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.
The parser for the output of the ctdb_fetch program
did not match the output that ctdb_fetch generates.
It seemed to rather come from the ctdb_bench test...
This patch adapts the parser to correctly interpret
the output of ctdb_fetch.
Michael
(This used to be ctdb commit 836b95f32724cf37e4f643f20653f78842613692)
The flags are user-visible, via tdb_get_flags/add_flags/remove_flags.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit 8f48e37c254e0852d4e2dea54b905ce5ef2b925d)
This version just wraps the reopen code, so we still re-grab the lock and do
the normal sanity checks.
The reason we do this at all is to avoid global fd limits, see:
http://forums.fedoraforum.org/showthread.php?t=210393
Note also that this whole reopen concept is fundamentally racy: if the parent
goes away before the child calls tdb_reopen_all, the database can be left
without an active lock and another TDB_CLEAR_IF_FIRST opener will clear it.
A fork_with_tdbs() wrapper could use a pipe to solve this, but it's hardly
elegant (what if there are other independent things which have similar needs?).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit 8d0d432ab7766d9c0f9868fd77e48b9b5cc5d9f9)
Make tdb transaction lock recursive (samba version)
This patch replaces 6ed27edbcd and
1a416ff13c, which fixed the bug where traversals
inside transactions would release the transaction lock early.
This solution is more general, and solves the more minor symptom that nested
traversals would also release the transaction lock early. (It was also suggestd in
Volker's comment in 6ed27ed).
This patch also applies to ctdb, if the traverse.c part is removed (ctdb's tdb
code never received the previous two fixes).
Tested using the testsuite from ccan (adapted to the samba code). Thanks to
Michael Adam for feedback.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Michael Adam <obnox@samba.org>
commit 760104188d
Author: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue Jul 21 16:23:35 2009 +0930
tdb: fix locking error
54a51839ea "Make tdb transaction lock
recursive (samba version)" was broken: I "cleaned it up" and prevented
it from ever unlocking.
To see the problem:
$ bin/tdbtorture -s 1248142523
tdb_brlock failed (fd=3) at offset 8 rw_type=1 lck_type=14 len=1
tdb_transaction_lock: failed to get transaction lock
tdb_transaction_start failed: Resource deadlock avoided
My testcase relied on the *count* being correct, which it was. Fixing that
now.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit ce19658ba13272238058e9b9bc03e62f48b737c0)
overallocate all records by 25%
This greatly reduces the fragmentation of databases where records
tend to grow slowly by a small amount each time. The case where this
is most seen is the ldb index records. Adding this overallocation
reduced the size of the resulting database by more than 20x when
running a test that adds 10k users.
(This used to be ctdb commit e72974e5cefabc7035399d16633f727f868caa61)
auto-repack in transactions that expand the tdb
The idea behind this is to recover from badly fragmented free
lists. Choosing the point where the file expands is fairly arbitrary,
but seems to work well.
(This used to be ctdb commit 233c52bfb087f636ad61e95c12616c02901f4f83)
imported the tdb_repack() code from CTDB
The tdb_repack() function repacks a TDB so that it has a single
freelist entry. The file doesn't shrink, but it does remove all
freelist fragmentation. This code originated in the CTDB vacuuming
code, but will now be used in ldb to cope with fragmentation from
re-indexing
(This used to be ctdb commit fe3ceb101a5a9c336973c2c6c31406bd8181c2fe)
make TDB_NOSYNC affect all the fsync/msync calls in transactions
During a transaction commit tdb normally uses fsync/msync calls to
make it crash safe. This can be disabled using the TDB_NOSYNC flag,
but it wasn't disabling all the code paths that caused a fsync/msync.
(This used to be ctdb commit e03980add02a28609a7a0a0c87ebc85419b98144)
commit b90863c0b7
Author: Howard Chu <hyc@highlandsun.com>
Date: Tue Mar 31 13:15:54 2009 +1100
Add tdb_transaction_prepare_commit()
Using tdb_transaction_prepare_commit() gives us 2-phase commits. This
allows us to safely commit across multiple tdb databases at once, with
reasonable transaction semantics
Signed-off-by: tridge@samba.org
(This used to be ctdb commit 4c3dac215a088947f645f727343997f5d47e3260)
Something, perhaps root_squash, causing permission denied on the test
file after we copy it over with scp. This sets the initial
permissions to be friendly and adds -p to the scp command to maintain
those friendly permissions.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 52f21f5a92eb14df7540a2ae9e212d936e646c06)
Add a "stopped" case to log events and stop the event script from
failing with an unknown event.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 7f67f7395e2233f0bba2e9662404aad49e13f645)
The parsing of "ctdb status -Y" output to determine various node
states was implemented very strictly. Therefore, the parsing broke
due to the addition of the new "stopped" state to the output of "ctdb
status -Y". This relaxes the parsing so that it should work for
versions prior to the introduction of the "stopped" state, as well as
future versions that add new states to the end of the list of bits in
output of "ctdb status -Y".
Similarly the check for cluster unhealthy (in _cluster_is_healthy())
now just checks for a single 1 in any bit in the "ctdb status -Y"
output, rather than checking for a particular number of 0s.
New tests
tests/simple/{41_ctdb_stop.sh,42_ctdb_continue.sh,43_stop_recmaster_yield.sh}
do rudimentary testing of the stop and continue functions.
Remove tests tests/simple/41_ctdb_ban.sh and
tests/simple/42_ctdb_unban.sh. They were both unreliable.
tests/simple/21_ctdb_disablemonitor.sh now schedules a restart, since
one will be required.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit 67c5bfb5f02c9d45a32d976021ede4fb2174dfe9)
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Signed-off-by: Wolfgang Mueller-Friedt <wolfmuel@de.ibm.com>
(This used to be ctdb commit 30cdad97706a9e9bb210120699aa939f6b16e8ca)
In ctdb_client.c:ctdb_transaction_commit(), after a failed
TRANS2_COMMIT control call (for instance due to the 1-second
being exceeded waiting for a busy node's reply), there is a
1-second gap between the transaction_cancel() and
replay_transaction() calls in which there is no lock on the
persistent db. And due to the lack of global state
indicating that a transaction is in progress in ctdbd, other nodes
may succeed to start transactions on the db in this gap and
even worse work on top of the possibly already pushed changes.
So the data diverges on the several nodes.
This change fixes this by introducing global state for a transaction
commit being active in the ctdb_db_context struct and in a db_id field
in the client so that a client keeps track of _which_ tdb it as
transaction commit running on. These data are set by ctdb upon
entering the trans2_commit control and they are cleared in the
trans2_error or trans2_finished controls. This makes it impossible
to start a nother transaction or migrate a record to a different
node while a transaction is active on a persistent tdb, including
the retry loop.
This approach is dead lock free and still allows recovery process
to be started in the retry-gap between cancel and replay.
Also note, that this solution does not require any change in the
client side.
This was debugged and developed together with
Stefan Metzmacher <metze@samba.org> - thanks!
Michael
(This used to be ctdb commit f88103516e5ad723062fb95fcb07a128f1069d69)
also check the returned status code in case the _stop() command failed
due to the eventscripts failing.
If this happens, make "ctdb stop" log an error to the console and try
the operation again.
(This used to be ctdb commit 20e82e0c48e07d1012549f5277f1f5a3f4bd10d1)
This does not modify any behaviour of the daemon itself other than showing this flag as ON in the ctdeb getcapabilities output
(This used to be ctdb commit fb337c151bd16ad5ad0c99431224451979d8c651)
