| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This flag to tdb_open/tdb_open_ex effects creation of a new database:
1) Uses the Jenkins lookup3 hash instead of the old gdbm hash if none is
specified,
2) Places a non-zero field in header->rwlocks, so older versions of TDB will
refuse to open it.
This means that the caller (ie Samba) can set this flag to safely
change the hash function. Versions of TDB from this one on will either
use the correct hash or refuse to open (if a different hash is specified).
Older TDB versions will see the nonzero rwlocks field and refuse to open
it under any conditions.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
If the caller to tdb_open_ex() doesn't specify a hash, and tdb_old_hash
doesn't match, try tdb_jenkins_hash.
This was Metze's idea: it makes life simpler, especially with the upcoming
TDB_INCOMPATIBLE_HASH flag.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
This is a better hash than the default: shipping it with tdb makes it easy
for callers to use it as the hash by passing it to tdb_open_ex().
This version taken from CCAN and modified, which took it from
http://www.burtleburtle.net/bob/c/lookup3.c.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Guenther
|
|
this might help reduce test times and load on test machines
|
|
This is Stefan Metzmacher <metze@samba.org>'s patch with minor changes:
1) Use the TDB_MAGIC constant so both hashes aren't of strings.
2) Check the hash in tdb_check (paranoia, really).
3) Additional check in the (unlikely!) case where both examples hash to 0.
4) Cosmetic changes to var names and complaint message.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
We must not endian-convert the magic string, just the rest.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Commit bc1c82ea137 "Fix tdb_check() to work with read-only tdb databases."
claimed to do this, but tdb_lockall_read() fails on read-only databases.
Also make sure we can still do tdb_check() inside a transaction (weird,
but we previously allowed it so don't break the API).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
We can end up with dead areas when we die during transaction commit;
tdb_check() fails on such a (valid) database.
This is particularly noticable now we no longer truncate on recovery;
if the recovery area was at the end of the file we used to remove it
that way.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
We saw tdb_lockall() take 71 seconds under heavy load; this is because Linux
(at least) doesn't prevent new small locks being obtained while we're waiting
for a big log.
The workaround is to do divide and conquer using non-blocking chainlocks: if
we get down to a single chain we block. Using a simple test program where
children did "hold lock for 100ms, sleep for 1 second" the time to do
tdb_lockall() dropped signifiantly. There are ln(hashsize) locks taken in
the contended case, but that's slow anyway.
More analysis is given in my blog at http://rusty.ozlabs.org/?p=120
This may also help transactions, though in that case it's the initial
read lock which uses this gradual locking routine; the update-to-write-lock
code is separate and still tries to update in one go.
Even though ABI doesn't change, minor version bumped so behavior change
can be easily detected.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
tdb_lockall() uses F_WRLCK internally, which doesn't work on a fd opened with O_RDONLY. Use tdb_lockall_read() instead.
Jeremy.
|
|
Guenther
|
|
Guenther
|
|
Commit 207a213c/24fed55d purported to fix the problem of signals during
tdb_new_database (which could cause a spurious short write, hence a failure).
However, the code is wrong: newdb+written is not correct.
Fix this by introducing a general tdb_write_all() and using it here and in
the tracing code.
Cc: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
We now use -fvisibilty=hidden to hide symbols from outside the tdb
shared library.
This also moved tdb_transaction_recover() into the tdb_private.h
header, as it should never have been a public API. For that reason we
are changing the version number. We're only doing a minor version
increment as it is extremely unlikely that anyone was actually using
tdb_transaction_recover() as its locking requirements were rather
unusual.
Pair-Programmed-With: Rusty Russell <rusty@samba.org>
|
|
|
|
|
|
|
|
|
|
tdb transactions were designed to be robust against the machine
powering off, but interestingly were never designed to handle the case
where an administrator kill -9's a process during commit. Because
recovery is only done on tdb_open, processes with the tdb already
mapped will simply use it despite it being corrupt and needing
recovery.
The solution to this is to check for recovery every time we grab a
data lock: we could have gained the lock because a process just died.
This has no measurable cost: here is the time for tdbtorture -s 0 -n 1
-l 10000:
Before:
2.75 2.50 2.81 3.19 2.91 2.53 2.72 2.50 2.78 2.77 = Avg 2.75
After:
2.81 2.57 3.42 2.49 3.02 2.49 2.84 2.48 2.80 2.43 = Avg 2.74
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
|
|
The current recovery code truncates the tdb file on recovery. This is
fine if recovery is only done on first open, but is a really bad idea
as we move to allowing recovery on "live" databases.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Now the transaction code uses the standard allrecord lock, that stops
us from trying to grab any per-record locks anyway. We don't need to
have special noop lock ops for transactions.
This is a nice simplification: if you see brlock, you know it's really
going to grab a lock.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
tdb_release_extra_locks() is too general: it carefully skips over the
transaction lock, even though the only caller then drops it. Change
this, and rename it to show it's clearly transaction-specific.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Now the transaction allrecord lock is the standard one, and thus is cleaned
in tdb_release_extra_locks(), _tdb_transaction_cancel() doesn't need to
know what type it is.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Centralize locking of all chains of the tdb; rename _tdb_lockall to
tdb_allrecord_lock and _tdb_unlockall to tdb_allrecord_unlock, and
tdb_brlock_upgrade to tdb_allrecord_upgrade.
Then we use this in the transaction code. Unfortunately, if the transaction
code records that it has grabbed the allrecord lock read-only, write locks
will fail, so we treat this upgradable lock as a write lock, and mark it
as upgradable using the otherwise-unused offset field.
One subtlety: now the transaction code is using the allrecord_lock, the
tdb_release_extra_locks() function drops it for us, so we no longer need
to do it manually in _tdb_transaction_cancel.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Records themselves get (read) locked by the traversal code against delete.
Interestingly, this locking isn't done when the allrecord lock has been
taken, though the allrecord lock until recently didn't cover the actual
records (it now goes to end of file).
The write record lock, grabbed by the delete code, is not suppressed
by the allrecord lock. This is now bad: it causes us to punch a hole
in the allrecord lock when we release the write record lock. Make this
consistent: *no* record locks of any kind when the allrecord lock is
taken.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
We were previously inconsistent with our "global" lock: the
transaction code grabbed it from FREELIST_TOP to end of file, and the
rest of the code grabbed it from FREELIST_TOP to end of the hash
chains. Change it to always grab to end of file for simplicity and
so we can merge the two.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
This was redundant before this patch series: it mirrored num_lockrecs
exactly. It still does.
Also, skip useless branch when locks == 1: unconditional assignment is
cheaper anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
This is pure overhead, but it centralizes the locking. Realloc (esp. as
most implementations are lazy) is fast compared to the fnctl anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Use our newly-generic nested lock tracking for the active lock.
Note that the tdb_have_extra_locks() and tdb_release_extra_locks()
functions have to skip over this lock now it is tracked.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
This never nests, so it's overkill, but it centralizes the locking into
lock.c and removes the ugly flag in the transaction code to track whether
we have the lock or not.
Note that we have a temporary hack so this places a real lock, despite
the fact that we are in a transaction.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Rather than a boutique lock and a separate nest count, use our
newly-generic nested lock tracking for the transaction lock.
Note that the tdb_have_extra_locks() and tdb_release_extra_locks()
functions have to skip over this lock now it is tracked.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Factor out two loops which find locks; we are going to introduce a couple
more so a helper makes sense.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Move locking intelligence back into lock.c, rather than open-coding the
lock release in transaction.c.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
In many places we check whether locks are held: add a helper to do this.
The _tdb_lockall() case has already checked for the allrecord lock, so
the extra work done by tdb_have_extra_locks() is merely redundant.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
tdb_transaction_lock() and tdb_transaction_unlock() do nothing if we
hold the allrecord lock. However, the two locks don't overlap, so
this is wrong.
This simplification makes the transaction lock a straight-forward nested
lock.
There are two callers for these functions:
1) The transaction code, which already makes sure the allrecord_lock
isn't held.
2) The traverse code, which wants to stop transactions whether it has the
allrecord lock or not. There have been deadlocks here before, however
this should not bring them back (I hope!)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Because fcntl locks don't nest, we track them in the tdb->lockrecs array
and only place/release them when the count goes to 1/0. We only do this
for record locks, so we simply place the list number (or -1 for the free
list) in the structure.
To generalize this:
1) Put the offset rather than list number in struct tdb_lock_type.
2) Rename _tdb_lock() to tdb_nest_lock, make it non-static and move the
allrecord check out to the callers (except the mark case which doesn't
care).
3) Rename _tdb_unlock() to tdb_nest_unlock(), make it non-static and
move the allrecord out to the callers (except mark again).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
The word global is overloaded in tdb. The global_lock inside struct
tdb_context is used to indicate we hold a lock across all the chains.
Rename it to allrecord_lock.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
The word global is overloaded in tdb. The GLOBAL_LOCK offset is used at
open time to serialize initialization (and by the transaction code to block
open).
Rename it to OPEN_LOCK.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Now tdb_open() calls tdb_transaction_cancel() instead of
_tdb_transaction_cancel, we can make it static.
Signed-off-by: Rusty Russell<rusty@rustcorp.com.au>
|
|
This is taken from the CCAN code base: rather than using tdb_brlock for
locking and unlocking, we split it into brlock and brunlock functions.
For extra debugging information, brunlock says what kind of lock it is
unlocking (even though fnctl locks don't need this). This requires an
extra argument to tdb_transaction_unlock() so we know whether the
lock was upgraded to a write lock or not.
We also use a "flags" argument tdb_brlock:
1) TDB_LOCK_NOWAIT replaces lck_type = F_SETLK (vs F_SETLKW).
2) TDB_LOCK_MARK_ONLY replaces setting TDB_MARK_LOCK bit in ltype.
3) TDB_LOCK_PROBE replaces the "probe" argument.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Signed-off-by: Matthias Dieter Wallnöfer <mwallnoefer@yahoo.de>
|
|
This might help on some filesystems
|
|
|
|
If a process (or the machine) dies after just after writing the
recovery head (pointing at the end of file), the recovery record will filled
with 0x42. This will not invoke a recovery on open, since rec.magic
!= TDB_RECOVERY_MAGIC.
Unfortunately, the first transaction commit will happily reuse that
area: tdb_recovery_allocate() doesn't check the magic. The recovery
record has length 0x42424242, and it writes that back into the
now-valid-looking transaction header) for the next comer (which
happens to be tdb_wipe_all in my tests).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
|
|
There was a bug in tdb where the
tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0, 1);
(ending the transaction-"mutex") was done before the
/* remove the recovery marker */
This means that when a transaction is committed there is a window where another
opener of the file sees the transaction marker while the transaction committer
is still fully functional and working on it. This led to transaction being
rolled back by that second opener of the file while transaction_commit() gave
no error to the caller.
This patch moves the F_UNLCK to after the recovery marker was removed, closing
this window.
|
|
We need to keep TDB_ALLOW_NESTING as default behavior,
so that existing code continues to work.
However we may change the default together with a major version
number change in future.
metze
|
|
Make the default be that transaction is not allowed and any attempt to create a nested transaction will fail with TDB_ERR_NESTING.
If an application can cope with transaction nesting and the implicit
semantics of tdb_transaction_commit(), it can enable transaction nesting
by using the TDB_ALLOW_NESTING flag.
(cherry picked from ctdb commit 3e49e41c21eb8c53084aa8cc7fd3557bdd8eb7b6)
Signed-off-by: Stefan Metzmacher <metze@samba.org>
|
