[SCM] CTDB repository - annotated tag ctdb-1.0.109 created - ctdb-1.0.109

Ronnie Sahlberg sahlberg at samba.org
Wed Dec 16 21:54:41 MST 2009


The annotated tag, ctdb-1.0.109 has been created
        at  0ef5d97eea0593775e29a64448cc596da1060743 (tag)
   tagging  99894a70fe2ebfe43daae7e88ff0fc9cab33e0fb (commit)
  replaces  ctdb-1.0.108
 tagged by  Ronnie Sahlberg
        on  Thu Dec 17 15:55:32 2009 +1100

- Log -----------------------------------------------------------------
tag for 1.0.109
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQBLKbnW2aJ36aon/y8RAh7HAJ93sAsji3UX1XJG2IAoQ2hMN5yzjwCdHyRI
cDufV0kMcguHHqe1NemwZLk=
=qcSW
-----END PGP SIGNATURE-----

Andrew Tridgell (7):
      tdb: allow reads after prepare commit
      added some more speed tests to tdbtool
      tdb: fixed the intermittent failure of tdbtorture in the build farm
      fixed tdbbackup to give tdb error messages (cherry picked from samba commit 08be1420ba52ef9bba90d0f811c7810841ee8568)
      make tdbbackup use transactions
      added basic testing of tdb_transaction_prepare_commit() in tdbtorture (cherry picked from samba commit 84547b8dba3c0cf4e20b3c50d9386081d475df6b)
      tdb: detect tdb store of identical records and skip

Christian Ambach (2):
      improve time jump logging
      reduce vacuuming lognoise

Günther Deschner (1):
      tdb: fix c++ build warning.

Holger Hetterich (1):
      Added a simple tdb integrity check to tdbtool. The command "check" runs traverse on the currently open tdb, and returns the number of entries if the integrity check is successful. (cherry picked from samba commit 42366bcbbdd42bb9d5821dfcc9dbe71a1eafa330)

Jelmer Vernooij (3):
      tdb: Add simple reimplementation of tdbdump in Python as an example of the tdb Python bindings. (This used to be commit 47d797f7885b1e7bcff724496ecb1990e8440eea) (cherry picked from samba commit 6bdd1425b75c8931965f0e5627f5a63dc6820a7c)
      Implement missing functions in pytdb. (cherry picked from samba commit 2da551bbcc6cab296769c193b0b82aaa6256cece)
      Make sure to not close tdb database more than once. (cherry picked from samba commit 6fe6983e4c960abc69d5fa80cbef534ae515209a)

Jeremy Allison (2):
      Add define guards around otherwise unused variable. Jeremy. (cherry picked from samba commit 4fc9f9c3f943cdeb27e37f0ee068cdd0da7cb00c)
      Remove unecessary msync. Jeremy. (cherry picked from samba commit 0bae1ef3de8fda5e1e2d641b14a408e627396912) (This used to be commit db2acaf46fdc38078b6b28b68909e289f6c9e0ec) (cherry picked from samba commit a1cf3ad5d655cf5a847df6d6299b3af1a30ec1e3)

Kirill Smelkov (5):
      tdb: kill last bits from swig
      tdb: fix typo in python's Tdb.get() docstring
      tdb: reset tdb->fd to -1 in tdb_close()
      tdb: add tests for double .close() in pytdb
      tdb: update README a bit

Matthias Dieter Wallnöfer (1):
      tdb tools: Mostly cosmetic adaptions

Michael Adam (27):
      call: lower the debug message "refusing migration while transction" to lvl INFO
      client: lower debug level of transaction-active-retry message to DEBUG
      client: lower level of commit retry message WARNING->DEBUG
      test: get value for --timelimit from environment var CTDB_TEST_TIMELIMIT in transaction test
      test: add test 54_ctdb_transaction_recovery.sh
      tests: remove persistent_safe write test.
      tests: remove the persistent_unsafe writes test.
      tests: remove the no_trans mode from ctdb_transaction.
      server: add a new control CTDB_CONTROL_TRANS3_COMMIT
      define CTDB_DB_SEQNUM_KEY - used with the new implementation of transactions.
      Add a new control CTDB_GET_DB_SEQNUM - fetch a persistent db's sequence number.
      tests: temporarily disable the transaction test tool.
      Revert "recovery: add special pull-logic for persistent databases"
      tdb: add scripts to extract library symbols (exports file) from headers
      tdb: add script to extract signatures from header files.
      tdb: add script/abi_checks.sh. check for abi changes without gcc magic.
      One would expect I could spell my name... (cherry picked from samba commit 0d120be36bfc561e3f679d081993ccc6bea2a401)
      tdb:mksyms: allow double pointer return value of functions.
      tdb:mksyms: allow characters after closing functions parenthesis.
      tdb:mksigs: ignore struct forward declarations.
      tdb:mksigs: correctly ignode multiline function typedefs
      tdb:mksigs: ignore symbols (like _DEPRECATED_) after closing function parentheses
      tdb:mksigs: normalize bool -> _Bool
      tdb:mksigs: allow PRINTF_ATTRIBUTE(..) macros function types as funcion args
      tdb:tdbtool: add the "speed" command to the help text.
      tdb:tdbtool: add transaction_start/_commit/_cancel commands.
      tdb:tdbtool: fix indentation.

Ronnie Sahlberg (20):
      make sure to also check that interfaces used for NATGW are ok
      Bond devices can have any name the user configures, so
      Merge commit 'rusty/script-report'
      From: Volker Lendecke <vl at samba.org>
      remove the variable "disable when unhealthy"
      cleanup:   remove a tunable we no longer use in the eventscripts any more :
      Rename the tunable EventScriptBanCount to EventScriptTimeoutCount
      From rusty:
      When setting up the logging, set the event to trigger a read of a log message from a child process as a child of the "log" structure and not the ctdb structure,
      ctdb_fetch requires the number of nodes being specified.
      add a new test tool that just locks and releases the same record over and over
      Merge commit 'obnox/ctdb-wip-trans3' into trans3
      Revert "Make fetch_locked more scalable"
      This is a dodgy patch.
      Author: Rusty Russell <rusty at rustcorp.com.au>
      Merge branch 'trans3'
      Revert "cleanup:   remove a tunable we no longer use in the eventscripts any more :"
      fix conflict in merge from metze
      fix a conflict in the merge from rusty
      version 1.0.109

Rusty Russell (50):
      eventscript: reduce code duplication for ending a script, and fix bug
      eventscript: marshall onto last_status immediately
      eventscript: use -ETIME for timeout status value
      talloc: save errno over talloc_free
      eventscript: check_executable() to centralize stat/perm checks
      eventscript: enhance script delete race check
      eventscript: use -ENOEXEC for disabled status value
      eventscript: handle and report generic stat/execution errors
      eventscript: simplify ctdb_run_event_script loop
      eventscript: replace other -1 returns with -errno
      eventscript: handle v. unlikely timeout race
      eventscript: expost ctdb_ban_self()
      eventscript: handle banning within the callbacks
      eventscript: don't make ourselves healthy if we're under ban_count
      eventscript: hoist work out of child process, into parent
      eventscript: split ctdb_run_event_script into multiple parts
      eventscript: make current_monitor_status_ctx serve as monitor_event_script_ctx
      eventscript: use direct script state pointer for current monitor
      eventscript: store from_user and script_list inside state structure
      eventscript: fork() a child for each script.
      eventscript: refactor forking code into fork_child_for_script()
      eventscript: remove unused ctbd_ctrl_event_script*
      eventscript: pass struct ctdb_log_state directly to ctdb_log_handler().
      eventscript: ctdb_fork_with_logging()
      eventscript: use scripts array directly, rather than separate list
      eventscript: record script status for all events
      eventscript: use an array rather than a linked list of scripts
      eventscript: get_current_script() helper
      eventscript: rename ctdb_monitoring_wire to ctdb_scripts_wire
      eventscript: use wire format internally for script status.
      eventscript: cleanup finished to take state arg
      eventscript: save state for all script invocations
      eventscript: get rid of ctdb_control_event_script_finished altogether
      ctdb: support --machinereadable (-Y) for scriptstatus
      eventscript: lock logging on timeout.
      eventscript: expost call names and enum
      ctdb: scriptstatus can now query non-monitor events
      eventscript: fix monitoring when killed by another script command
      ctdb: don't print OUTPUT: for DISABLED scripts
      There is one signedness issue in tdb which prevents traverses of TDB records over the 2G offset on systems which support 64 bit file offsets. This fixes that case.
      lib/tdb: don't overwrite TDBs with different version numbers.
      lib/tdb: TDB_TRACE support (for developers)
      lib/tdb: wean off TDB_ERRCODE.
      lib/tdb: add -t (always use transactions) option to tdbtorture
      lib/tdb: add tdb_check()
      lib/tdb: make tdbtool use tdb_check() for "check" command
      Add --valgringing flag instead of --nosetsched
      Remove RT priority, use niceness.
      ctdb: use mlockall, cautiously
      eventscript: remove cb_status, fix uninitialized bug when monitoring aborted

Simo Sorce (1):
      Fix release script with newer versins of git

Slava Semushin (1):
      lib/tdb/tools/tdbtorture.c: fixed memory leak.

Stefan Metzmacher (49):
      events/50.samba: only use wbinfo --ping-dc if available
      Fix all warnings in source3 with gcc4.3. Jeremy. (cherry picked from samba commit 07e0094365e8dc360a83eec2e7cf9b1d5d8d6d00)
      The msync manpage reports that msync *must* be called before munmap. Failure to do so may result in lost data. Fix an ifdef check, I really think we meant to check HAVE_MMAP here. (cherry picked from samba commit 74c8575b3f3b90ea21ae6aa7ccd95947838af956) (This used to be commit 8fd54bb55f0c23bd025d1719abcbe75c6a2ea8ac) (cherry picked from samba commit b39e332bd7803dd4ce8267f750c497b5deb837ee)
      Add simple manually written replacement for the tdb module. (cherry picked from samba commit 2a61fd41e9d4d2e3d7b47a816544c88bb369bc03)
      py: Properly increase the reference counter of Py_None. (cherry picked from samba commit d2c70d24e12293d9b4272eb310a6a4c4582b2d92)
      s3/s4 build: Fix Py_RETURN_NONE to work with python versions < 2.4 (cherry picked from samba commit 61a23c5eea2203721a70a4ad41bfdc6cd1d6a2a7)
      Avoid using a utility header for Python replacements included in Samba, since this will not be shipped with talloc/tdb/tevent/etc. (cherry picked from samba commit ba5d6e6d706ebf07640ee24d701255cbb4803525)
      lib/tdb: include replace.h and system/filesys.h in pytdb.c
      s3 build: Remove unused fstat check to fix a bunch of HAVE_FSTAT warnings (cherry picked from samba commit 2c2545d45af9c7479bf032fff1263e102926a495)
      lib/tdb: sync build files from samba master
      tdb: rename 'struct list_struct' into 'struct tdb_record'
      tdb: always set tdb->tracefd to -1 to be safe on goto fail
      tdb: add TDB_DISALLOW_NESTING and make TDB_ALLOW_NESTING the default behavior
      tdb: change version to 1.2.0 after adding TDB_*ALLOW_NESTING
      ctdb: pass TDB_DISALLOW_NESTING to all tdb_open/tdb_wrap_open calls
      config: load 'ctdb' config before 'nfs' config in statd-callout
      server: create tdbs with 0600 permissions in ctdb_local_attach()
      server: add db_directory_state to ctdb_context
      server: only do the mkdir() calls for db_directory* once at the start
      server: create recdb.tdb.X in /var/ctdb/state/
      server: create vactun.tdb.X under /var/ctdb/state
      server: create vactune.tdb.X with 0600 permissions
      server: open /var/ctdb/state/persistent_health.tdb.X on startup
      server/recovery: update flags on nodes before syncing dbs
      server: move error handling to a 'fail' label in ctdb_control_transaction_commit()
      server: Use tdb_check to verify persistent tdbs on startup
      server: add CTDB_CONTROL_DB_SET_HEALTHY and CTDB_CONTROL_DB_GET_HEALTH
      client: add ctdb_ctrl_getdbhealth()
      tools/ctdb: display db health in "ctdb getdbmap"
      tools/ctdb: add "ctdb getdbstatus <dbname>"
      tools/ctdb: let "ctdb restoredb" and "ctdb wipedb" mark the db as healthy on all
      client: make ctdb_dumpdb_record() public
      client: add "ctdb dumpdbbackup <filename>"
      ctdb: disallow "ctdb backupdb" on unhealthy databases
      ctdb: add machinereadable output fot "ctdb -Y getdbmap"
      ctdb: print out some hints how to debug a "ctdb catdb" failure
      config: try to use tdbtool <tdb> check instead of tdbdump for persistent db checks
      config: add CTDB_MAX_PERSISTENT_CHECK_ERRORS option
      doc/ctdb.1: document "ctdb wipedb"
      doc/ctdb.1: document UNHEALTHY for "ctdb getdbmap"
      doc/ctdb.1: document -Y output fot ctdb getdbmap
      doc/ctdb.1: document "ctdb dumpdbbackup <file>"
      doc/ctdb.1: add "See also" for ctdb getdbmap
      doc/ctdb.1: document "ctdb getdbstatus <dbname>"
      doc/ctdb.1: make clear the database is specified by name for "ctdb backupdb"
      doc/ctdb.1: update example "ctdb listvars" output
      doc: fix docbook warnings for ctdb.1 and onnode.1 manpages
      doc: regenerate manpages
      ctdb: pass TDB_DISALLOW_NESTING to all tdb_open/tdb_wrap_open calls

Tim Prouty (2):
      tdb: Fix some recently introduced warnings in tdbtool (cherry picked from samba commit c299833bf8e6506c793d6e8283743949aaac9ef4)
      s3/s4: Fix "shadows a global declaration" warning (cherry picked from samba commit e48a5cd5d4fc2626b09af16a0ac68c638d081437)

Volker Lendecke (7):
      Run only one event for each epoll_wait/select call
      Rename a struct member for clarity
      Tiny simplification of ctdb_queue_packet()
      Import "talloc_array_length" from upstream talloc
      Make fetch_locked more scalable
      Attempt to fix bug 5684
      tdb: Fix a C++ warning (cherry picked from samba commit be88a126ea50ea7e41583f402013c63234305e48)

-----------------------------------------------------------------------


-- 
CTDB repository


More information about the samba-cvs mailing list