IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
in_fd is coming soon.
Fix coding style violations in the affected and adjacent lines.
Modernise some debug macros and make them more consistent (e.g. drop
logging of errno when strerror(errno) is already logged.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14084
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This is the core logic from ctdb_ip_to_pnn(), so re-implement that
that function using ctdb_ip_to_node().
Something similar (ctdb_ip_to_nodeid()) was recently removed in commit
010c1d77cd because it wasn't required.
Now there is a use case.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14084
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Compiling with -Wsign-compare complains:
1047 | && (call->call_id == CTDB_FETCH_WITH_HEADER_FUNC)) {
| ^~
struct ctdb_call is a protocol element, so we can't simply change it.
Found by csbuild.
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): Wed Aug 14 10:29:59 UTC 2019 on sn-devel-184
Compiling with -Wsign-compare complains:
ctdb/server/ctdb_call.c:831:12: warning: comparison of integer expressions of different signedness: ‘int’ and ‘uint32_t’ {aka ‘unsigned int’} [-Wsign-compare]
831 | if (count <= ctdb_db->statistics.hot_keys[0].count) {
| ^~
and
ctdb/server/ctdb_call.c:844:13: warning: comparison of integer expressions of different signedness: ‘int’ and ‘uint32_t’ {aka ‘unsigned int’} [-Wsign-compare]
844 | if (count <= ctdb_db->statistics.hot_keys[i].count) {
| ^~
Found by cs-build.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Compiling with -Wsign-compare complains:
ctdb/server/ctdb_daemon.c: scope_hint: In function ‘daemon_queue_send’
ctdb/server/ctdb_daemon.c:259:40: warning: comparison of integer expressions of different signedness: ‘int’ and ‘uint32_t’ {aka ‘unsigned int’} [-Wsign-compare]
...
The struct ctdb_queue member out_queue_length is actually uint32_t, so
just return that type.
Found by csbuild.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Should be used when public IP addresses are not assigned.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
If there is a failure it will always be the last test run. Don't
obscure this by following it with a summary.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
These files now pass, so we might as well keep them in good shape.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This can be used to find ctdb_run_tests and ctdb_local_daemons.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Given other improvements, this is now needlessly complex.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
SC2045 Iterating over ls output is fragile. Use globs.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
SC2034 summary appears unused. Verify use (or export if used externally)
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
SC2155 Declare and assign separately to avoid masking return values
The wscript changes require an identical change in local_daemons.sh.
While touching the lines in wscript, escape the backslashes to make
them literal backslashes.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
SC2164 Use 'cd ... || exit' or 'cd ... || return' in case cd fails
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
SC2086 Double quote to prevent globbing and word splitting
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Commit 90de5e0594 remove the processing
for this option but forgot to remove it from the getopts command.
Versions of ShellCheck >= 0.4.7 warn on this, so it is worth fixing.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14086
RN: Fix onnode test failure with ShellCheck >= 0.4.7
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
On older systems like SLES 11 without POSIX robust mutexes, but with glib robust
mutexes where all the functions are available but have a "_np" suffix,
compilation fails in:
ctdb/tests/src/test_mutex_raw.c.239.o: In function `worker':
/root/samba-4.10.6/bin/default/../../ctdb/tests/src/test_mutex_raw.c:129: undefined reference to `pthread_mutex_consistent'
ctdb/tests/src/test_mutex_raw.c.239.o: In function `main':
/root/samba-4.10.6/bin/default/../../ctdb/tests/src/test_mutex_raw.c:285: undefined reference to `pthread_mutex_consistent'
/root/samba-4.10.6/bin/default/../../ctdb/tests/src/test_mutex_raw.c:332: undefined reference to `pthread_mutexattr_setrobust'
/root/samba-4.10.6/bin/default/../../ctdb/tests/src/test_mutex_raw.c:363: undefined reference to `pthread_mutex_consistent'
collect2: ld returned 1 exit status
This could be fixed by using libreplace system/threads.h instead of pthreads.h
directly, but as there has been a desire to keep test_mutex_raw.c standalone and
compilable without other external depenencies then libc and libpthread, make the
tool developer build only. This should get the average user over the cliff.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14038
RN: Fix compiling ctdb on older systems lacking POSIX robust mutexes
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
The "init" event is only run once so don't bother caching the
configured value of the recovery lock. Add some extra error checking.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Fri Jul 26 04:52:04 UTC 2019 on sn-devel-184
These tests currently do not do what is expected. They test the
default case.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
If the lock file is inaccessible or the inode number changes then the
lock is lost, so exit. This allows the recovery daemon to trigger an
election. The ensuing recovery will re-take the lock.
By default the lock file is checked every 60 seconds. A lot can
happen in 60 seconds but being more aggressive and accessing the lock
too often could result in a performance issue for the cluster
filesystem.
An new optional 2nd argument is added, which is the lock file re-check
time in seconds.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This will allow more conditions to be waited on via additional
sub-requests. At the moment this just completes when the parent wait
completes.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Put the checking for the process being immediately re-parented into
the computation too. This will be very rare and doing it
consistently makes testing saner.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
There is no need to wait until the parent kills the helper. The
parent will get the initial response, indicating contention or
similar, and will then get a separate event indicating that the pipe
is gone.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This makes the code more explicit and makes testing easier due to less
dependencies.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
clang warns:
ctdb/server/ctdb_mutex_fcntl_helper.c:61:3: warning: Value stored to 'fd' is never read
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
One of these had a missing space, so this implicitly fixes it. It
also drops the need to unnecessarily include common.h, which comes
with some dependency baggage.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Only do this if the recovery lock is unset. Log every minute for the
first 10 minutes, then every 10 minutes, then every hour.
This is useful for determining whether a split brain occurred. It is
particularly useful if logging failed or was throttled at startup, so
there is no evidence of the split brain when it began.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
In tini, allow_empty_value=false causes the parser to ignore the lines
without '=' sign, but lines with nothing after '=' sign are allowed and
cause empty string ("") to be passed as a value.
This is counter-intuitive, so conf requires special handling for empty
values (which are treated as invalid).
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Regression introduced by commit
2558f96da1. count should be signed
because list_of_connected_nodes() returns -1 on failure. Variable i
is used in both signed and unsigned contexts, so add new signed
variable j for use in signed context.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14017
CTDB should start as a disabled unit (systemd) in most of the
distributions and, when trying to enable it for the first time, user
should get an unconfigured, or similar, error.
Depending on /etc/ctdb/nodes file will give a clear direction to final
user on what is needed in order to get cluster up and running. It should
work like previous ENABLED=NO variables in SySV like initialization
scripts.
Signed-off-by: Rafael David Tinoco <rafaeldtinoco@ubuntu.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This makes it consistent with print-socket.
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 Jul 5 06:19:11 UTC 2019 on sn-devel-184
csbuild doesn't like the hack where variable buf is initialised to
itself to avoid an unused variable warning. buf is unused so remove
it instead.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This can never be NULL. It could probably be NULL in the past when
"all database" locks existed.
There are paths where is is checked for NULL and then later
dereferenced, causing static analysers to produce spurious warnings.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This avoids static analysers continuing analysis after calls to these
functions and producing incorrect warnings.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Otherwise ret == 0 is returned from successful call to
ctdb_int32_pull().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
According to the documentation, sendto() should either send the packet
as given or return with an error. However, given that it can return
the number of bytes sent, treat the theoretical error of a short
packet send separately, since errno would not be set in this case.
Similarly, treat a short packet recv() separately from an error where
errno is set.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This avoids an unnecessary signed/unsigned comparison issue.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
clang reports:
ctdb/protocol/protocol_types.c:5191:3: warning: Value stored to 'ret' is never read
Found by csbuild.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Indexing by PNN is wrong.
This also removes a signed/unsigned comparison because the PNN is not
compared to -1 anymore.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Node ID is a poorly defined concept, indicating the slot in the node
map where the IP address was found. This signed value also ends up
compared to num_nodes, which is unsigned, producing unwanted warnings.
Just return the PNN because this what both callers really want.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Static analysis finds that earlier in the call path, ctdb_string_len()
checks for NULL, so complains that a NULL value can be passed to
strlen() here. Avoid this by adding an assert().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The dummy reader should never be called, so contains an assert on the
buffer length that should always trigger. Just abort() instead.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
These are all cases comparing a number of bytes written (int or
ssize_t) with a size_t, so casting to size_t is appropriate.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Change declarations of variable and parameters, usually loop variables
and limits, from int to unsigned int, size_t or uint32_t.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This improves readability. Also, the asserts involving this
expression get more complicated in the next commit, so this will keep
those asserts within a single line.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
... and does not just contain whitespace.
Otherwise NULL can be passed as the first argument to execv().
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Instead of taking exclude_pnn as a parameter, calculate it from an
include_self_parameter, which is passed through from the 2 calling
functions.
While doing this, fix a signed/unsigned comparison issue by declaring
the new exclude_pnn local variable as an unsigned type.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The next commit will change the type of this function, which is only
used in this file. So, make it static to isolate the change.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This has been broken for 10 years since commit
9616959bd6, which introduced the
separate filtering. This commit was missing a redirect of the output
of stderr_filter() to stderr.
Since nobody depends on the separate filtering (i.e. nobody reported a
bug), just return to combining stdout and stderr, and filtering them
together.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This filter no longer does anything useful in this context. By
default it adds a pipeline with trailing cat process. In many
contexts, stdout of the process being run is still open so the cat
process will stay around and will stop onnode from exiting.
The filters should all go away because they are simply an example of
code that is trying to be too clever while causing unfortunate corner
cases.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
I don't think anyone uses this and it causes complications.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Yes, the other callers check the return value of ctdb_lockdb_mark().
However, this is called in a void function and ctdb_lockdb_mark() has
already printed any error message. All we can do is explicitly ignore
the return value.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
To make this much clearer, move the declaration into the scope where
it is used.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This fixes warnings about signed versus unsigned comparisons.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Simple cases where variables and function parameters need to be
declared as an unsigned type instead of an int.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Simple cases where variables need to be declared as an unsigned type
instead of an int.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Simple cases where a variables and function parameters need to be
declared as an unsigned type instead of an int.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The incremented value of argc is indeed never used. Leave it as a
comment to warn anyone cutting and pasting the code.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
In one case, given triviality of change, add missing braces and fix
whitespace.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Simple cases where a variable (usually a loop variable) needs to be
declared as an unsigned type (unsigned int or size_t) instead of an
int.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
All the top-level callers pass size_t.
Drop the ternary operator. The value of hsize is always positive
because it is unsigned.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
There's no point using unsigned here. tdb_traverse() returns an int
for the number of records traversed and the number of empty records
can't exceed this value.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This needs an extra variable because variable i has been used in both
signed and unsigned contexts.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
These are the simple cases where a variable (usually a loop variable)
needs to be declared as an unsigned type (usually unsigned int or
size_t) instead of an int.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
net.ipv4.tcp_tw_recycle has been removed from Linux 4.12 but, still,
makes sense to check its existence. Unfortunately, current check does
not test for the procfs file existence. This commit fixes the issue.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13984
Signed-off-by: Rafael David Tinoco <rafaeldtinoco@ubuntu.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Jun 4 23:31:24 UTC 2019 on sn-devel-184
The caller should be able to call TALLOC_FREE() on the returned
strings.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Fixes
ctdb/server/ipalloc_lcp2.c:61: error: shiftTooManyBitsSigned: Shifting signed 32-bit value by 31 bits is undefined behaviour <--[cppcheck]
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andreas Schneider <asn@samba.org>
Old war story completely from memory, I could not find the commit that
introduced TDB_SEQNUM so far...:
Back in the days when ctdb was initially developed, TDB_SEQNUM's only
user was the notify.tdb that held one huge record for all notify
records. With that use case in mind it made perfect sense to keep the
SEQNUM stable locally, sacrificing precision. By now notify.tdb is
long gone, an the only user of TDB_SEQNUM right now is brlock.tdb,
which contains special case code for the imprecise ctdb implementation
of TDB_SEQNUM.
With this commit, that special code can go: The TDB_SEQNUM will also
increment when just the DMASTER header field changes, indicating to
smbd that someone else might have changed the record. This will of
course increase the SEQNUM frequency, but it should not increase the
load on ctdb: If you look at the brlock.c workaround, it just does not
do the caching that is possible with precise TDB_SEQNUMs working.
How did I get here? I want to move brl_num_read_oplocks() from
brlock.tdb into locking.tdb, and for that I need precise TDB_SEQNUMs
for locking.tdb.
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): Fri May 24 00:42:17 UTC 2019 on sn-devel-184
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 May 23 18:08:36 UTC 2019 on sn-devel-184
state is always freed before exiting this function, so allocate fde
off it instead of long-lived ctdb context.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13943
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
There is a chance that restoring IP addresses to the test node will
result in different IP addresses being assigned to that node.
Removing a single IP address may then fail (or be a no-op) if it is
done after the restore.
So, swap the single IP address removal to happen first, then restore,
then remove all IP addresses.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
ctdb reloadips will fail if it can't disable takover runs. The most
likely reason for this is that there is already a takeover run in
progress. We can't predict when this will happen, so retry if this
occurs.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Otherwise, when looping tests for a long time, nodes are unable to
connect to each other.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
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): Mon May 13 08:42:44 UTC 2019 on sn-devel-184
ctdb_control_db_attach() and ctdb_control_db_detach() assume that any
control with client ID 0 comes from another daemon and treat it
specially.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13930
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
2b5dbb3525 fixed builds with an explicit
--with-libcephfs but broke builds against system Ceph libraries. This
change handles both cases.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Thu May 9 04:24:56 UTC 2019 on sn-devel-184
sock_socket_start_recv() might not fill sockpath if we return early.
Found by GCC 9.
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
error: ‘%04d’ directive writing between 4 and 11 bytes into a region of
size 5 [-Werror=format-overflow=]
sprintf(key, "key%04d", i);
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
If the directory is always cleaned up then it is not possible to look
at daemon logs to debug test failures.
This target is only really used by autobuild.py, which (optionally)
cleans up the parent directory anyway.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
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): Tue May 7 06:56:01 UTC 2019 on sn-devel-184
Since commit 0e9ead8f28 daemons have
been shut down after each test, so this option no longer has anything
to do with killing daemons.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Sometimes the detach test fails:
Check detaching single test database detach_test1.tdb
BAD: database detach_test1.tdb is still attached
Number of databases:4
dbid:0x5ae995ee name:detach_test4.tdb path:tests/var/simple/node.0/db/volatile/detach_test4.tdb.0
dbid:0xd84cc13c name:detach_test3.tdb path:tests/var/simple/node.0/db/volatile/detach_test3.tdb.0
dbid:0x8e8e8cef name:detach_test2.tdb path:tests/var/simple/node.0/db/volatile/detach_test2.tdb.0
dbid:0xc62491f4 name:detach_test1.tdb path:tests/var/simple/node.0/db/volatile/detach_test1.tdb.0
Number of databases:3
dbid:0x5ae995ee name:detach_test4.tdb path:tests/var/simple/node.1/db/volatile/detach_test4.tdb.1
dbid:0xd84cc13c name:detach_test3.tdb path:tests/var/simple/node.1/db/volatile/detach_test3.tdb.1
dbid:0x8e8e8cef name:detach_test2.tdb path:tests/var/simple/node.1/db/volatile/detach_test2.tdb.1
Number of databases:4
dbid:0x5ae995ee name:detach_test4.tdb path:tests/var/simple/node.2/db/volatile/detach_test4.tdb.2
dbid:0xd84cc13c name:detach_test3.tdb path:tests/var/simple/node.2/db/volatile/detach_test3.tdb.2
dbid:0x8e8e8cef name:detach_test2.tdb path:tests/var/simple/node.2/db/volatile/detach_test2.tdb.2
dbid:0xc62491f4 name:detach_test1.tdb path:tests/var/simple/node.2/db/volatile/detach_test1.tdb.2
*** TEST COMPLETED (RC=1) AT 2019-04-27 03:35:40, CLEANING UP...
When issued from a client, the detach control re-broadcasts itself
asynchronously to all nodes and then returns success. The controls to
some nodes to do the actual detach may still be in flight when success
is returned to the client. Therefore, the test should wait for a few
seconds to allow the asynchronous controls to complete.
The same is true for the attach control, so workaround the problem in
the attach test too.
An alternative is to make the attach and detach controls synchronous
by avoiding the broadcast and waiting for the results of the
individual controls sent to the nodes. However, a simple
implementation would involve adding new nested event loops.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This sometimes fails, apparently due to a cat process in onnode
getting EAGAIN. The conclusion is that tests that process large
amounts of output should not depend on a sub-shell delivering that
output into a shell variable.
Change try_command_on_node() to leave all of the output in file
$outfile and just put the first 1KB into $out. $outfile is removed
after each test completes.
Change the implementation of sanity_check_output() to use $outfile
instead of $out.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
All callers are currently passed $out. Global variable $out is used
in many other places so use it here to simplify the interface and make
future changes simpler.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13924
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
CTDB's system memory monitoring in 05.system.script monitors both main
memory and swap. The swap monitoring was originally based on
the (possibly incorrect, see below) idea that swap space stacks on top
of main memory, so that when a system starts filling swap space then
this is supposed to be a good sign that the system is running out of
memory. Additionally, performance on a Linux system tends to be
destroyed by the I/O associated with a lot of swapping to spinning
disks.
However, some platforms default to creating only 4GB of swap space
even when there is 128GB of main memory. With such a small swap to
main memory ratio, memory pressure can force swap to be nearly full
even when a significant amount of main memory is still available and
the system is performing well. This suggests that checking swap
utilisation might be less than useful in many circumstances.
So, remove the separate swap space checking and change the memory
check to cover the total of main memory and swap space.
Test function set_mem_usage() still takes an argument for each of main
memory and swap space utilisation. For simplicity, the same number is
now passed twice to make the intended results comprehensible. This
could be changed later.
A couple of tests are cleaned up to no longer use hard-coded
/proc/meminfo and ps output.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This is to help us notice when ctdbd is using the full capacity of a
CPU, so is saturated.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13895
In run_proc, there was an implicit assumption that when a process exits,
fd event (pipe between parent and child) would be processed first and
signal event (SIGCHLD for the child) would be processed later.
However, that is not the case. SIGCHLD can be received asynchronously
any time even when the pipe data has not fully been read. This causes
run_proc to miss some of the output from child process in tests.
When SIGCHLD is being processed, if the pipe between parent and child is
still open, then do an explict read from the pipe to ensure we read any
data still in the pipe before closing the pipe.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Fri Apr 12 08:19:29 UTC 2019 on sn-devel-144
We also can not assume that nodes can be marked as connected via only
the keepalive mechanism. Keepalives are not sent to disconnected
nodes so, in the absence of other packets (e.g. broadcasts), 2 nodes
may never become marked as connected to each other.
Revert to marking nodes as connected in the TCP transport code. If a
connection is to a non(-operational) ctdbd then it will revert to
disconnected after a short while and may actually flap. This should
be rare.
This reverts commit 66919db3d7.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13888
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
I thought this was being triggered during automated testing.
However, it appears that a poor choice of fixed ports for NFS RPC
services was the real problem. Revert, since the original behaviour
may be useful.
This reverts commit f1a1c300e1.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The new string conversion wrappers detect and flag errors
which occured during the string to integer conversion.
Those modifications required an update of the callees
error checks.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
The new string conversion wrappers detect and flag errors
which occured during the string to integer conversion.
Those modifications required an update of the callees
error checks.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Christof Schmitt <cs@samba.org>
If a data packet arrives which exceeds the queue's current buffer size,
the buffer needs to be increased to hold the full packet. Once the packet
is processed the buffer size should be decreased to its standard size again.
This test case verifies this process.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Christof Schmitt <cs@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Wed Apr 10 00:17:37 UTC 2019 on sn-devel-144
Some test scenarios require access to the created queue.
Prepare the test_setup function to provide it as additional parameter.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Christof Schmitt <cs@samba.org>
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Sat Apr 6 11:51:55 UTC 2019 on sn-devel-144
While 0 may indicate that all threads have exited after being stuck,
it may also indicate that nfsd should not be running due to being shut
down.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Sun Mar 31 11:47:44 UTC 2019 on sn-devel-144
The alternative seems to be to try something via CTDB_NFS_CALLOUT.
That would be complicated and seems like overkill for something this
simple.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
The situation for NFS config has got more complicated and is probably
broken in statd-callout on Debian-like systems at the moment. Allow
several alternative configuration names to be tried. Stop after the
first that is found and loaded.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
The tests are written around the default of sysvinit-redhat. Add
support for systemd-redhat.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
At least Red Hat and Debian appear to use (a variant of?) the upstream
systemd units for NFS, so adding support for these services is
relatively easy. Distributions using Sys-V init can patch the
call-out to use the relevant Sys-V init services.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
When an NFS check restarts a failed service by hand then systemd will
be unable to stop or start this service again because (at least) the
PID file will be wrong. Do this via the NFS Linux kernel call-out
instead. Allow the call-out to use the services instead of doing
manual restarts. Add variables for mount, status and rquotad services
to support this.
Adding systemd NFS services to the call-out will follow.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
There will be more of these variable for other services so, for
readability, it makes sense for them to start with "nfs_".
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13860
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
We do this by removing the confusing mandatory option to
conf.SAMBA_CHECK_PYTHON{,_HEADERS}(), instead just use the value of
--disable-python internally
This follows the default minimum of Python 3.4 and keeps things consistent
with the main Samba build where --disable-python is required to skip building
python bindings.
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
We now used the default of 3.4 from conf.SAMBA_CHECK_PYTHON()
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
There is no need to write SAMBA_VERSION_STRING as CTDB_VERSION_STRING.
Wherever required use SAMBA_VERSION_STRING directly.
Avoids the confusion with two version.h files.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13789
Signed-off-by: Amitay Isaacs <amitay@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Fri Mar 15 06:31:50 UTC 2019 on sn-devel-144
This way we don't get constant rebuild as SAMBA_VERSION_STRING
is "4.7.0pre1.DEVELOPERBUILD" for the binaries under bin/
instead of "4.7.0pre1.GIT.59e51f6".
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13789
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This can be used to test the version checking logic. Cache the
version to avoid re-checking the environment variable each time.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
This file is included by local_daemons.sh, which is not a bash script
and wait_until() uses the "local" keyword. Prefixing variable names
with '_' to indicate that they are local changes a lot of lines in
this function. So, fix indentation, potential quoting problems and
other ShellCheck hits while touching this function.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
This isn't used anywhere that requires it to be exported, but the lack
of consistency will cause problems and confusion at some later stage.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
ctdb_sock_addr values are hashed in some contexts. This means that
all of the memory used for the ctdb_sock_addr should be consistent
regardless of how parsing is done. The first 2 cases are just sanity
checks but the 3rd case involving an IPv4-mapped IPv6 address is the
real target of this test addition.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13839
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
Failed to kill the tcp connection that using IPv4-mapped IPv6 address
(e.g. ctdb_killtcp eth0 ::ffff:192.168.200.44:2049
::ffff:192.168.200.45:863).
When the ctdb_killtcp is used to kill the tcp connection, the IPs and
ports in the connection will be parsed to conn.client and conn.server
(call stack: main->ctdb_sock_addr_from_string->ip_from_string). In
the ip_from_string, as we are using IPv4-mapped IPv6 addresses, the
ipv6_from_string will be used to parse ip to addr.ip6 first. The next
step the ipv4_from_string will be used to reparse ip to addr.ip.
As a result, the data that dump from conn.server is "2 0 8 1 192 168
200 44 0 0 0 0 0 0 0 0 0 0 255 255 192 168 200 44 0 0 0 0", the data
from conn.client is "2 0 3 95 192 168 200 45 0 0 0 0 0 0 0 0 0 0 255 255
192 168 200 45 0 0 0 0". The connection will be add to conn_list by
ctdb_connection_list_add. Then the reset_connections_send uses conn_list
as parameter to start to reset connections in the conn_list.
In the reset_connections_send, the database "connections" will be
created. The connections from conn_list will be written to the
database(call db_hash_add), and use the data that dump from conn_client
and conn_server as key.
In the reset_connections_capture_tcp_handler, the
ctdb_sys_read_tcp_packet will receive data on the raw socket. And
extract the IPs and ports from the tcp packet. when extracting IP and
port, the tcp4_extract OR tcp6_extract will be used. Then we got the
new conn.client and conn.server. the data that dump from the
conn.server is "2 0 8 1 192 168 200 44 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0", the data from conn.client is "2 0 3 95 192 168 200 45 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0". Finally, we use the data as key to check
if this connection is one being reset(call db_hash_delete). The
db_hash_delete will return ENOENT. Because the two key that being used
by db_hash_delete and db_hash_add are different.
So, the TCP RST will be NOT sent for the connection forever. We should
initialize addr struct to zero before reparsing as IPV4 in the
ip_from_string.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13839
Signed-off-by: Zhu Shangzhong <zhu.shangzhong@zte.com.cn>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
CTDB_CLUSTER_MUTEX_HELPER can't be depended on because it is only set
when the tests are not installed and setting it unconditionally for
this particular use would be wrong.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@samba.org>
Compilation currently fails, as ctdb_mutex_ceph_rados_helper doesn't
include or link against the samba-util library. Revert back to the
previous strtoull() behaviour, which works fine.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Mar 1 18:34:18 UTC 2019 on sn-devel-144
263/386] Compiling ctdb/server/ctdb_recovery_helper.c
In file included from ../../server/ctdb_recovery_helper.c:24:0:
../../server/ctdb_recovery_helper.c: In function ‘main’:
../../../lib/talloc/talloc.h:911:34: error: ‘mem_ctx’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
#define TALLOC_FREE(ctx) do { if (ctx != NULL) { talloc_free(ctx); ctx=NULL; } } while(0)
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Jeremy Allison <jra@samba.org>
In order to detect an value overflow error during
the string to integer conversion with strtoul/strtoull,
the errno variable must be set to zero before the execution and
checked after the conversion is performed. This is achieved by
using the wrapper function strtoul_err and strtoull_err.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
In order to detect an value overflow error during
the string to integer conversion with strtoul/strtoull,
the errno variable must be set to zero before the execution and
checked after the conversion is performed. This is achieved by
using the wrapper function strtoul_err and strtoull_err.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
In order to detect an value overflow error during
the string to integer conversion with strtoul/strtoull,
the errno variable must be set to zero before the execution and
checked after the conversion is performed. This is achieved by
using the wrapper function strtoul_err and strtoull_err.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
In order to detect an value overflow error during
the string to integer conversion with strtoul/strtoull,
the errno variable must be set to zero before the execution and
checked after the conversion is performed. This is achieved by
using the wrapper function strtoul_err and strtoull_err.
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Ralph Böhme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This code is difficult to read and there really is no common code
between the 2 cases. For example, there is no need to split a
filename into words. Separating each of the 2 cases into its own
function makes the logic much easier to understand.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
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): Mon Feb 25 03:40:16 CET 2019 on sn-devel-144
Under the covers, a command is always used. However, there is no way
of testing of the code path where a command is explicitly configured.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Currently this will wait forever. It really needs a timeout in case
the cluster filesystem (or other lock mechanism) is completely wedged.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
We really shouldn't see unknown errors. They probably represent a
misconfigured recovery lock or similar.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Add an explicit case for a timeout and clean up the other messages.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
If nested events occur while the file descriptor handler is still
active then chaos can ensue. For example, if a node is banned and the
lock is explicitly cancelled (e.g. due to election loss) then
double-talloc-free()s abound.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13800
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
ctdbd will start without a recovery lock configured. It will log a
message saying that this is not optimal. However, a careless user may
overlook both this message and the importance of setting a recovery
lock. If the existing example configuration is uncommented then the
directory containing it will be created (by 01.reclock.script) and the
failure (i.e. multiple nodes able to take the lock) will be confusing.
Instead, change the example setting to one that will result in banned
nodes, encouraging users to consciously configure (or deconfigure) the
recovery lock. Tweak the corresponding comment.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13790
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13791
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Fri Feb 22 03:51:37 CET 2019 on sn-devel-144
In order to calculate the number of bytes correctly which
are to be read into the buffer, the buffer.offset must be taken
into account.
This patch fixes a regression introduced by 382705f495.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13791
Signed-off-by: Swen Schillig <swen@linux.ibm.com>
Reviewed-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
It isn't currently possible to distinguish these 2 cases.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Checking that the database contains 0 records cause a traverse. This
may take a lock and cause vacuuming to fail (or be deferred for a
particular record/chain). Minimise the chance of this happening by
only checking for 0 records every 10 seconds.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
When the number of fast path vacuuming runs is 0 then a full vacuuming
run is done. This means the first one is a full run, which is almost
certainly not what is intended.
Combine the 2 conditionals to only flag a full vacuuming run when the
count exceeds the configured limit. This means that the
full_vacuum_run flag is set in both parent and child, but this is
harmless... and is better than getting it wrong.
Also tweak the comparison to be less-than-or-equal, since the zeroth
run needs to be counted.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This allows to build packages even when python3 is not available by
setting PYTHON variable.
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
../../tools/ctdb.c: In function 'str_to_data':
../../tools/ctdb.c:624: warning: 'data.dsize' may be used uninitialized in this function
../../tools/ctdb.c:624: warning: 'data.dptr' may be used uninitialized in this function
Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
This can result in Unix domain socket names that are too long.
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): Tue Dec 18 05:31:00 CET 2018 on sn-devel-144
If CTDB_BASE is relative then symbolic link targets will be incorrect.
Don't force CTDB_BASE to be absolute because this can result in Unix
domain socket names that are too long.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
If CTDB_BASE is relative then the symbolic link target will be
incorrect.
Don't force CTDB_BASE to be absolute because this can result in Unix
domain socket names that are too long.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Event scripts (well, statd_callout) can change directory, causing
stubs to be unable to locate EVENTSCRIPTS_TESTS_VAR_DIR if it is
relative.
Don't force TEST_VAR_DIR to be absolute because this can result in
Unix domain socket names that are too long.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
If TEST_VAR_DIR is relative then symbolic link targets will be
incorrect.
Don't force TEST_VAR_DIR to be absolute because this can result in
Unix domain socket names that are too long.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The lock may have been lost due to a failure in the underlying locking
mechanism. This could be due to quorum loss or similar. It is best
to call an election to confirm that this node should still be master.
At worst, the node will reelect itself, fail to take the lock and then
ban itself. This is a suitable outcome for a node that has been
partitioned from others in the cluster.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This provides finer resolution while still maintaining a reasonable
maximum. In this case the top bucket contains any hop counts
>= 16384, compared to the current situation where the top bucket contains
hop counts >= 268435456.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This avoids creating a mini-configure in the configure script.
Users wishing to use python2 to build need to specify PYTHON=
to both ./configure and make
After we merged the python3 change, it became clear that relying on systems prefixing
the correct python just causes trouble and make debugging harder, so only use $PYTHON
for the override, not the default case
This essentially reverts a660b7fb8e but
leaves the files more consistent.
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Noel Power <npower@samba.org>
Make sure default make and configure for all now defaults
to building with python3.
To build a samba (or sub component e.g. talloc etc.) with python3
./configure && make
To build a samba (or sub component e.g. talloc etc.) with python2
PYTHON=python ./configure && PYTHON=python make
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
The code does not implement saving the record data to a file, so update
the usage info accordingly.
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Mon Dec 10 05:02:13 CET 2018 on sn-devel-144
The received packet is copied into a newly allocated memory chunk for further
processing by the assigned callback. Once this is done, the memory is free'd.
This is repeated for each received packet making the memory allocation / free
an expensive task. To optimize this process, a memory pool is defined which
is sized identically to the queue's buffer.
During tests it could be seen that more than 95% of all messages were sized
below the standard buffer_size of 1k.
Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Christof Schmitt <cs@samba.org>
Autobuild-User(master): Christof Schmitt <cs@samba.org>
Autobuild-Date(master): Fri Dec 7 23:27:16 CET 2018 on sn-devel-144
The memmove operation is quite expensive, therefore,
a new buffer attribute "offset" is introduced to support
an optimized buffer processing.
The optimization is to "walk" through the buffer and process
each packet until the buffer is fully processed (empty)
without requiring any memmove.
Only if a packet is in-complete, the buffer content is moved
and the new data is read from the queue.
This way almost all memmove operations are eliminated.
Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Christof Schmitt <cs@samba.org>
Since 4.9.0, the log messages can be confusing if a required database
directory does not exist. Explicitly check for database directories,
logging a clear error and exiting if one is missing.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13696
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): Mon Dec 3 06:56:41 CET 2018 on sn-devel-144
Ideally this would just involve using "test -r". However, operating
system security features may mean that kernel stacks are not readable
even though they appear to be.
Instead, try reading that stack of a process on the test node. If
that succeeds then so should reading the stack of the "stuck" sleep
process in the test.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13684
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Tim Beale <timbeale@catalyst.net.nz>
Autobuild-User(master): Tim Beale <timbeale@samba.org>
Autobuild-Date(master): Thu Nov 15 08:15:32 CET 2018 on sn-devel-144
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Nov 8 11:03:11 CET 2018 on sn-devel-144
Explicitly background ctdbd instead.
This has the advantage of leaving stdin open. ctdbd can then be
enhanced to exit when stdin closes, allowing better cleanup in a test
environment.
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): Tue Nov 6 10:30:14 CET 2018 on sn-devel-144
popt uses an int in place of a bool, so declare an extra int and make
the conversion explicit.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
These aren't used by simple tests but they will be useful for
integrating ctdbd local daemons into other test suites and for
debugging.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
The etc-ctdb/ subdirectory containing the event script moves into the
top-level tests/ directory because the subdirectory is really now
owned by local_daemons.sh instead of simple/.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
This provides a separate script for handling local daemons. It can be
used for testing outside of the CTDB simple test suite. It is
installed as ctdb_local_daemons.
The logic is copied from ctdb/tests/simple/scripts/local_daemons.bash.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Always create an empty event script directory. If $TEST_SUBDIR is
unset then don't use it to look for etc-ctdb/.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Don't calculate this locally as _tools_dir. Add it to PATH
unconditionally - this may result in duplicate entries in PATH but the
resulting code is simpler.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Don't calculate this locally as _test_bin_dir. Just calculate
top_dir, source script_install_paths.sh and use
$CTDB_SCRIPT_TESTS_BINDIR.
Don't bother sanity checking if TEST_BIN_DIR is set. It will go away
soon.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
