selftest: Remove samba3.smb2.lock.*.rw-exclusive from flapping file

This test demonstrates a problem with byte range locks and AIO.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Dec  6 05:19:37 CET 2013 on sn-devel-104

selftest: Run smb2.lock tests also against AIO share

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

selftest: Introduce share for testing AIO

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

s3: Return correct error code from SMB2 AIO read failure

This is similar to commit 27e20d5d60ea8aa526bcb7c2dfc18dd2de0bb97b which
fixed the same case for SMB2 writes: When sending the AIO read fails,
return the real error instead of mapping it to NT_STATUS_FILE_CLOSED.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

s3-aio: Use correct locking context for SMB2

The synchronous SMB2 reads and writes use open_persistent_id. The AIO
codepathes have to use the same, otherwise a write will conflict with a
lock on the same open file.

Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

s3-torture: Call cli_flush().

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Fri Dec  6 03:07:32 CET 2013 on sn-devel-104

krb5_wrap: Move function declaration outside the function.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3-libsmb: Fix a memory leak in get_pdc_ip().

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3-torture: Make sure status is used initialized.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3-torture: Fix several memory leaks in delete_fn().

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3-utils: Fix a memory leak in smbget.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Use talloc for asprintf.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Fix out of bound array access.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s3-torture: Comment out dead code in spoolss test.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Fix some compile warnings in spoolss.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Make sure handles are correctly initialized.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Make sure the handles are initialized correctly.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-torture: Make sure the handles are not used uninitialized.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

s4-libcli: Add smb2_util_handle_empty().

Will be used in the next commit.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

torture: match Windows responses to bad shadow copy IDs

test_fsrvp_bad_id() issues DeleteShareMapping requests with invalid
shadow copy IDs, expect Windows Server 2012 return codes.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

torture: test duplicate shares in FSRVP shadow-copy set

MS-FSRVP 3.1.4.4 documents the requirement for failure when duplicate
volume names are encountered on AddToShadowCopySet.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

smbd: Fix a panic when a smb2 brlock times out

Found by Peter Somogyi.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Dec  5 21:21:35 CET 2013 on sn-devel-104

smbtorture: New torture test for bug #9870.

Not fetching the latest modification time on a folder if we have read locks on it.

Prove we should just rely on the mtime value from the underlying
filesystem, even with an open handle.

  BUG: https://bugzilla.samba.org/show_bug.cgi?id=9870

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Thu Dec  5 10:05:06 CET 2013 on sn-devel-104

smbd - allow updates on directory write times on open handles.

If we set a non-null 'old timestamp' in the share mode database
when creating a directory handle, this prevents mtime (write time)
updates from being seen by clients, as we will always return the
timestamp stored in the database whilst the handle is open.

For files this is ok, as we update the stored timestamp
ourselves when we write to the handle. For directories
we should just rely on the mtime value from the underlying
filesystem.

Torture test to follow.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=9870

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>

idmap_cache: Use gencache_parse

This avoids a few tallocs and brings down user CPU a bit more

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Thu Dec  5 03:06:10 CET 2013 on sn-devel-104

idmap_cache: Use an fstring instead of talloc_asprintf

In a test doing one million uid2sid calls this brings down user CPU from
1.3 seconds to 0.9 seconds. And it saves a few code lines.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

messaging3: Do not go through messages.tdb for self-sends

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

torture: enum snapshots after FSRVP creation

Issue an SMB2 ENUM_SNAPSHOTS ioctl following FSRVP snapshot creation to
cover Explorer previous file version use-cases.

This test will fail against Windows Server 2012, as FSRVP created
snapshots are not exposed via the ENUM_SNAPSHOTS ioctl.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

test: add fake_snap.pl for snapshot simulation

The script simulates snapshots by simply copying file data from the base
path to a snapshot path located under .snapshots/@GMT-%Y.%m.%d-%H.%M.%S/

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb:tests: update README

explain how to run individual tests and test collections and remove mention of
tests/scripts/run_tests which does not exist any more.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb/tests/scripts: Promote scripts/run_tests

The wrapper is not needed anymore.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/scripts: Simplify the top-level test scripts

At the moment run_tests.sh has quite fragile argument processing.  It
needs that annoying "--" between options and tests.  The random
default (mktemp -d) for TEST_VAR_DIR is wrong and is worked around in
various places.

Instead:

* Change the default behaviour to print a summary, add new option -N
  to turn off summary, and remove old -s option.

* Change the default behaviour to run integration tests with local
  daemons, add new options -c to run on a cluster, remove old -l
  option.

* Make $testdir/var the default if the tests are not installed, and
  $(mktemp -d ) the default if tests are installed.

* Move the default tests for local/cluster into scripts/run_tests.

run_tests.sh (and the run_cluster_tests.sh symlink) should behave as
before but with slightly more reasonable defaults.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/scripts: Sort command-line options

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/integration: Decentralise the daemon restart code

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/integration: Update daemons shutdown pseudo-test

Don't scatter the TEST_LOCAL_DAEMONS logic around the code.  Limit it
to the local daemons file.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/simple: Local daemons version of setup_ctdb() overrides

setup_ctdb() doesn't need to do anything on a cluster.  To avoid a
conditional, just override it for local daemons.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/integration: Remove some unused functions

This was the start of some refactorisation that was never completed.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/integration: Remove the time logging code

This hasn't been required for a long time and is probably broken.  If
it is needed in future then we know where to find it.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/simple: Move the local daemons code to its own file

This is just a straight move.  The clever stuff will follow.  :-)

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb/tests/integration: Update NFS tickles test and supporting code

This currently requires an eventscript to be dynamically installed.
This eventscript is only used to help determine when a monitor event
has occurred.  This code is horrible and fragile.

A better way is to just monitor the output of "ctdb scriptstatus".
When changes it changes then a monitor event has occurred.

Also remove the old code that checks for tickle information in shared
storage.  CTDB hasn't done things this way for a long time.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

lib/replace remove orphaned code

this is not compiled and used anymore

Signed-off-by: Christian Ambach <ambi@samba.org>
Reviewed-By: Jelmer Vernooij <jelmer@samba.org>
Autobuild-User(master): Christian Ambach <ambi@samba.org>
Autobuild-Date(master): Wed Dec  4 22:55:12 CET 2013 on sn-devel-104

s3:utils remove orphaned code

this does not even compile at all.. looks like a real orphan

Signed-off-by: Christian Ambach <ambi@samba.org>
Reviewed-By: Jelmer Vernooij <jelmer@samba.org>

smbd: Avoid calling notify_filter_string for low debuglevels

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Dec  4 20:09:39 CET 2013 on sn-devel-104

ldb: use of NULL pointer bugfix

Autobuild-User(master): Simo Sorce <idra@samba.org>
Autobuild-Date(master): Tue Dec  3 21:13:53 CET 2013 on sn-devel-104

s4-rpc: dnsserver: Ignore duplicate dns zones from multiple locations

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Kai Blin <kai@samba.org>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Sat Nov 30 06:15:52 CET 2013 on sn-devel-104

s4-dns: Ignore duplicate dns zones from multiple locations in dlz_bind9

BIND DLZ module currently loads DNS zones from DNS partitions and domain
partitions using following prefixes:

   CN=MicrosoftDNS,DC=DomainDnsZones
   CN=MicrosoftDNS,DC=ForestDNSZones
   CN=MicrosoftDNS,CN=System

Windows supports DNS zones duplicated in DNS partitions and domain
partition and updates both of them simultaneously.

BIND DLZ module can handle DNS zones stored either in DNS partitions
or domain partition, but not both.  This patch ignores duplicate zones
from domain partition and allows BIND9 to work with AD with duplicate
DNS zones.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Kai Blin <kai@samba.org>

netcmd/dns: Catch wildcard patterns when querying for name

DNS query should either be '@' to represent entire zone or a fixed string
and not wildcard search pattern.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Kai Blin <kai@samba.org>

fail authentication for single group name which cannot be converted to sid

furthermore if more than one name is supplied and no sid is converted
then also fail.

Bug: https://bugzilla.samba.org/show_bug.cgi?id=8598

Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andreas Schneider <asn@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Fri Nov 29 15:45:11 CET 2013 on sn-devel-104

replace: fix typo in variable name

13550a2b5eed57084a5d9671d9493a9a2e08d7e3 added a typo causing
compilation failure.

Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Thu Nov 28 18:16:27 CET 2013 on sn-devel-104

pam_winbind: Use strlcat in safe_append_string

We have that available via libreplace, so use it.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Thu Nov 28 14:33:32 CET 2013 on sn-devel-104

replace: Don't run over dst in strlcat

If "d" is not 0-terminated, the pure strlen will read beyond the end
of the given bufsize. strlcat in libbsd deliberately avoids this, so we
should do the same.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: David Disseldorp <ddiss@samba.org>

ctdb:tests: in the stub "ip link show" command use echo instead of cat

This case of "ip link show" does not break autobuild with
"Broken pipe" messages, but let's be consistent.

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Thu Nov 28 09:23:03 CET 2013 on sn-devel-104

autobuild: Run ctdb regression tests

Pair-Programmed-with: Michael Adam <obnox@samba.org>

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

autobuild: build the "ctdb" target with socket-wrapper enabled

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

ctdb:test: remove unused ip2ipmask from integration.bash

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

ctdb:tests:76_ctdb_pdb_recovery: change from using ctdb pstore to ctdb ptrans.

This removes the requirement to create a temporary file
and hence makes this test runnable against local daemons
and against a real cluster without further changes.

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

ctdb:tests:76_ctdb_pdb_recovery: fix a typo in a message

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

ctdb:tests:76_ctdb_pdb_recovery: fix a typo in a message

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

ctdb:tests: in the stub ip command, avoid broken pipe by using echo instead of cat

This fixes running "make autotest" from autobuild, since
it prevents irritating error output in delete_ip_from_iface()
when calling ip addr list ... | grep -Fq "inet ..." .

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

ctdb:tests: ignore SIGPIPE in unit tests

This makes scripts called in the unit tests behave like
when called from ctdbd which ignodes SIGPIPE.
This also makes the scrips behave the same when
called from "make autotest" directly and via autobuild (python).

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

ctdb:tests: add path of ip command to extra output in the error case

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

ctdb:tests: remove old now unused script test/recover.sh

Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>

ctdb:build: Update mkversion.sh to use samba-* tags

To test CTDB on a cluster we need to be able to build test RPMs with
relatively sane version numbers.  This is a minimal change to allow
that to happen, until CTDB is integrated into the Samba build system.

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

Cope with first element in hdb_method having a different name in different heimdal versions.

It's called `interface_version` in older Heimdal versions and
`version` in newer versions.

Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Thu Nov 28 04:17:55 CET 2013 on sn-devel-104

Use heim_octet_string typedef.

The struct version is not provided by newer versions of heimdal.

Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

Remove no longer used et_deps.pl.

Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

Remove no longer used asn1_deps.pl.

Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

Add a basic guide on pytalloc.

Signed-off-by: Jelmer Vernooij <jelmer@samba.org>
Reviewed-By: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Jelmer Vernooij <jelmer@samba.org>
Autobuild-Date(master): Thu Nov 28 02:24:45 CET 2013 on sn-devel-104

ctdb:tests/simple: Nobody looks at /tmp/recloop.out so use /dev/null instead

Otherwise this should use mktemp, something should look at the output
and the file should be removed.  :-)

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Wed Nov 27 20:39:00 CET 2013 on sn-devel-104

ctdb:tests: run_tests should ignore bogus test directories

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb:tests: New "autotest" Makefile.in target

This needs to run with socket wrapper and needs to stop after the 1st
error.

Pair-Programmed-With: Michael Adam <obnox@samba.org>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>

Signed-off-by: Martin Schwenke <martin@meltin.net>
Signed-off-by: Michael Adam <obnox@samba.org>
Signed-off-by: Amitay Isaacs <amitay@gmail.com>

ctdb:tests: Add -S option to support socket wrapper

Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>

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

ctdb:tests: Rework unit test result filtering

Using a variable is too fragile, so use a function instead.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb:tests: remove unused setup_nmap_output_filter()

The tests that used it have gone.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb:tests/integration: Be more careful when killing ctdbd

Also match $TEST_VAR_DIR in the socket name.  This means that we'll
only ever kill ctdbd process belong to our own test run.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb: Update NEWS

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-scripts: Be careful when generating unique pids for stack traces

sort expects the data to be line based, so make it so.

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

ctdb-config: Simplify the default CTDB configuration file

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Pair-programmed-with: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-scripts: Replace hard-coded /var/ctdb with CTDB_VARDIR

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

ctdb-scripts: Set defaults for CTDB_DBDIR and CTDB_DBDIR_PERSISTENT

If these configuration variables are not defined, then there should
a default fallback.  This is a workaround till CTDB compile time
configuration can be accessed at runtime.

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

ctdb-eventscripts: Perform share check before NFS RPC checks in 60.ganesha

If NFS RPC checks do restart Ganesha, then it's possible that share
check can fail prematurely.

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

ctdb-tools/ctdb: Improve error checking when parsing node string

If a node isn't numeric then it is silently converted to 0.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-recoverd: Only respond to currently queued ipreallocated requests

Otherwise new requests can come in during the latter parts of the
takeover run when the IP allocation algorithm has already run, and the
new requests will be dequeued even though they haven't really be
processed.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-scripts: Add an early exit to statd-callout's notify case

If $statd_state is empty then the loop will run once and print
spurious errors.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-eventscripts: Remove the nfs_statd_update() call from 60.ganesha

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-tests/integration: Neaten up some of the persistent database tests

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-tools/ctdb: Fix tstore command to generate ltdb header internally

This fixes an alignment discrepancy on 32-bit vs 64-bit platforms.

  sizeof(struct ctdb_ltdb_header) = 20  (32-bit)
                                  = 24  (64-bit)

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

ctdb-tests/takeover: Fix bogus test description

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-tests/simple: User sleep_for() instead of sleep

Progress...

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-tests/simple: Update persistent DB tests

* Low level DB checks should ignore the sequence number record.

* A restart is needed after messing with the RecoverPDBBySeqNum
  tunable.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-recoverd: For persistent databases a sequence number of 0 is valid

Otherwise recovery ends up done by RSN when it is unnecessary.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-locking: Use vfork instead of fork to exec helpers

There is a significant overhead using fork() over vfork(), specially
when the child process execs a helper.  The overhead is in memory space
and time.

    # strace -c ./test_fork 1024 200
    count=1024, size=204800, total=200M
    failed fork=0
    time for fork() = 4879.597000 us
    % time     seconds  usecs/call     calls    errors syscall
    ------ ----------- ----------- --------- --------- ----------------
    100.00    4.543321        3304      1375       375 clone
      0.00    0.000071           0      1033           mmap
      0.00    0.000000           0         1           read
      0.00    0.000000           0         3           write
      0.00    0.000000           0         2           open
      0.00    0.000000           0         2           close
      0.00    0.000000           0         3           fstat
      0.00    0.000000           0         3           mprotect
      0.00    0.000000           0         1           munmap
      0.00    0.000000           0         3           brk
      0.00    0.000000           0         1         1 access
      0.00    0.000000           0         1           execve
      0.00    0.000000           0         1           arch_prctl
    ------ ----------- ----------- --------- --------- ----------------
    100.00    4.543392                  2429       376 total

    # strace -c ./test_vfork 1024 200
    count=1024, size=204800, total=200M
    failed fork=0
    time for fork() = 82.041000 us
    % time     seconds  usecs/call     calls    errors syscall
    ------ ----------- ----------- --------- --------- ----------------
     96.47    0.001204           1      1000           vfork
      3.53    0.000044           0      1033           mmap
      0.00    0.000000           0         1           read
      0.00    0.000000           0         3           write
      0.00    0.000000           0         2           open
      0.00    0.000000           0         2           close
      0.00    0.000000           0         3           fstat
      0.00    0.000000           0         3           mprotect
      0.00    0.000000           0         1           munmap
      0.00    0.000000           0         3           brk
      0.00    0.000000           0         1         1 access
      0.00    0.000000           0         1           execve
      0.00    0.000000           0         1           arch_prctl
    ------ ----------- ----------- --------- --------- ----------------
    100.00    0.001248                  2054         1 total

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

ctdb-common: Refactor code to keep track of child processes

This code can then be used to track child processes created with vfork().

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

ctdb-scripts: Run a single instance of debug_locks.sh at a give time

This prevents spamming of logs if multiple lock requests are waiting
and keep timing out.

Also, improve the logging format with separators.

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

ctdb-locking: Update current lock statistics when lock is scheduled

When a child process is created for a lock request, the current locks
statistics should be updated immediately.  This will provide accurate
information on number of active lock requests.

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

ctdb-locking: Do not merge multiple lock requests to avoid unfair scheduling

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

ctdb-locking: Implement active lock requests limit per database

This limit was currently a global limit and not per database.  This
prevents any database freeze lock requests from getting scheduled if
the global limit was reached.

Only individual record requests should be limited and database freeze
requests should always get scheduled.

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

ctdb-scripts: Rewrite statd-callout to avoid 10 minute lag

This is naive and assumes no performance problems when updating
persistent DBs.  It also does no error handling.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Pair-programmed-with: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Michael Adam <obnox@samba.org>

ctdb-client: Treat empty __db_sequence_number__ record as 0

This fixes the issue of transaction commit failing due to an empty
__db_sequence_number__ record in persistent database left by previous
cancelled transaction.

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

ctdb-doc: Update ctdb.1 - primarily to add pdelete/pfetch/pstore/ptrans

Also:

* More <refentryinfo> above <refmeta> to make the XML valid.

* Describe DB argument in introduction and use it for database
  commands.

* Remove unnecessary format="linespecific" from <screen> tags, since
  it will not be allowed in DocBook 5.0.

* Sort the items in "INTERNAL COMMANDS".

* Update/simplify some command descriptions.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Michael Adam <obnox@samba.org>