1
0
mirror of https://github.com/samba-team/samba.git synced 2024-12-23 17:34:34 +03:00
Commit Graph

143 Commits

Author SHA1 Message Date
Martin Schwenke
a5be2c245d ctdb-daemon: Store node addresses as ctdb_sock_addr rather than strings
Every time a nodemap is contructed the node IP addresses all need to
be parsed.  This isn't very productive use of CPU.

Instead, parse each string once when the nodes file is loaded.  This
results in much simpler code.

This code also removes the use of ctdb_address.  Duplicating the port
is pointless without an abstraction layer around ctdb_address.  If
CTDB gets an incompatible transport in the future then add an
abstraction layer.

Note that the infiniband code is not updated.  Compilation of the
infiniband code is already broken.  Fixing it will be a separate,
properly tested effort.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
2015-03-23 12:23:12 +01:00
Martin Schwenke
39d2fd330a ctdb-recoverd: Abort when daemon can take recovery lock during recovery
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>

Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Fri Feb 13 09:48:15 CET 2015 on sn-devel-104
2015-02-13 09:48:15 +01:00
Martin Schwenke
432d677489 ctdb-recoverd: Improve error messages on recovery lock coherence fail
When the daemon is able to take the recovery lock during recovery we
might as well guess that the cluster filesystem has a lock coherence
problem and print a more useful message.  This will be more helpful to
those trying out cluster filesystems that don't have lock coherence or
that are difficult to setup.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2015-02-13 07:19:07 +01:00
Martin Schwenke
1d6ed91f55 ctdb-recoverd: Simplify ctdb_recovery_lock()
Have it just silently take or fail to take the lock, except on an
unexpected failure (where it should log an error).

This means that when it is called we need to keep the old behaviour
and explicitly release the lock.  In do_recovery() the lock is
released and a message is printed before attempting to take the lock.
In the daemon sanity check the lock must be released in the error path
if it is actually taken.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2015-02-13 07:19:07 +01:00
Martin Schwenke
db32a2bce5 ctdb-recoverd: New function ctdb_recovery_unlock()
Unlock the recovery lock file.  This way knowledge of the file
descriptor isn't sprinkled throughout the code.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2015-02-13 07:19:07 +01:00
Martin Schwenke
72701be663 ctdb-recoverd: New function ctdb_recovery_have_lock()
True if this recovery daemon holds the lock.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2015-02-13 07:19:07 +01:00
Martin Schwenke
d110fe2318 ctdb-daemon: Mark tunable VerifyRecoveryLock as obsolete
It is pointless having a recovery lock but not sanity checking that it
is working.  Also, the logic that uses this tunable is confusing.  In
some places the recovery lock is released unnecessarily because the
tunable isn't set.

Simplify the logic by assuming that if a recovery lock is specified
then it should be verified.

Update documentation that references this tunable.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2015-02-13 07:19:07 +01:00
Michael Adam
a59fb322d6 ctdb: improve helpfulness of debug message when taking reclock fails
Print out the errno if the fcntl call.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Richard Sharpe <rsharpe@samba.org>

Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Fri Jan  9 04:25:02 CET 2015 on sn-devel-104
2015-01-09 04:25:02 +01:00
Martin Schwenke
acf26089f1 ctdb-util: Rename db_wrap to tdb_wrap and make it a build subsystem
This makes it consistent with Samba, to ease transition.

Update unit test code to link to with tdb_wrap instead of including
db_wrap.c.

There are some potential whitespace fixes in this commit that have
been ignored.  CTDB's lib/tdb_wrap will be deleted after the
transition to Samba's lib/tdb_wrap, so there's no point polishing it
too much.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2014-09-10 01:36:15 +02:00
Martin Schwenke
b0f9d33058 ctdb: Fix some "declarations after code" problems
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2014-09-10 01:36:14 +02:00
Martin Schwenke
c1558adeaa ctdb: Use sys_read() and sys_write() to ensure correct signal interaction
... and avoid compiler warnings in some cases.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2014-08-21 04:46:13 +02:00
Amitay Isaacs
f87b7f664f ctdb-vacuum: Use existing function ctdb_marshall_finish
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Volker Lendecke <vl@samba.org>

Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Wed Jul 23 09:44:00 CEST 2014 on sn-devel-104
2014-07-23 09:44:00 +02:00
Amitay Isaacs
2855173dac ctdb-daemon: Do not thaw databases if recovery is active
This prevents ctdb tool from thawing databases prematurely in
thaw/wipedb/restoredb commands if recovery is active.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
2014-07-07 13:29:50 +02:00
Amitay Isaacs
7aa20ccb5c ctdb-daemon: No need to call event scripts with CTDB_CALLED_BY_USER
This was added to support external monitoring using CTDB event scripts.
However, it was never used.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
2014-01-16 11:41:12 +11:00
Amitay Isaacs
6d1b74f052 ctdb-server: Coverity fixes
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>
2013-11-19 17:13:03 +01:00
Amitay Isaacs
ae30b61255 vacuuming: Fix vacuuming bug where requests keep bouncing between nodes (part 2)
This is caused by corruption of a record header such that the records
on two nodes point to each other as dmaster.  This makes a request for
that record bounce between nodes endlessly.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit f0853013655ac3bedf1b793de128fb679c6db6c6)
2013-08-14 16:55:51 +10:00
Amitay Isaacs
ee8d573069 vacuuming: Fix vacuuming bug where requests keep bouncing between nodes (part 1)
This is caused by corruption of a record header such that the records
on two nodes point to each other as dmaster.  This makes a request for
that record bounce between nodes endlessly.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit a610bc351f0754c84c78c27d02f9a695e60c5b0f)
2013-08-14 16:55:51 +10:00
Sumit Bose
d039f799ac Check return value of tdb_delete()
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 5cdcc3d45d358ddbcd7e864898eed9cbd9935429)
2013-07-11 15:16:55 +10:00
Amitay Isaacs
1c21f37e57 ctdbd: Set process names for child processes
This helps distinguish processes in process list in top, perf, etc.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 2493f57ce268d6fe7e4c40a87852c347fd60d29e)
2013-07-10 14:33:19 +10:00
Mathieu Parent
d82b9ae410 build: Fix tdb.h path to enable building with system TDB library
(This used to be ctdb commit f8bf99de3a5f56be67aaa67ed836458b1cf73e86)
2013-06-14 16:45:27 +10:00
Amitay Isaacs
140336383b ctdbd: Log node state transitions at higher debug level
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit db31dc48bd3135e9242af08bb79b67a17a2b1668)
2013-05-29 17:47:15 +10:00
Amitay Isaacs
a002c6ec12 vacuum: Reduce the priority of non-critical error
Since the complete database is not locked when the receive_records
control is received, it's possible that we may not be able to obtain
lock on a chain.  We will try again to store this record.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

(This used to be ctdb commit 32723c9efdad1c6ca4aa53f308ccd9bef1aadfff)
2013-05-24 14:22:16 +02:00
Martin Schwenke
6d9667f01c ctdbd: Add new runstate CTDB_RUNSTATE_FIRST_RECOVERY
This adds more serialisation to the startup, ensuring that the
"startup" event runs after everything to do with the first recovery
(including the "recovered" event).

Given that it now takes longer to get to the "startup" state, the
initscript needs to wait until ctdbd gets to "first_recovery".

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit ed6814ff0a59ddbb1c1b3128b505380f60d7aeb7)
2013-05-24 14:08:07 +10:00
Martin Schwenke
5aeae9744e ctdbd: Log a message when recovery master changes
Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-Programmed-With: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 1f96ea08f9a39dfe537c9b957ac512c84dc76f91)
2013-05-23 16:17:18 +10:00
Martin Schwenke
fa16cccf02 ctdbd: Remove the "stopped" event
It isn't used, superceded by "ipreallocated".

Signed-off-by: Martin Schwenke <martin@meltin.net>

(This used to be ctdb commit c2bb8596a8af6406ef50e53953884df9d6246a96)
2013-05-06 13:38:21 +10:00
Michael Adam
217d2ad7b8 recover: use CTDB_REC_RO_FLAGS where appropriate
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-By: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit b5a8791268e938d7e017056e0e2bd2cbec1fa690)
2013-04-24 18:49:08 +10:00
Michael Adam
527976d02a vacuum: introduce the RECEIVE_RECORDS control
This in preparation of turning the vacuming on the lmaster into
into a two phase process:

- First the node sends the list of records to be vacuumed
  to all other nodes with this new RECEIVE_RECORDS control.
  The remote nodes should store the lmaster's empty current copy.
- Only those records that could be stored on all other nodes
  are processed further. They are send to all other nodes with
  the TRY_DELETE_RECORDS control as before for deletion.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-By: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit e397702e271af38204fd99733bbeba7c1db3a999)
2013-04-24 18:47:32 +10:00
Michael Adam
b1a6289b44 ctdbd: unimplement the unused SET_DMASTER control
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 2e92deef5221ee651028ef87138b3113f1fece91)
2013-04-17 12:44:08 +02:00
Amitay Isaacs
30299c387f daemon: On shutdown, destroy timed events that check if recoverd is active
When CTDB is shutting down, recovery daemon is stopped, but the
event that checks if recovery daemon is still alive is not destroyed.
So recovery master is restarted during shutdown if CTDB daemon takes
longer to shutdown.

There are two processes that check if recovery daemon is working.

1. ctdb_check_recd() - which checks every 30 seconds if the recovery
   daemon process exists.

2. ctdb_recd_ping_timeout() - which is triggered when recovery daemon
   fails to ping CTDB daemon.

Both the events are periodic and need to be destroyed when shutting down.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 746168df2e691058e601016110fae818c6a265c3)
2013-01-09 13:20:26 +11:00
Michael Adam
f5b15e21c5 ctdb:recover: fix a comment typo
Signed-off-by: Michael Adam <obnox@samba.org>

(This used to be ctdb commit 5067392d2e06795559f25828b65c129608b65c0b)
2013-01-05 01:15:19 +01:00
Amitay Isaacs
08ffbc342c ctdb_recover: Replace static locking functions with locking API
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 4456a01d8f54ca6c771d7488048de5f638477d21)
2012-10-20 02:48:44 +11:00
Ronnie Sahlberg
e7d21834ae RECOVER: When we pull databases during recovery, we used to reallocate the databuffer for each entry added. This would normally not be an issue, but for cases where memory is fragmented, this could start to cost significant cpu if we need to reallocate and move to a different region.
Change this to instead preallocate , by default, 10MByte chunks to the data buffer.
This significantly reduces the number of potential reallocate and move  operations that may be required.

Create a tunable to override/change how much preallocation should be used.

(This used to be ctdb commit 1f262deaad0818f159f9c68330f7fec121679023)
2012-05-25 12:34:06 +10:00
Ronnie Sahlberg
26322d257d DEBUG: Add checks for and print debug messages when 1) a database contains very many records, 2) when a database is very big, 3) when a single record is very big.
Add tunables to control when to log these instances and allow it to be completely turned off by setting the threshold to 0

(This used to be ctdb commit 9ed58fef4991725f75509433496f4d5ffae0ae87)
2012-05-21 13:26:13 +10:00
Ronnie Sahlberg
a57eba2bb4 Track all child process so we never send a signal to an unrelated process (our child died and kernel wrapped the pid-space and reused the pid for a different process
Wrap all creation of child processes inside ctdb_fork() which is used to track all processes we have spawned.
Capture SIGCHLD to track also which child processes have terminated.

Wrap kill() inside ctdb_kill() and make sure that we never send a !0 signal to a child process pid that has already terminated (and might have been replaced with a

(This used to be ctdb commit f73a4b1495830bcdd094a93732a89dd53b3c2f78)
2012-05-03 14:03:26 +10:00
Amitay Isaacs
4392591555 Remove explicit include of lib/tevent/tevent.h.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 0681014ca5ed2a9b56f63fdace7f894beccf8a9a)
2012-04-13 17:28:14 +10:00
Amitay Isaacs
e2d83970e9 recovery: Add prototypes for tdb internal functions
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

(This used to be ctdb commit 98ac99c4a79fe2ee024890bb27c3ca68dc02d434)
2012-03-30 12:33:28 +11:00
Ronnie Sahlberg
fa3a06246a STICKY: add prototype code to make records stick to a node to "calm" down if they are found to be very hot and accessed by a lot of clients.
This can improve performance and stop clients from having to chase a rapidly migrating/bouncing record

(This used to be ctdb commit d0d98f7e45e5084b81335b004d50bddc80cdc219)
2012-03-20 17:12:19 +11:00
Ronnie Sahlberg
6f83805183 READONLY: skip vacuuming or deleting records with readonly delegations.
they are hot. wait until they have been revoked before we recall them.

(This used to be ctdb commit 7417d994c2a159f71d27d4bcd2f53684862eece3)
2012-02-29 16:09:24 +11:00
Ronnie Sahlberg
e3cdf429b6 ReadOnly: revokechild_active is a list, not a context.
Dont reset the pointer to NULL after deleting the first entry, loop deleting one entry
at a time until they are all gone or we will leak some memory and possibly a process.

(This used to be ctdb commit 8a86ac72088ad9f64ca83218c704f84c9abe00b6)
2011-09-13 18:47:18 +10:00
Ronnie Sahlberg
206a3c0c66 ReadOnly: add a new control to activate readonly lock capability for a database.
let all databases default to not support this  until enabled through this control

(This used to be ctdb commit 908a07c42e5135a3ba30a625fc4f4e4916de197a)
2011-09-01 11:08:18 +10:00
Ronnie Sahlberg
a0d4d240c3 ReadOnly: add a readonly flag to the getdbmap control and show the readonly setting in ctdb getdbmap output
(This used to be ctdb commit 4cac9ad7d9c9ca657a247a6c215476399c7d2210)
2011-09-01 10:28:15 +10:00
Ronnie Sahlberg
63dc96cdb2 ReadOnly: Change the ctdb_db structure to keep a uint8_t for flags instead of a boolean for
the persistent flag.
This is the same size as the original boolean but allows ut to add additional flags for the database

(This used to be ctdb commit 7462761638d25880ad46024ad4ef21667eb99a98)
2011-09-01 10:21:55 +10:00
Ronnie Sahlberg
9729d3e339 ReadOnly: Check the readonly flag instead of whether the tdb pointer is NULL or not
(This used to be ctdb commit 01314c2cb3a480917d6a632b83c39f0a48bba0e7)
2011-08-23 10:41:52 +10:00
Ronnie Sahlberg
59d8d9b695 ReadOnly: Once recovery has finished, make sure to free all revoke child processes and trigger the destructors for all deferred calls to re-queue the original packets to the input packet processing function
(This used to be ctdb commit 530a78aa05910beeca0867c4dbe226d4ce73f946)
2011-08-23 10:30:57 +10:00
Ronnie Sahlberg
b01dc029ca ReadOnly: After recovering all databases, make sure to clear out the tracking database used to track delegations and revoke. This is because the recovery will implicitely result in a revoke of all delegations.
(This used to be ctdb commit b5520933b9922d6af6f59f535824e1cdacb9f774)
2011-08-23 10:24:44 +10:00
Ronnie Sahlberg
6ff039d444 ReadOnly: After performing a recovery, clear out all flags related to readonly delegations and revoke
(This used to be ctdb commit 9985a97e11688f3f688bb84e1180fd57c42077f4)
2011-08-23 10:24:18 +10:00
Ronnie Sahlberg
a1abcd41e0 Restart recovery dameon if it looks like it hung.
Dont shutdown ctdbd completely, that only makes the problem worse.

(This used to be ctdb commit 221ecc2509f6d267d1854c1042ff945a620510bb)
2011-03-07 06:39:10 +11:00
Ronnie Sahlberg
8acb677c9c Deferred attach : at early startup, defer any db attach calls until we are out of recovery.
(This used to be ctdb commit eeaabd579841f60ab2c5b004cbbb1f5de2bfe685)
2011-03-01 12:13:34 +11:00
Michael Adam
40e922f4e6 recover: finish pending trans3 commits when a recovery is finished.
When the end_recovery control is received, pending trans3 commits are
finished. During the recovery, all the actions like persistent_callback
and persistent_store_timeout had been disabled to let the recovery do
its job. After the recover is completed, send the reply to the waiting
clients.

(This used to be ctdb commit f7dfeb7143f574c2434f7dd16917380dfd1f4f64)
2011-02-24 10:35:26 +01:00
Ronnie Sahlberg
3f819741ad ctdb: hold transaction locks during freeze, mark during recover.
Make the ctdb parent "mark" the transaction lock once the child process
has frozen/locked the entire database.
This stops the ctdb daemon from using  a blocking fcntl() locking on the tdb during the
read traverse during recovery.

CQ 1021388

(This used to be ctdb commit 52ee2b3ce822344d0f55ac040fe25f6ec5c0d7c2)
2011-01-18 14:07:44 +11:00