Mathieu Parent [Sat, 26 Mar 2011 10:55:30 +0000 (11:55 +0100)]
Manpage for ping_pong
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
af75d3e37412e03d3978073edbe6dee78f265c3c)
Martin Schwenke [Mon, 23 May 2011 05:33:12 +0000 (15:33 +1000)]
onnode: fix natgwlist nodespec
This hasn't worked for a while if ever.
We treat this case specially because the output has 2 works on the 1st
line. We also handle the error case where /etc/ctdb_natgw_nodes
exists but none of the other $NATGW_* configuration is done.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
66e89797c7866d207a5bbf1836f52d70dba7cea6)
Martin Schwenke [Mon, 23 May 2011 05:24:52 +0000 (15:24 +1000)]
onnode: fix get_nodes_with_status()
Setting IFS and looping though items with colons in them doesn't work.
Change this to read through the output line by line. The header line
needs to be thrown away by throwing away everything up to the 1st
newline.
Keep stderr from the "ctdb status" command, otherwise debugging is
impossible.
On error, append any output from ctdb to onnode's error message.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
d60592cf99999f10344a05ef0571fb300bb9d97c)
Martin Schwenke [Tue, 17 May 2011 04:26:55 +0000 (14:26 +1000)]
onnode: Remove an unnecessary comment.
The comment about $CTDB_NODES_SOCKETS is meaningless. The code ti
refers to works just find with $CTDB_NODES_SOCKETS.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
74e69a564bac653dadfffe8b08145b9b3be16e61)
Martin Schwenke [Tue, 17 May 2011 04:24:30 +0000 (14:24 +1000)]
onnode: Future-proof get_nodes_with_status().
The current code requires knowledge of the number of status bits
output by "ctdb status -Y".
This changes the code to be completely general.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
e1788f25fde3d1f26bf4831a331741aa280f6fbc)
Martin Schwenke [Tue, 17 May 2011 03:25:08 +0000 (13:25 +1000)]
onnode: Exit with error for unknown command-line flags.
Use of "local" was masking errors in command-line processing.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
ca80adda7517b43147ef30156ae34c66b29fa2bd)
Martin Schwenke [Tue, 17 May 2011 03:20:51 +0000 (13:20 +1000)]
onnode: Be defensive when listing IPs of nodes with designated status.
The current version gives the last item left after stripping the known
fields. If an insufficent number of status fields is stripped then
this would return a residual status field value, which turned out to
be a valid IP address for localhost... so no error occurs.
This change means that the node number is stripped and any residual
status field value will stay appended, causing an error the first time
this command is tested.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
74715e6ec7b67c6f0e863aa51c87279758d6bf91)
Martin Schwenke [Tue, 17 May 2011 03:18:11 +0000 (13:18 +1000)]
onnode - Fix long standing bug in onnode healthy/ok/connected/con.
When the output of "ctdb status -Y" changed to add an extra status
column we didn't fix onnode.
This adds a match for the extra column.
Signed-off-by: Martin Schwenke <martin@meltin.net>
(This used to be ctdb commit
793febaebd3d484ddfbbcb47aaa0cdf3cfc1a00d)
Mathieu Parent [Wed, 23 Mar 2011 21:16:18 +0000 (22:16 +0100)]
Fix bashism
... again ;-)
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
2266586c1839af032622be54dc7f71e39d2bd9ef)
Ronnie Sahlberg [Thu, 12 May 2011 08:58:07 +0000 (18:58 +1000)]
Merge branch 'master' of ssh://git.samba.org/data/git/ctdb
(This used to be ctdb commit
307e915459c26a728a1ec16bd735d983d493df53)
Ronnie Sahlberg [Thu, 12 May 2011 00:24:46 +0000 (10:24 +1000)]
When using multiple VLANs, some funky stuff can sometimes happen when
adding/removing IP addresses causing routes might be dropped by the system.
The easiest workaround for this is to unconditionally try to reapply
all static routes for all interfaces once ipreallocation has finished,
not just adding them back on the affected interface.
This worksaround a funky issue in
CQ S1023538
(This used to be ctdb commit
84600d1f53632d5fe76c308727f31f61b5ec1010)
Michael Adam [Wed, 11 May 2011 14:13:47 +0000 (16:13 +0200)]
doc: regenerate ctdb docs
(This used to be ctdb commit
2d67186e5acd5aa8cb3eb1f4fbd4a41153c52e96)
Luk Claes [Wed, 11 May 2011 15:53:59 +0000 (17:53 +0200)]
doc/ctdb.1.xml: update listvars documentation
Signed-off-by: Luk Claes <luk@debian.org>
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
afd96d5990815019b1f9ddc8b78a05f86eca0421)
Luk Claes [Wed, 11 May 2011 14:13:47 +0000 (16:13 +0200)]
doc: regenerate ctdb docs
Signed-off-by: Luk Claes <luk@debian.org>
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
39f595cad5321c64e2b1e72fe7b4bbb720f4b906)
Luk Claes [Wed, 11 May 2011 14:11:40 +0000 (16:11 +0200)]
doc/ctdb.1.xml: Fix typo
s/poerwoff/poweroff/
Bug 8124
Signed-off-by: Luk Claes <luk@debian.org>
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
a6d2f1bd552dba33640acb7a0b8110534debd4ce)
Ronnie Sahlberg [Wed, 11 May 2011 09:50:09 +0000 (19:50 +1000)]
Remove all checking of GPFS from ctdb_diagnostics
CQ S1023524
(This used to be ctdb commit
4cddba08b46db0a56a86b32403a41b89cd097317)
Ronnie Sahlberg [Mon, 9 May 2011 22:25:27 +0000 (08:25 +1000)]
If samba fails to start for some reason, make this cause the startup event to fail too, so that ctdbd will re-try the startup event later.
Or else this will leave samba not running.
CQ S1023394
(This used to be ctdb commit
f90485b08d32cbe56050718a3b28ca0fe1d64e0f)
Ronnie Sahlberg [Mon, 9 May 2011 20:19:34 +0000 (06:19 +1000)]
Dont exit from checking interfaces once we have found one interface that is not
in use by public addresses. this can happen when we have removed existing interfaces/ip addresses and prevents us from verifying the status of other interfaces
(This used to be ctdb commit
d67955b42f7627be9dae995230c8fcbb8a948ec2)
Ronnie Sahlberg [Sun, 8 May 2011 20:35:33 +0000 (06:35 +1000)]
Remove logging of spam/errors from the 10.interfrace
script if/when we have for example NATGW configured but no public addresses defined on that interface
CQ S1023378
(This used to be ctdb commit
8837daa424732aeb5a20814b1709c345a97a0e09)
Michael Adam [Wed, 4 May 2011 12:28:26 +0000 (14:28 +0200)]
packaging: add ltdbtool and its manpage to the RPM
(This used to be ctdb commit
ce6409dc7d059701f0fe4b57e7c05c38c66629c5)
Michael Adam [Wed, 4 May 2011 12:25:48 +0000 (14:25 +0200)]
install the ltdbtool manpage with "make install"
(This used to be ctdb commit
ffbff1affed8301831387e23b4f8f824d9f78e20)
Michael Adam [Wed, 4 May 2011 11:44:59 +0000 (13:44 +0200)]
install ltdbtool with "make install"
(This used to be ctdb commit
991ea66e5ed0eb7ab256dc8e3118dc78462d4752)
Michael Adam [Wed, 4 May 2011 11:44:10 +0000 (13:44 +0200)]
build "ltdbtool" in "make all"
(This used to be ctdb commit
d91e80c698a7706460e9ee74bd4f5a9ab0a7b9b1)
Gregor Beck [Wed, 4 May 2011 12:17:04 +0000 (14:17 +0200)]
ltdbtool: add manpage html + roff
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
992baa4215bfc1b29fd153ccb7c42bb0cb66fa4f)
Gregor Beck [Wed, 4 May 2011 12:14:54 +0000 (14:14 +0200)]
ltdbtool: add manpage
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
2ed3603274cd38dde4ae98eef653e9a9de631eb5)
Gregor Beck [Thu, 14 Apr 2011 10:51:59 +0000 (12:51 +0200)]
add ltdbtool - a standalone ltdb tool
This this is a tool to handle (dump and convert) ctdb's local tdb
copies (ltdbs) without connecting to a ctdb daemon.
It can be used to
* dump the contents of a ltdb, printing
the ctdb record header information
* dump a non-clustered tdb database (like tdbdump)
* convert between an ltdb and a non-clustered tdb
(adding or removing ctdb headers)
* convert between 64 and 32 bit ltdbs
(the ctdb record headers differ by 4 bytes of padding)
usage: bin/ltdbtool dump [-p] [-s{0|32|64}] <idb>
bin/ltdbtool convert [-s{0|32|64}] [-o{0|32|64}] <idb> <odb>
Pair-Programmed-With: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
efcf2815711cd5371633614fb91273bd0a786da0)
Gregor Beck [Thu, 14 Apr 2011 10:55:57 +0000 (12:55 +0200)]
ctdb catdb: fix escaping of '"' and '\'
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
2b5cb0841fd813cd54be170c305a828885e0f038)
Ronnie Sahlberg [Wed, 4 May 2011 01:34:17 +0000 (11:34 +1000)]
Dont call the UPDATE event if both old and new interface is the same.
CQ S1018175
(This used to be ctdb commit
6a74515f0a1e24d97cee3ba05d89133aac7ad2b7)
Ronnie Sahlberg [Tue, 3 May 2011 22:54:02 +0000 (08:54 +1000)]
Cleanup of logging messages/spamming
Reduce an infomational message about not performing ip reallocation
from NOTICE(the default) to INFO.
These messages are normal during startup or when stopped/banned when
we will be in recovery mode for a while.
Remove a messager in the loop waiting for initial startup to complete about
the generation being invalid. It is always invalid at this stage before we have
finished initial recovery.
Rate-limit the informational messages for CTDB_WAIT_UNTIL_RECOVERED
so that we only print them once per second for the first 60 seconds and after that only once per 10 minutes.
These messages are normal during startup, but we should not be logging them every second for cases where we will remain in recovery mode during startup for an extended period of time.
Such as if suspended or permabanned.
CQ S1023302
(This used to be ctdb commit
3a0af8780dc595acbed880f288fcbc4f62c862fb)
Ronnie Sahlberg [Tue, 12 Apr 2011 21:51:36 +0000 (07:51 +1000)]
bonding mode 4 monitoring:
we can not just check if MII Status is up for bonding mode 4, since the kernel will always report the bond device as UP
even if all cables are disconneccted.
For mode 4, ignore the status of the bond device and instead chek if at least one slave interface is up
when determining if the device is good or bad
(This used to be ctdb commit
a6930cec6d9503dba18b9d4839d87a1c1a8ddba2)
Ronnie Sahlberg [Mon, 11 Apr 2011 19:24:43 +0000 (05:24 +1000)]
If the eventscript is finished but state->ctdb is NULL,
log an error and return.
(Need to find root cause for this is soo too.)
(This used to be ctdb commit
2e80d53b73fcba58ed5a72bab66c051691ccf719)
Ronnie Sahlberg [Sun, 10 Apr 2011 19:56:14 +0000 (05:56 +1000)]
IFACE handling. Assume links are always good on nstartup (they almost always
Simplify the handling of setting the links in the 10.interface eventscript
and remove the optimization to only call setifacelink on state change
to make the code simpler to read.
If a take ip event fails, flag the node as unhealthy.
Add a check to the interface script to check if the interface exists
or if it has been deleted.
So that we can capture and become UNHELTHY if someone deletes an interface
we are using to host public addresses.
(This used to be ctdb commit
4ab63d2a7262aff30d5eced184c294c9c9dd4974)
David Disseldorp [Tue, 29 Mar 2011 09:08:39 +0000 (11:08 +0200)]
web: use the new git repository url on the download page
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
b36818888fac7ebbed26fcdd2dd1d426e3d2f8f0)
Ronnie Sahlberg [Wed, 6 Apr 2011 00:26:27 +0000 (10:26 +1000)]
NATGW: dont set arp_ignore in 11.natgw anymore since we no longer
need this for the natgw functionality
(This used to be ctdb commit
bf3bf2967e3781c918e33b3a210e68e0ccca0c51)
Michael Adam [Tue, 5 Apr 2011 11:58:09 +0000 (13:58 +0200)]
packaging: remove the dependency to tdbtool and tdbdump from the spec file
The init script does now check for the availability of tdbdump
and "tdbtool check" and issues warnings if they are not available.
This can remove a dependency loop with building samba RPMs.
(This used to be ctdb commit
c7652c4038e012b7ef9bc1da352dd2c02d60dc29)
Michael Adam [Tue, 5 Apr 2011 11:50:00 +0000 (13:50 +0200)]
ctdb.init: print a warning when tdbdump is found but tdbtoo or "tdbtool check" is not available
(This used to be ctdb commit
afb26e38b617b85cdac14a7cd6dd3c85b8fddbc4)
Michael Adam [Tue, 5 Apr 2011 11:43:56 +0000 (13:43 +0200)]
ctdb.init: check for availability of "tdbtool check" and "tdbdump"
Print a warning if neither is available.
(This used to be ctdb commit
4137d2a7d31cdce22847cebfc0239cfe2d8e937c)
Mathieu Parent [Tue, 22 Mar 2011 23:16:27 +0000 (00:16 +0100)]
Correction of spelling errors
* continous -> continuous
* activete -> activate
(thanks to lintian)
See https://bugzilla.samba.org/show_bug.cgi?id=6935
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
fb6987c2f747d6dbf9bb3899a480124d1c242a90)
Ronnie Sahlberg [Mon, 21 Mar 2011 03:25:53 +0000 (14:25 +1100)]
This needs more testing first
Revert "ctdbd: call tdb_reopen_all() in freeze child."
This reverts commit
3d9828861c771a060923f3181fa8224e0122bffc.
(This used to be ctdb commit
55c3446c9ba82d24b1d7db92bc3611fd8027b7fb)
Rusty Russell [Mon, 21 Mar 2011 02:37:17 +0000 (13:07 +1030)]
ctdbd: call tdb_reopen_all() in freeze child.
In theory, the ctdbd parent shouldn't be holding any locks, but it's a good
idea to always call tdb_reopen_all() after a fork().
(This used to be ctdb commit
3d9828861c771a060923f3181fa8224e0122bffc)
Rusty Russell [Mon, 21 Mar 2011 02:33:01 +0000 (13:03 +1030)]
ctdbd: fix lock held on error ("ctdb_req_dmaster from non-")
We should release the lock on the record before returning; otherwise the
recovery (which tries to freeze the database) will fail. Symptoms are as
follows:
ctdbd: pnn 15 dmaster request for new-dmaster 19 from non-master 1 real-dmaster=5 key
f049c3c8 dbid 0x6cf2837d gen=
1148812532 curgen=
1148812532 c->rsn=2 header.rsn=15 reqid=
2147483585 keyval=0x4f464e49
ctdbd: ctdb_req_dmaster from non-master. Force a recovery.
...
ctdbd: freeze_lock-1:server/ctdb_freeze.c:55 Failed to lock database registry.tdb
CQ:
1022545
(This used to be ctdb commit
38b2dbe0605816742e74e2b8a811eaba99c7e12d)
Ronnie Sahlberg [Wed, 16 Mar 2011 03:55:58 +0000 (14:55 +1100)]
Deferred attach: create the timed event as a child context of the da context we want to delete.
Othwervise the da context can be timed out and talloc_free()d
but the event for this already freed object will still trigger,
causing a talloc error and shutdown.
CQ S1022515
(This used to be ctdb commit
2fd27bdedb1e0d6558c07e1b74fc8e70ddf593dc)
Ronnie Sahlberg [Sun, 13 Mar 2011 22:55:28 +0000 (09:55 +1100)]
IP reallocation. If a public address is already hosted on the node when we startup, log a warning message but do not cause the recovery to fail.
CQ S1022356
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
89f8169c24da96c1fdd0ac19b8a1e0e1df01a72a)
Ronnie Sahlberg [Sun, 13 Mar 2011 00:30:52 +0000 (11:30 +1100)]
Vacuuming: initialize a variable to avoid a harmless valgrind hit
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
ad709e99bcad7a4884f2336663d161ba61307ae5)
Ronnie Sahlberg [Fri, 11 Mar 2011 22:42:07 +0000 (09:42 +1100)]
Dont allow clients to connect to databases untile we are well past and through
the initial recovery phase
CQ S1022412
Signed-off-by: Michael Adam <obnox@samba.org>
(This used to be ctdb commit
e02bbd915b7151c615ff64f09ad9abc9720bef7d)
Michael Adam [Fri, 11 Mar 2011 15:05:44 +0000 (16:05 +0100)]
vacuum: fix a comment typo
(This used to be ctdb commit
a16dc65b4602da5ce2c16578bec2e7882aff240d)
Michael Adam [Fri, 11 Mar 2011 14:57:45 +0000 (15:57 +0100)]
vacuum: use insert_record_into_delete_queue in ctdb_local_schedule_for_deletion.
This is to take advantage of the hash collision handling and logging
also in ctdb_local_schedule_for_deletion.
(This used to be ctdb commit
52193b6692091e341ed7a81dbd9a61ae49a8aac5)
Michael Adam [Fri, 11 Mar 2011 14:55:52 +0000 (15:55 +0100)]
vacuum: refactor insert_record_into_delete_queue out of ctdb_control_schedule_for_deletion
(This used to be ctdb commit
be4b63ee18933524f780df5c313447e5ef0786d1)
Michael Adam [Fri, 11 Mar 2011 13:57:15 +0000 (14:57 +0100)]
vacuum: raise a debug level from INFO to DEBUG
when overwriting an existing entry in the delete_queue.
(This used to be ctdb commit
f28e636cc4a04ef982672d5f569ad6b6b963db1f)
Michael Adam [Thu, 3 Feb 2011 15:32:23 +0000 (16:32 +0100)]
ctdb_ltdb_store_server: honour the AUTOMATIC record flag
Do not delete empty records that carry this flag but store
them and schedule them for deletetion. Do not store the flag
in the ltdb though, since this is internal only and should not
be visible to the client.
(This used to be ctdb commit
f898ff21fa338358179e79381215b13a6bc77c53)
Michael Adam [Thu, 3 Feb 2011 15:30:52 +0000 (16:30 +0100)]
ltdb: add the CTDB_REC_FLAG_AUTOMATIC to the initial header in ctdb_ltdb_fetch()
Signals that this record was not created by a client level store.
(This used to be ctdb commit
69d34983a37b0324ff7610b8dfdcd8d13bf81c54)
Michael Adam [Thu, 3 Feb 2011 15:27:42 +0000 (16:27 +0100)]
ctdb_private.h: add record flag CTDB_REC_FLAG_AUTOMATIC
This is a flag that shall signa that a record has been automatically generated by ctdb
and not by an explicit client store operation. This will be used in the ctdb_ltdb_fetch
operation which stores an empty record with default initial header before trying to
migrate the record from the dmaster when the record does not exist in the local tdb.
(This used to be ctdb commit
46381a3cb58ccc11422af8f7798c80ea8d72294f)
Michael Adam [Tue, 28 Dec 2010 12:19:22 +0000 (13:19 +0100)]
ctdb_ltdb_store_server: add ability to send SCHEDULE_FOR_DELETION control to ctdb_ltdb_store.
(This used to be ctdb commit
ab2711701999a5ecc23a36b3d9ba8e94f92e4c87)
Michael Adam [Tue, 21 Dec 2010 17:08:11 +0000 (18:08 +0100)]
ctdb_ltdb_store_server: Improve debug message in ctdb_ltdb_store when store or delete fails.
(This used to be ctdb commit
2559b2a45eb11834da3b0e0963e24351c8b7477f)
Michael Adam [Tue, 21 Dec 2010 16:50:52 +0000 (17:50 +0100)]
ctdb_ltdb_store_server: always store the data when ctdb_ltdb_store() is called from the client
This also fixes a segfault since ctdb_lmaster uses the vnn_map.
(This used to be ctdb commit
e58c8f51f27e468897af5210b80e5f5f45c3c4bb)
Michael Adam [Fri, 10 Dec 2010 13:13:50 +0000 (14:13 +0100)]
ctdb_ltdb_store_server: implement fastpath vacuuming deletion based on VACUUM_MIGRATED flag.
When the record has been obtained by the lmaster as part of the vacuuming-fetch
handler and it is empty and never been migrated with data, then such records
are deleted instead of being stored. These records have automatically been
deleted when leaving the former dmaster, so that they vanish for good when
hitting the lmaster in this way. This will reduces the load on traditional
vacuuming.
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit
c9b65f3602f51bcbf0e6d82c12076c31e4aebe38)
Michael Adam [Fri, 3 Dec 2010 14:29:21 +0000 (15:29 +0100)]
ctdb_ltdb_store_server: delete an empty record that is safe to delete instead of storing locally.
When storing a record that is being migrated off to another node
and has never been migrated with data, then we can safely delete it
from the local tdb instead of storing the record with empty data.
Note: This record is not deleted if we are its lmaster or dmaster.
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit
3cca0d4b48325d86de2cb0b44bb7811a30701352)
Michael Adam [Thu, 30 Dec 2010 17:19:32 +0000 (18:19 +0100)]
server: Use the ctdb_ltdb_store_server() in the ctdb daemon for non-persistent dbs
This is realized by adding a ctdb_ltdb_store_fn function pointer to the db
context and filling it in the attach procedure for non-persistent dbs.
(This used to be ctdb commit
df49ec44de80affa5ccc637dec12a20a26e8706e)
Michael Adam [Thu, 30 Dec 2010 16:44:51 +0000 (17:44 +0100)]
server: create a server variant ctdb_ltdb_store_server() of ctdb_ltdb_store().
This is supposed to contain logic for deleting records that are safe
to delete and scheduling records for deletion. It will be called in
server context for non-persistent databases instead of the standard
ctdb_ltdb_store() function.
(This used to be ctdb commit
23631ffc152486aed9ce5b69a391e52bc4947833)
Michael Adam [Tue, 28 Dec 2010 12:14:23 +0000 (13:14 +0100)]
daemon: fill ctdb->ctdbd_pid early
(This used to be ctdb commit
3da1e2e30bf34622f08e6ecd5b8fe55684e5007a)
Michael Adam [Tue, 21 Dec 2010 14:29:46 +0000 (15:29 +0100)]
test: send SCHEDULE_FOR_DELETION control from randrec test.
(This used to be ctdb commit
30aa55b3efc6fbd4078f93da386b6aeb337c1a0c)
Michael Adam [Tue, 21 Dec 2010 14:29:23 +0000 (15:29 +0100)]
client: add accessor function ctdb_header_from_record_handle().
(This used to be ctdb commit
cf57efd440ccc3db381386f4749bfcbf8ac5ecae)
Michael Adam [Tue, 28 Dec 2010 12:13:34 +0000 (13:13 +0100)]
vacuum: add ctdb_local_schedule_for_deletion()
(This used to be ctdb commit
b70bc141d84f7355d2c6c901961b7366db566980)
Michael Adam [Tue, 21 Dec 2010 13:25:48 +0000 (14:25 +0100)]
server: implement a new control SCHEDULE_FOR_DELETION to fill the delete_queue.
(This used to be ctdb commit
680223074e992b32ccf6f42cb80c3fa93074fee7)
Michael Adam [Tue, 8 Mar 2011 23:57:55 +0000 (00:57 +0100)]
control: add a new control opcode CTDB_CONTROL_SCHEDULE_FOR_DELETION
(This used to be ctdb commit
4cebfa33db3c7effa087f753530c52b2dd8550e6)
Michael Adam [Tue, 8 Mar 2011 23:56:25 +0000 (00:56 +0100)]
control: add macro CHECK_CONTROL_MIN_DATA_SIZE.
This is for the control dispatcher to check whether the input data has
a required minimum size.
(This used to be ctdb commit
2038e745db33cc5c3b4e2db8a00a57ede03906a2)
Michael Adam [Thu, 23 Dec 2010 10:54:09 +0000 (11:54 +0100)]
vacuum: lower level of hash collision debug message to INFO
(This used to be ctdb commit
b9bdef46fedfbc543263b67cfee3e896773cd8e8)
Michael Adam [Wed, 22 Dec 2010 23:27:27 +0000 (00:27 +0100)]
vacuum: add statistics output to the fast and full traverse runs.
(This used to be ctdb commit
3addd28aa73883b3b05888e309d19db0eb67eab9)
Michael Adam [Tue, 21 Dec 2010 13:19:00 +0000 (14:19 +0100)]
vacuum: refactor insert_delete_record_data_into_tree() out of add_record_to_delete_tree()
for reuse in filling the delete_queue.
(This used to be ctdb commit
7bbb12695c24da25671f1c39a411295d35870d2c)
Michael Adam [Mon, 20 Dec 2010 20:43:41 +0000 (21:43 +0100)]
vacuum: change all Vacuum*Interval tunables to default to 10
So, by default we have a fastpath vacuuming every 10 seconds and
full blown db-traverse vacuuming once every 10 minutes.
(This used to be ctdb commit
4f0ace982dbb5b4f9c035dbf4cb0ae74cd18d81b)
Michael Adam [Mon, 20 Dec 2010 20:30:39 +0000 (21:30 +0100)]
vacuum: disable full db-traverse vacuuming runs when VacuumFastPathCount == 0
(This used to be ctdb commit
571683e7c48aeed8ce41c584d016ced7ff0d2e2d)
Michael Adam [Mon, 20 Dec 2010 17:03:38 +0000 (18:03 +0100)]
vacuum: Only run full vacuumig (db traverse) every VacuumFastPathCount times.
(This used to be ctdb commit
23b8c8c5fc8604ee0bd6da1f4b5152277eb5f1c0)
Michael Adam [Mon, 20 Dec 2010 16:54:04 +0000 (17:54 +0100)]
vacuum: reset the fast path count in the event handle if it exceeds the limit.
(This used to be ctdb commit
91e6d36a190b1c9e4c8b18f7833e51c5c9a67574)
Michael Adam [Mon, 20 Dec 2010 16:49:29 +0000 (17:49 +0100)]
vacuum: bump the number of fast-path runs in the vacuum child destructor
(This used to be ctdb commit
c0668bfe0bb4e69988ae34d875568d08539e6fb9)
Michael Adam [Mon, 20 Dec 2010 16:44:02 +0000 (17:44 +0100)]
vacuum: add a fast_path_count to the vacuum_handle.
(This used to be ctdb commit
53a39d0cc5ea251c2189ec8178ccb769fa046c43)
Michael Adam [Mon, 20 Dec 2010 16:42:25 +0000 (17:42 +0100)]
Add a tunable VacuumFastPathCount.
This will control how many fast-path vacuuming runs wil have to
be done, before a full vacuuming will be triggered, i.e. one with
a db-traversal.
(This used to be ctdb commit
0d997ec7e61a7bee2cb05456f9c7d5e6f7a44797)
Michael Adam [Mon, 20 Dec 2010 16:25:35 +0000 (17:25 +0100)]
vacuum: traverse the delete_queue befor traversing the database.
(This used to be ctdb commit
04c335f9195a5fd83c91a57d06b1e4eaa511844e)
Michael Adam [Mon, 20 Dec 2010 16:24:32 +0000 (17:24 +0100)]
vacuum: add delete_queue_traverse() for traversal of the delete_queue.
(This used to be ctdb commit
5eee05c4d256c08f4ee60a1a69efda6844e39729)
Michael Adam [Tue, 21 Dec 2010 10:22:50 +0000 (11:22 +0100)]
vacuum: reduce indentation in add_record_to_delete_tree()
This simplyfies the logical structure a bit by using early return.
(This used to be ctdb commit
4d32908fdcec120426536a761e1d0be60f076198)
Michael Adam [Mon, 20 Dec 2010 16:11:27 +0000 (17:11 +0100)]
vacuum: refactor new add_record_to_delete_tree() out of vacuum_traverse().
This will be reused by the traversal of the delete_queue list.
(This used to be ctdb commit
4407e5a7fb045ce56b6d902f7116de663ea648cb)
Michael Adam [Mon, 20 Dec 2010 15:41:13 +0000 (16:41 +0100)]
vacuum: skip adding records to list of records to send to lmaster on lmaster
This list is skipped afterwards when the lists are processed.
(This used to be ctdb commit
e99834c1a2eea60f7f974c0689ae0a65cfe178ff)
Michael Adam [Mon, 20 Dec 2010 15:31:27 +0000 (16:31 +0100)]
vacuum: refactor new add_record_to_vacuum_fetch_list() out of vacuum_traverse().
This is the function that fills the list of records to send to each lmaster
with the VACUUM_FETCH message.
This function will be reused in the traverse function for the delete_queue.
(This used to be ctdb commit
d4ab790c1f679e833eb97816762fcfcee15ccb10)
Michael Adam [Mon, 20 Dec 2010 09:55:53 +0000 (10:55 +0100)]
server: rename ctdb_repack_db() to ctdb_vacuum_and_repack_db()
(This used to be ctdb commit
6c603f85726d2efac9710af7c4875ded2ca7230e)
Michael Adam [Fri, 17 Dec 2010 01:22:02 +0000 (02:22 +0100)]
When wiping a database, clear the delete_queue.
(This used to be ctdb commit
731a6011ce4a1301f86eacb039955745f2b5d866)
Michael Adam [Fri, 17 Dec 2010 00:53:25 +0000 (01:53 +0100)]
vaccum: clear the fast-path vacuuming delete_queue after creating the vacuuming child.
Maybe we should keep a copy for the case that the vacuuming fails?
(This used to be ctdb commit
f19fe5b45748a6998c6950a5b1db7ec2c4468c1c)
Michael Adam [Fri, 17 Dec 2010 00:38:09 +0000 (01:38 +0100)]
When attaching to a non-persistent DB, initialize the delete_queue.
(This used to be ctdb commit
0aff1b61dd1b683c6739478008a5b014b933df50)
Michael Adam [Wed, 22 Dec 2010 13:50:53 +0000 (14:50 +0100)]
Add a delete_queue to the ctdb database context struct.
This list will be filled by the client using a new
delete control. The list will then be used to implement
a fast-path vacuuming that will traverse this list instead
of traversing the database.
(This used to be ctdb commit
9bbedf786b26bb074f668b31f29a9032af958673)
Michael Adam [Fri, 10 Dec 2010 13:11:38 +0000 (14:11 +0100)]
call: becoming dmaster in VACUUM_MIGRATION, set the VACUUM_MIGRATED record flag
This temporary flag is used for the local record storage function to
decide whether to delete an empty record which has never been migrated
with data as part of the fast-path vacuuming process or, or to store
the record.
(This used to be ctdb commit
c11ca778ee90444c44dee0a629cd2eefa3a1f75e)
Michael Adam [Fri, 10 Dec 2010 13:07:21 +0000 (14:07 +0100)]
call: hand the submitted record_flags to local record storage function.
(This used to be ctdb commit
4079b8bf7a57a27a45d29784a1b0a414c778e552)
Michael Adam [Fri, 10 Dec 2010 13:02:33 +0000 (14:02 +0100)]
call: transfer the record flags in the ctdb call packets.
This way, the MIGRATED_WITH_DATA information can be transported
along with the records. This is important for vacuuming to function
properly.
The record flags are appended to the data section of the ctdb_req_dmaster
and ctdb_reply_dmaster structs.
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit
945187d64cfc7bd30a0c3b0d548cbe582d95dde3)
Michael Adam [Fri, 10 Dec 2010 12:59:37 +0000 (13:59 +0100)]
server: in the VACUUM_FETCH handler, add the VACUUM_MIGRAION to the call flags
This way, the records coming in via this handler, can be treated appropriately.
Namely, they can be deleted instead of being stored when the meet the fast-path
vacuuming criteria (empty, never migrated with data...)
(This used to be ctdb commit
fb5d832104970320359b3e474eb291ca3d629380)
Michael Adam [Fri, 10 Dec 2010 12:57:01 +0000 (13:57 +0100)]
add a new record flag CTDB_REC_FLAG_VACUUM_MIGRATED.
This is to be used internally. The purpose is to flag a record
as been migrated by a VACUUM_MIGRATION, which is triggered by
a VACUUM_FETCH message as part of the vacuuming. The local store
routine will base its decision whether to delete or to store
the record (among other things) upon the value of this flag.
This flag should never be stored in the local database copies.
(This used to be ctdb commit
dd2449c422f323f9b5485e45107a9cc5acc09e08)
Michael Adam [Fri, 10 Dec 2010 13:22:55 +0000 (14:22 +0100)]
call: Move definition of call flags down to the definition of the flags field.
(This used to be ctdb commit
86c844fb08a7fd33e94f56b8d5e43278120e1162)
Michael Adam [Fri, 10 Dec 2010 13:24:40 +0000 (14:24 +0100)]
call: add new call flag CTDB_CALL_FLAG_VACUUM_MIGRATION
This is to be used when the CTDB_SRVID_VACUUM_FETCH message
triggers the migration of deleted records to the lmaster.
The lmaster can then delete records that have not been
migrated with data instead of storing them.
(This used to be ctdb commit
455cc6616e10b7f09589f9b87cb60f591bb502b0)
Michael Adam [Fri, 3 Dec 2010 14:24:06 +0000 (15:24 +0100)]
recoverd: in a recovery, set the MIGRATED_WITH_DATA flag on all records
Those records that are kept after recovery, are non-empty, and
stored identically on all nodes. So this is as if they had been
migrated with data.
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
(This used to be ctdb commit
101be642e492a3a54231e2e3e6553a59380fe702)
Michael Adam [Fri, 3 Dec 2010 14:21:51 +0000 (15:21 +0100)]
server: when we migrate off a record with data, set the MIGRATED_WITH_DATA flag
(This used to be ctdb commit
f5fb232117886186066ab3430fdd2307cba94960)
Michael Adam [Thu, 3 Feb 2011 11:15:41 +0000 (12:15 +0100)]
vacuum: check lmaster against num_nodes instead of vnn_map->size
When lmaster is bigger than the biggest recorded node number,
then exit the traverse with error.
(This used to be ctdb commit
3930c7796b72bbf275bbca8aaeceec3e705a964b)
Michael Adam [Thu, 3 Feb 2011 16:47:36 +0000 (17:47 +0100)]
vacuum: reduce indentation of the loop sending VACUUM_FETCH controls
This slightly improves the code structure in that loop.
(This used to be ctdb commit
bc4990e600c53433a924a0d70e3488a5a6bdc1ff)
Michael Adam [Thu, 3 Feb 2011 11:26:45 +0000 (12:26 +0100)]
vacuum: correctly send TRY_DELETE_RECORDS ctrl to all active nodes
Originally, the control was sent to all records in the vnn_map, but
there was something still missing here:
When a node can not become lmaster (via CTDB_CAPABILITY_LMASTER=no)
then it will not be part of the vnn_map. So such a node would
be active but never receive the TRY_DELETE_RECORDS control from a
vacuuming run.
This is fixed in this change by correctly building the list of
active nodes first in the same way that the recovery process does it.
(This used to be ctdb commit
49247df4a47a8a107fa7dd7b187e69e243e6bdbe)
Michael Adam [Thu, 3 Feb 2011 11:18:58 +0000 (12:18 +0100)]
vacuum: in ctdb_vacuum_db, fix the length of the array of vacuum fetch lists
This patch fixes segfaults in the vacuum child when at least one
node has been stopped or removed from the cluster:
The size of the vnn_map is only the number of active nodes
(that can be lmaster). But the node numbers that are referenced
by the vnn_map spread over all configured nodes.
Since the array of vacuum fetch lists is referenced by the
key's lmaster's node number later on, the array needs to
be of size num_nodes instad of vnn_map->size.
(This used to be ctdb commit
136508e3f4dd0acc210dde938ad59ef38b63d3a1)