all in-progress operations after recovery. In some
cases it's not possible to know which in-progess locks
to recover, so recover all. (Most require recovery
in nodir mode anyway since rehashing changes most
master nodes.)
- Change the way nodir mode is enabled, from a command
line mount arg passed through gfs2, into a sysfs
file managed by dlm_controld, consistent with the
other config settings.
- Allow recovering MSTCPY locks on an rsb that has not
yet been turned into a master copy.
- Ignore RCOM_LOCK and RCOM_LOCK_REPLY recovery messages
from a previous, aborted recovery cycle. Base this
on the local recovery status not being in the state
where any nodes should be sending LOCK messages for the
current recovery cycle.
- Hold rsb lock around dlm_purge_mstcpy_locks() because it
may run concurrently with dlm_recover_master_copy().
- Maintain highbast on process-copy lkb's (in addition to
the master as is usual), because the lkb can switch
back and forth between being a master and being a
process copy as the master node changes in recovery.
- When recovering MSTCPY locks, flag rsb's that have
non-empty convert or waiting queues for granting
at the end of recovery. (Rename flag from LOCKS_PURGED
to RECOVER_GRANT and similar for the recovery function,
because it's not only resources with purged locks
that need grant a grant attempt.)
- Replace a couple of unnecessary assertion panics with
error messages.
Signed-off-by: David Teigland <teigland@redhat.com>
Change some existing error/debug messages to
collect more useful information, and add
some new error/debug messages to address
recently found problems.
Signed-off-by: David Teigland <teigland@redhat.com>
These new callbacks notify the dlm user about lock recovery.
GFS2, and possibly others, need to be aware of when the dlm
will be doing lock recovery for a failed lockspace member.
In the past, this coordination has been done between dlm and
file system daemons in userspace, which then direct their
kernel counterparts. These callbacks allow the same
coordination directly, and more simply.
Signed-off-by: David Teigland <teigland@redhat.com>
Put all the calls to recovery barriers in the same function
to clarify where they each happen. Should not change any behavior.
Also modify some recovery debug lines to make them consistent.
Signed-off-by: David Teigland <teigland@redhat.com>
Instead of creating our own kthread (dlm_astd) to deliver
callbacks for all lockspaces, use a per-lockspace workqueue
to deliver the callbacks. This eliminates complications and
slowdowns from many lockspaces sharing the same thread.
Signed-off-by: David Teigland <teigland@redhat.com>
If a node is removed from a lockspace, and then added back before the
dlm is notified of the removal, the dlm will not detect the removal
and won't clear the old state from the node. This is fixed by using a
list of added nodes so the membership recovery can detect when a newly
added node is already in the member list.
Signed-off-by: David Teigland <teigland@redhat.com>
To prevent the master of an rsb from changing rapidly, an unused rsb is kept
on the "toss list" for a period of time to be reused. The toss list was
being cleared completely for each recovery, which is unnecessary. Much of
the benefit of the toss list can be maintained if nodes keep rsb's in their
toss list that they are the master of. These rsb's need to be included
when the resource directory is rebuilt during recovery.
Signed-off-by: David Teigland <teigland@redhat.com>
Introduce a per-lockspace rwsem that's held in read mode by dlm_recv
threads while working in the dlm. This allows dlm_recv activity to be
suspended when the lockspace transitions to, from and between recovery
cycles.
The specific bug prompting this change is one where an in-progress
recovery cycle is aborted by a new recovery cycle. While dlm_recv was
processing a recovery message, the recovery cycle was aborted and
dlm_recoverd began cleaning up. dlm_recv decremented recover_locks_count
on an rsb after dlm_recoverd had reset it to zero. This is fixed by
suspending dlm_recv (taking write lock on the rwsem) before aborting the
current recovery.
The transitions to/from normal and recovery modes are simplified by using
this new ability to block dlm_recv. The switch from normal to recovery
mode means dlm_recv goes from processing locking messages, to saving them
for later, and vice versa. Races are avoided by blocking dlm_recv when
setting the flag that switches between modes.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
New features: lock timeouts and time warnings. If the DLM_LKF_TIMEOUT
flag is set, then the request/conversion will be canceled after waiting
the specified number of centiseconds (specified per lock). This feature
is only available for locks requested through libdlm (can be enabled for
kernel dlm users if there's a use for it.)
If the new DLM_LSFL_TIMEWARN flag is set when creating the lockspace, then
a warning message will be sent to userspace (using genetlink) after a
request/conversion has been waiting for a given number of centiseconds
(configurable per node). The time warnings will be used in the future
to do deadlock detection in userspace.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Some common, non-error messages should use log_debug instead of log_error
so they can be turned off.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This fixes the following gcc warnings generated on
the architectures where uint64_t != unsigned long long (e.g. ppc64).
fs/dlm/rcom.c:154: warning: format '%llx' expects type 'long long unsigned int', but argument 4 has type 'uint64_t'
fs/dlm/rcom.c:154: warning: format '%llx' expects type 'long long unsigned int', but argument 5 has type 'uint64_t'
fs/dlm/recoverd.c:48: warning: format '%llx' expects type 'long long unsigned int', but argument 3 has type 'uint64_t'
fs/dlm/recoverd.c:202: warning: format '%llx' expects type 'long long unsigned int', but argument 3 has type 'uint64_t'
fs/dlm/recoverd.c:210: warning: format '%llx' expects type 'long long unsigned int', but argument 3 has type 'uint64_t'
Signed-off-by: Ryusuke Konishi <ryusuke@osrg.net>
Signed-off-by: Patrick Caulfield <pcaulfie@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Requests that arrive after recovery has started are saved in the
requestqueue and processed after recovery is done. Some of these requests
are purged during recovery if they are from nodes that have been removed.
We move the purging of the requests (dlm_purge_requestqueue) to later in
the recovery sequence which allows the routine saving requests
(dlm_add_requestqueue) to avoid filtering out requests by nodeid since the
same will be done by the purge. The current code has add_requestqueue
filtering by nodeid but doesn't hold any locks when accessing the list of
current nodes. This also means that we need to call the purge routine
when the lockspace is being shut down since the add routine will not be
rejecting requests itself any more.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Red Hat BZ 211914
The previous patch "[DLM] fix aborted recovery during
node removal" was incomplete as discovered with further testing. It set
the bit for the RS_LOCKS barrier but did not then wait for the barrier.
This is often ok, but sometimes it will cause yet another recovery hang.
If it's a new node that also has the lowest nodeid that skips the barrier
wait, then it misses the important step of collecting and reporting the
barrier status from the other nodes (which is the job of the low nodeid in
the barrier wait routine).
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Red Hat BZ 211914
When many nodes are joining a lockspace simultaneously, the dlm gets a
quick sequence of stop/start events, a pair for adding each node.
dlm_controld in user space sends dlm_recoverd in the kernel each stop and
start event. dlm_controld will sometimes send the stop before
dlm_recoverd has had a chance to take up the previously queued start. The
stop aborts the processing of the previous start by setting the
RECOVERY_STOP flag. dlm_recoverd is erroneously clearing this flag and
ignoring the stop/abort if it happens to take up the start after the stop
meant to abort it. The fix is to check the sequence number that's
incremented for each stop/start before clearing the flag.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Red Hat BZ 211914
With the new cluster infrastructure, dlm recovery for a node removal can
be aborted and restarted for a node addition. When this happens, the
restarted recovery isn't aware that it's doing recovery for the earlier
removal as well as the addition. So, it then skips the recovery steps
only required when nodes are removed. This can result in locks not being
purged for failed/removed nodes. The fix is to check for removed nodes
for which recovery has not been completed at the start of a new recovery
sequence.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
When a new lockspace was being created, the recoverd thread was being
started for it before the lockspace was added to the global list of
lockspaces. The new thread was looking up the lockspace in the global
list and sometimes not finding it due to the race with the original thread
adding it to the list. We need to add the lockspace to the global list
before starting the thread instead of after, and if the new thread can't
find the lockspace for some reason, it should return an error.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
When we abort one recovery to do another, break out of the ping_members()
routine more quickly, and wake up the dlm_recoverd thread more quickly
instead of waiting for it to time out.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This is the core of the distributed lock manager which is required
to use GFS2 as a cluster filesystem. It is also used by CLVM and
can be used as a standalone lock manager independantly of either
of these two projects.
It implements VAX-style locking modes.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steve Whitehouse <swhiteho@redhat.com>
