ldb: Bump version to 2.0.12

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10730: lib ldb: Check if ldb_lock_backend_callback called twice

Prevent use after free issues if ldb_lock_backend_callback is called
twice, usually due to ldb_module_done being called twice. This can happen if a
module ignores the return value from function a function that calls
ldb_module_done as part of it's error handling.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10730: s4 dsdb vlv_pagination: Prevent repeat call of ldb_module_done

Check the return code from vlv_results, if it is not LDB_SUCCESS
ldb_module_done has already been called, and SHOULD NOT be called again.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10730: s4 dsdb paged_results: Prevent repeat call of ldb_module_done

Check the return code from paged_results, if it is not LDB_SUCCESS
ldb_module_done has already been called, and SHOULD NOT be called again.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10730: dsdb: Ban the combination of paged_results and VLV

This (two different paging controls) makes no sense and fails against
Windows Server 1709.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: dsdb: Fix crash when vlv and paged_results are combined

The GUID is not returned in the DN for some reason in this (to be banned)
combination.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: selftest: Add test to show that VLV and paged_results are incompatible

As tested against Windows Server 1709

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: vlv: Another workaround for mixing ASQ and VLV

This is essentially an alternative patch, but without the correct
behaviour.  Instead this just avoids a segfault.

Included in case we have something simialr again in
another module.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: selftest: Add test to confirm VLV interaction with ASQ

Tested against Windows 1709.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: vlv: Do not re-ASQ search the results of an ASQ search with VLV

This is a silly combination, but at least try and keep the results sensible
and avoid a double-dereference.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10730: vlv: Use strcmp(), not strncmp() checking the NULL terminated control OIDs

The end result is the same, as sizeof() includes the trailing NUL, but this
avoids having to think about that.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14364

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

VERSION: Bump version up to 4.11.11...

and re-enable GIT_SNAPSHOT.

Signed-off-by: Karolin Seeger <kseeger@samba.org>
(cherry picked from commit 08a51254198537395e9a6ea7a98fd627a491bf15)

VERSION: Disable GIT_SNAPSHOT for the 4.11.10 release.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

WHATSNEW: Add release notes for Samba 4.11.10.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

ldb_ldap: fix off-by-one increment in lldb_add_msg_attr

Fix regression introduced by commit ce2bf5c72b6423fff680b3d6a9042103a6cdda55

lldb_add_msg_attr() calls ldb_msg_add_empty() which, in turn, calls
calls _ldb_msg_add_el() which already increments msg->num_elements by one.

As a result, msg->num_elements is bigger than the actual number of
elements and any iteration over elements would step over elements array
boundary.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14413
Signed-off-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri Jun 19 08:35:33 UTC 2020 on sn-devel-184

(cherry picked from commit 990a0fc4a0481aed817fad7575d8df453fbe7af9)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Wed Jun 24 12:39:28 UTC 2020 on sn-devel-184

lib/ldb: add unit test for ldb_ldap internal code

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14413
Signed-off-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 36bd6edd8a603f3aa34aff81c43ef26efd3ad4cf)

s3: libsmbclient: Finish unifing bad iconv behavior across CORE NT1 SMB2 protocols.

On bad name conversion, exit the directory listing with an error, but leave the
connection intact. We were already checking for finfo->name == NULL here,
but were ignoring it and not reporting an error.

Remove the knownfail.d/bad_iconv file as we now
behave the same across CORE/NT1/SMB2.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14374

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue May 12 21:32:44 UTC 2020 on sn-devel-184

(cherry picked from commit 393da520e43bd3a28feb231bcd9fd5308a3daa4a)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Thu Jun  4 11:47:26 UTC 2020 on sn-devel-184

s3: libsmb: In SMB2 return NT_STATUS_INVALID_NETWORK_RESPONSE if name conversion ended up with a NULL filename.

Can happen if namelen == 0.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14374

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 753115a8d19f6ac8cd28305748fc6d888679dccc)

s3: libsmb: In SMB1 old protocol - return NT_STATUS_INVALID_NETWORK_RESPONSE if name conversion ended up with a NULL filename.

Can happen if namelen == 0.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14374

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit b10de0bb64fe022e6b066584013dfb0bdf2ade96)

s3: selftest: Add test_smbclient_iconv.sh to check client behavior on bad name conversion.

SMB2 and NT1 fail this, CORE already returns NT_STATUS_INVALID_NETWORK_RESPONSE
on bad conversion.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14374

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(back-ported from commit e016671d34c24c4768df774425ec743b88e30015)

s3: selftest: Add share definition [bad_iconv] in fileserver.

Creates a utf8 valid filename within that is invalid in CP850.
Useful to test smbclient list directory character set conversions.

https://bugzilla.samba.org/show_bug.cgi?id=14374

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(back-ported from commit a9651d6bc2b6dea8adc859ce21c2431253868887)

winbindd: Fix a use-after-free when winbind clients exit

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14382

Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>

Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Tue May 19 10:45:06 UTC 2020 on sn-devel-184

(cherry picked from commit 68380ebaa60c64311cc1081f700d571abbf69f4f)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Fri May 22 14:40:24 UTC 2020 on sn-devel-184

s3: lib: Paranoia around use of snprintf copying into a fixed-size buffer from a getenv() pointer.

Post checks for overflow/error.

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon May 18 23:42:57 UTC 2020 on sn-devel-184

(cherry picked from commit dd1f750293ef4361455a5d5b63fc7a89495715b7)

s3:gencache: Allow to open gencache as read-only

This allows client tools to access the cache for ready-only operations
as a normal user.

Example:
    net ads status

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14370

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Fri May 15 14:40:32 UTC 2020 on sn-devel-184

(cherry picked from commit 04f0c45475de383a0be4ca355ab9aa7784e61c27)

lib:util: Add test for path_expand_tilde()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14370

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(backported from commit a15bd5493b696c66c6803d8ca65bc13f1cfcdf0a)

lib:util: Add path_expand_tilde()

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14370

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 15457254be0ab1235c327bd305dfeee19b2ea7a1)

docs-xml: Fix usernames in pam_winbind manpages

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14358

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Wed Apr 29 09:44:55 UTC 2020 on sn-devel-184

(cherry picked from commit 3abd92d7824e803f1ff53425088ebee30b58894b)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Thu May 14 15:42:16 UTC 2020 on sn-devel-184

libsmb: Protect cli_oem_change_password() from rprcnt<2

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14366
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): Tue May  5 17:12:04 UTC 2020 on sn-devel-184

(cherry picked from commit f80c97cb8da64f3cd9904e2e1fd43c29b691166d)

libsmb: Protect cli_RNetServerEnum against rprcnt<6

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14366
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit ce8b70df7bd63e96723b8e8dc864f1690f5fad7b)

libsmb: Protect cli_RNetShareEnum() against rprcnt<6

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14366
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit 4a9fe4efefa67d6f24efcbe29722a43fc4859fdc)

libsmb: Fix indentation in cli_RNetShareEnum()

Also remove a level of indentation with a "goto done;"

Best review with "git show -b", almost no code change

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14366
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
(cherry picked from commit ae91d67a247424d4ddc89230f52365558d6ff402)

vfs_shadow_copy2: implement case canonicalisation in shadow_copy2_get_real_filename()

unix_convert() can't do this for us in snapdirseverywhere mode, so we do it
ourselves.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14350

Signed-off-by: Ralph Boehme <slow@samba.org>
(Similar to commit a3d1ac2a597e2441d6855db566306298ae5db99b)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Wed May 13 14:44:57 UTC 2020 on sn-devel-184

s3/lib: add is_gmt_token()

This is not present in master as master has been converted to use struct
smb_filename.twrp instead of @GMT string tokens as part of the path.

Signed-off-by: Ralph Boehme <slow@samba.org>

smbd: make get_real_filename_full_scan() public

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14350

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(backported from commit aa5f19ddf1dec1ac4386441929bca94727f30ee6)
[Conflicts: source3/smbd/proto.h: more functions are missing in 4.12]

CI: add two tests for shadow_copy2 VFS module

Note that the test "fetch a previous version of a regular file via non-canonical
basepath" doesn't fail by "luck" because it runs into the "creating file"
optimisation in unix_convert().

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14350

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 6557777c86d72a185b3fe4061a8b5791fd748924)

ldb: Bump version to 2.0.11

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14330

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>

lib ldb: lmdb init var before calling mdb_reader_check

Initilalise "stale" to zero before passing a pointer to it to
mdb_reader_check.

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Tue Apr  7 12:29:00 UTC 2020 on sn-devel-184

(cherry picked from commit d1f4002b914efb31aa34a59e7c93d80f3174727c)

lib ldb: lmdb clear stale readers on write txn start

In use process failures and Bind9 shut downs leave stale entries in the
lmdb reader table.  This can result in lmdb filling it's database file, as
the free list can not be reclaimed due to the stale reader.

In this fix we call mdb_reader_check at the start of each transaction,
to free any stale readers.  As the default maximum number of readers is
127, this should not impact on performance to any great extent.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14330

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Mar 31 01:26:07 UTC 2020 on sn-devel-184

(cherry picked from commit 89041a6d18a1d091ea713e6986cac5ca66c2b481)

ldb tests: Confirm lmdb free list handling

Add cmocka tests to confirm lmdb's handling of the free list.

As a result of lmdb's MVCC (Multiversion Concurrency Control) long
running read transactions or stale readers (read transactions where the
process exited without ending the transaction) can cause the database to
run out of space.

Items in the free list are only reused when they would not be visible in
a read transaction.  So long running read transactions prevent entries
in the free list being reused, and the database can run out of space.

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
(cherry picked from commit fc13304d1575ad6bc6e04cdb3eedf46d3c3678c7)

VERSION: Bump version up to 4.11.10...

and re-enable GIT_SNAPSHOT.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

VERSION: Disable GIT_SNAPSHOT for the 4.11.9 release.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

WHATSNEW: Add release notes for Samba 4.11.9.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

s3: pass DCE RPC handle type to create_policy_hnd

Various RPC services expect policy handles of a specific type.

s3 RPC server did not allow to create policy handles with a specific
type while actually requiring that policy handle type itself in some
places.

Make sure we are able to specify the policy on-wire handle type when
creating the policy handle. The changes follow s4 DCE RPC server
implementation.

The original logic to always set on-wire handle type to 0 can be tracked
down to commit fdeea341ed1bae670382e45eb731db1b5838ad21 when we didn't
really know about differences in on-wire handle types.

All but LSA trusted domain RPC calls do not check the on-wire handle
type in s3 RPC server.

Fixes trusted domain operations when Samba RPC client attempts to call
s3 RPC server to perform lsa_lsaRSetForestTrustInformation in FreeIPA.
This fix is a pre-requisite for FreeIPA-FreeIPA forest trust.

Signed-off-by: Alexander Bokovoy <ab@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Apr 28 22:55:29 UTC 2020 on sn-devel-184

(cherry picked from commit c7a4578d06427a82ead287f0c5248c1a54cc9336)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Mon May  4 11:35:51 UTC 2020 on sn-devel-184

ctdb: Fix a memleak

Bug: https://bugzilla.samba.org/show_bug.cgi?id=14348
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): Fri Apr 17 08:32:35 UTC 2020 on sn-devel-184

(cherry picked from commit ad4b53f2d95de3d3609f3a23d96602f3dd516705)

libsmb: Don't try to find posix stat info in SMBC_getatr()

This wrongly used "frame" instead of "fname", which can never have
worked. A first attempt to fix in 51551e0d53fa6 caused a few followup
patches in an attempt to clean up the test failures 51551e0d53fa6
introduced. They were reverted after a few discussions. So rather than
changing behaviour, just remove the code that introduced the valgrind
error again.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Mar 20 05:06:07 UTC 2020 on sn-devel-184

(cherry picked from commit 39c910fd9cba3caf7414274b678b9eee33d7e20b)

VERSION: Bump version up to 4.11.9.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

Merge tag 'samba-4.11.8' into v4-11-test

samba: tag release samba-4.11.8

VERSION: Disable GIT_SNAPSHOT for the 4.11.8 release.

o CVE-2020-10700: Use-after-free in Samba AD DC LDAP Server with ASQ
o CVE-2020-10704: LDAP Denial of Service (stack overflow) in Samba AD DC

Signed-off-by: Karolin Seeger <kseeger@samba.org>

WHATSNEW: Add release notes for Samba 4.11.8.

Signed-off-by: Karolin Seeger <kseeger@samba.org>

CVE-2020-10704 libcli ldap: Check search request lengths.

Check the search request lengths against the limits passed to
ldap_decode.

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: libcli ldap_message: Add search size limits to ldap_decode

Add search request size limits to ldap_decode calls.

The ldap server uses the smb.conf variable
"ldap max search request size" which defaults to 250Kb.
For cldap the limit is hard coded as 4096.

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: S4 ldap server: Limit request sizes

Check the size of authenticated and anonymous ldap requests and reject
them if they exceed the limits in smb.conf

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: smb.conf: Add max ldap request sizes

Add two new smb.conf parameters to control the maximum permitted ldap
request size.

Adds:
   ldap max anonymous request size       default 250Kb
   ldap max authenticated request size   default 16Mb

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: ldapserver tests: Limit search request sizes

Add tests to ensure that overly long (> 256000 bytes) LDAP search
requests are rejected.

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: lib util asn1: Check parse tree depth

Check the current depth of the parse tree and reject the input if the
depth exceeds that passed to asn1_init

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: libcli ldap: test recursion depth in ldap_decode_filter_tree

Add tests to check that ASN.1 ldap requests with deeply nested elements
are rejected.  Previously there was no check on the on the depth of
nesting and excessive nesting could cause a stack overflow.

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10704: lib util asn1: Add ASN.1 max tree depth

Add maximum parse tree depth to the call to asn1_init, which will be
used to limit the depth of the ASN.1 parse tree.

Credit to OSS-Fuzz

REF: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=20454
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14334

Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

CVE-2020-10700: ldb: Bump version up to 2.0.10.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14331

Signed-off-by: Karolin Seeger <kseeger@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

CVE-2020-10700: dsdb: Do not permit the ASQ control for the GUID search in paged_results

ASQ is a very strange control and a BASE search can return multiple results
that are NOT the requested DN, but the DNs pointed to by it!

Thanks to Andrei Popa <andrei.popa@next-gen.ro> for finding,
reporting and working with us to diagnose this issue!

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14331

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10700: ldb: Always use ldb_next_request() in ASQ module

We want to keep going down the module stack, and not start from the top again.

ASQ is above the ACL modules, but below paged_results and we do not wish to
re-trigger that work.

Thanks to Andrei Popa <andrei.popa@next-gen.ro> for finding,
reporting and working with us to diagnose this issue!

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14331

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

CVE-2020-10700: dsdb: Add test for ASQ and ASQ in combination with paged_results

Thanks to Andrei Popa <andrei.popa@next-gen.ro> for finding,
reporting and working with us to diagnose this issue!

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14331

Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Gary Lockyer <gary@catalyst.net.nz>

VERSION: Bump version up to 4.11.8...

and re-enable GIT_SNAPSHOT.

Signed-off-by: Karolin Seeger <kseeger@samba.org>
(cherry picked from commit 46e19f9f40258855d84e747bcb02019262e72057)

s3/librpc/crypto: Fix double free with unresolved credential cache

We free gse_ctx->k5ctx but then free it again in the
talloc dtor. This patch just lets the talloc dtor handle
things and removes the extra krb5_free_context

Failed to resolve credential cache 'DIR:/run/user/1000/krb5cc'! (No credentials cache found)
==30762== Invalid read of size 8
==30762==    at 0x108100F4: k5_os_free_context (in /usr/lib64/libkrb5.so.3.3)
==30762==    by 0x107EA661: krb5_free_context (in /usr/lib64/libkrb5.so.3.3)
==30762==    by 0x7945D2E: gse_context_destructor (gse.c:84)
==30762==    by 0x645FB49: _tc_free_internal (talloc.c:1157)
==30762==    by 0x645FEC5: _talloc_free_internal (talloc.c:1247)
==30762==    by 0x646118D: _talloc_free (talloc.c:1789)
==30762==    by 0x79462E4: gse_context_init (gse.c:241)
==30762==    by 0x794636E: gse_init_client (gse.c:268)
==30762==    by 0x7947602: gensec_gse_client_start (gse.c:786)
==30762==    by 0xBC87A3A: gensec_start_mech (gensec_start.c:743)
==30762==    by 0xBC87BC6: gensec_start_mech_by_ops (gensec_start.c:774)
==30762==    by 0xBC8167F: gensec_spnego_client_negTokenInit_step (spnego.c:633)
==30762==  Address 0x17259928 is 40 bytes inside a block of size 496 free'd
==30762==    at 0x4C2F50B: free (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==30762==    by 0x79462CA: gse_context_init (gse.c:238)
==30762==    by 0x794636E: gse_init_client (gse.c:268)
==30762==    by 0x7947602: gensec_gse_client_start (gse.c:786)
==30762==    by 0xBC87A3A: gensec_start_mech (gensec_start.c:743)
==30762==    by 0xBC87BC6: gensec_start_mech_by_ops (gensec_start.c:774)
==30762==    by 0xBC8167F: gensec_spnego_client_negTokenInit_step (spnego.c:633)
==30762==    by 0xBC813E2: gensec_spnego_client_negTokenInit_start (spnego.c:537)
==30762==    by 0xBC84084: gensec_spnego_update_pre (spnego.c:1943)
==30762==    by 0xBC83AE5: gensec_spnego_update_send (spnego.c:1741)
==30762==    by 0xBC85622: gensec_update_send (gensec.c:449)
==30762==    by 0x551BFD0: cli_session_setup_gensec_local_next (cliconnect.c:997)
==30762==  Block was alloc'd at
==30762==    at 0x4C306B5: calloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==30762==    by 0x107EA7AE: krb5_init_context_profile (in /usr/lib64/libkrb5.so.3.3)
==30762==    by 0xB853215: smb_krb5_init_context_common (krb5_samba.c:3597)
==30762==    by 0x794615B: gse_context_init (gse.c:209)
==30762==    by 0x794636E: gse_init_client (gse.c:268)
==30762==    by 0x7947602: gensec_gse_client_start (gse.c:786)
==30762==    by 0xBC87A3A: gensec_start_mech (gensec_start.c:743)
==30762==    by 0xBC87BC6: gensec_start_mech_by_ops (gensec_start.c:774)
==30762==    by 0xBC8167F: gensec_spnego_client_negTokenInit_step (spnego.c:633)
==30762==    by 0xBC813E2: gensec_spnego_client_negTokenInit_start (spnego.c:537)
==30762==    by 0xBC84084: gensec_spnego_update_pre (spnego.c:1943)
==30762==    by 0xBC83AE5: gensec_spnego_update_send (spnego.c:1741)
==30762==

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14344
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Tue Apr 14 22:55:51 UTC 2020 on sn-devel-184

(cherry picked from commit 34f8ab774d1484b0e60dbdec8ad2a1607ad92122)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Wed Apr 15 14:34:51 UTC 2020 on sn-devel-184

testprogs: Add 'net ads join createupn' test also verifying the keytab

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14336

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Mon Apr  6 19:09:53 UTC 2020 on sn-devel-184

(cherry picked from commit c4be195da2845be4f64e47883e3c911dedd90e48)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Tue Apr  7 10:51:07 UTC 2020 on sn-devel-184

s3:libads: Fix ads_get_upn()

This adds the userPrincipalName to ads_find_machine_acct() which
fetches the data for us.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14336

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
(cherry picked from commit ec69752cb963ae850568d3f4905d2941e485627e)

smbd: let delayed update handler also update on-disk timestamps

Let delayed update handler also update on-disk timestamps by calling
trigger_write_time_update_immediate().

trigger_write_time_update_immediate() sets fsp->update_write_time_on_close to
false which prevents updating the write-time on close if there was ever only one
write to the file.

Besides resetting fsp->update_write_time_on_close and setting the on-disk timestamps
trigger_write_time_update_immediate() takes the same steps as the removed code.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14320

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Thu Mar 19 03:05:40 UTC 2020 on sn-devel-184

(back ported from commit 81c1a14e3271aeed7ed4fe6311171b19ba963555)

smbd: let mark_file_modified() always call trigger_write_time_update()

Preperatory change: the next commit will reset fsp->update_write_time_on_close
in the event handler, so this change ensures it gets set again for any
subsequent write.

This will NOT always result in a write-time update because
trigger_write_time_update() has its own only-once logic using the internal
variable fsp->update_write_time_triggered.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14320

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 53de2da7acfc24513082190502d93306c12b7434)

smbd: remove stat call from mark_file_modified()

This stat dates back to d03453864ab1bc5fd3b4a3abaf96176a006c102b where the call
to trigger_write_time_update() had been to the file IO codepath. It was present
there for other reasons: to setup the write-cache based on the file's size.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14320

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit 2c19d27113036d607850f370bb9afd62856d671e)

smbd: avoid double chdir() in chdir_current_service()

Since 8e81090789e4cc3ba9e5aa792d4e52971909c894 we're doing chdir() twice, first
into conn->connectpath, then into conn->origpath.

Before commit 8e81090789e4cc3ba9e5aa792d4e52971909c894 if
chdir(conn->connectpath) succeeded, we wouldn't do the second chdir().

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14256
RN: smbd does a chdir() twice per request

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Feb  6 11:44:07 UTC 2020 on sn-devel-184

(backported from commit f705629a171c1411131164f3adff36175154c093)

smbd: flush pending writetime update when setting timestamps file

Cf the explanations in the previous commits.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14150

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(back-ported from commit 7b90fe69a865ae8648b6548eabbcf2fa8237ebd8)

smbd: flush pending writetime update when flushing file

Cf the explanations in the previous commit.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14150

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(back-ported from commit d99d5bf2c6d0a818ef2f3920e0c93fac38761c36)

smbd: always flush pending write time update when setting filesize

We need to flush a pending write time update even when we're setting the
filesize to current filesize.

Note that we're already doing it this way in the relevant places listed my
dochelp@MS in

https://lists.samba.org/archive/cifs-protocol/2019-December/003364.html

  Cleanup (= Close)
  SetBasicInfo
  SetAllocationInfo
  SetEndOfFileInfo
  SetValidDataLengthInfo
  Flush
  FSCTL_SET_ENCRYPTION
  FSCTL_OFFLOAD_WRITE

Cleanup (= Close):

  Already implemented by update_write_time_on_close() and friends.

SetBasicInfo:

  Currently doesn't flush pending updates. Fixed by a subsequent commit.

SetAllocationInfo:

  smb_set_file_allocation_info() when setting a file's allocation size.

SetEndOfFileInfo:

  Currently doesn't flush pending updates. Fixed by a subsequent commit.

SetValidDataLengthInfo:

  Not implemented, returns NT_STATUS_NOT_SUPPORTED which seems wrong btw, as
  SetValidDataLengthInfo IS listed in MS-SMB2 2.2.39.

Flush:

  Currently doesn't flush pending updates. Fixed by subsequent commit.

FSCTL_SET_ENCRYPTION:

  Windows 2016 doesn't flush a pending writetime update, verified with a
  smbtorture test.

FSCTL_OFFLOAD_WRITE:

  NT_STATUS_NOT_IMPLEMENTED

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14150

Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(back-ported from commit 79d7d6b9d01b8547f16b74a62926d0b471f18c39)

nsswitch: fix use-after-free causing segfault in _pam_delete_cred

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14327

Guenther

Signed-off-by: Guenther Deschner <gd@samba.org>
Reviewed-by: Alexander Bokovoy <ab@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Mon Mar 30 13:01:20 UTC 2020 on sn-devel-184

(cherry picked from commit 047b0d8ab534c7a10a8572fd9f21e2456fd30710)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Tue Mar 31 16:14:54 UTC 2020 on sn-devel-184

ctdb-recoverd: Avoid dereferencing NULL rec->nodemap

Inside the nested event loop in ctdb_ctrl_getnodemap(), various
asynchronous handlers may dereference rec->nodemap, which will be
NULL.

One example is lost_reclock_handler(), which causes rec->nodemap to be
unconditionally dereferenced in list_of_nodes() via this call chain:

  list_of_nodes()
  list_of_active_nodes()
  set_recovery_mode()
  force_election()
  lost_reclock_handler()

Instead of attempting to trace all of the cases, just avoid leaving
rec->nodemap set to NULL.  Attempting to use an old value is generally
harmless, especially since it will be the same as the new value in
most cases.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14324

Reported-by: Volker Lendecke <vl@samba.org>
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): Tue Mar 24 01:22:45 UTC 2020 on sn-devel-184

(cherry picked from commit 716f52f68b248ae7cfd66479b3fc678c4a0d8b38)

Autobuild-User(v4-11-test): Karolin Seeger <kseeger@samba.org>
Autobuild-Date(v4-11-test): Mon Mar 30 11:22:26 UTC 2020 on sn-devel-184

ctdb-daemon: Don't allow attach from recovery if recovery is not active

Neither the recovery daemon nor the recovery helper should attach
databases outside of the recovery process.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 147afe77de372ddb9c180228d6fe1b04cca4610f)

ctdb-daemon: Remove more unused old client database functions

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 052f1bdb9cf78f53f584edd32f81ae8e01e8e86e)

ctdb-recovery: Remove old code for creating missing databases

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 3a66d181b6f6199fca362fcb0aa06513645b589d)

ctdb-recovery: Create database on nodes where it is missing

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 76a8174279f42486b36cc41d5831d4e6613f172e)

ctdb-recovery: Fetch database name from all nodes where it is attached

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit e6e63f8fb8194634135bf34cda18f6cc8ff69a7c)

ctdb-recovery: Pass db structure for each database recovery

Instead of db_id and db_flags.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 1bdfeb3fdc06947a607957ab3d114f97bad5d7d7)

ctdb-recovery: GET_DBMAP from all nodes

This builds a complete list of databases across the cluster so it can
be used to create databases on the nodes where they are missing.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit c6f74e590f602e2ed38fe293468770a5e669aefa)

ctdb-recovery: Replace use of ctdb_dbid_map with local db_list

This will be used to build a merged list of databases from all nodes,
allowing the recovery helper to create missing databases.

It would be possible to also include the db_name field in this
structure but that would cause a lot of churn.  This field is used
locally in the recovery of each database so can continue to live in
the relevant state structure(s).

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 4c0b9c36050a0ed8a180d4ac1853224089528e8e)

ctdb-daemon: Respect CTDB_CTRL_FLAG_ATTACH_RECOVERY when attaching databases

This is currently only set by the recovery daemon when it attaches
missing databases, so there is no obvious behaviour change.  However,
attaching missing databases can now be moved to the recovery helper as
long as it sets this flag.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 7e5a8a4884ea87bb985fe0e2b65ff130fc2ba8aa)

ctdb-recovery: Use CTDB_CTRL_FLAG_ATTACH_RECOVERY to attach during recovery

ctdb_ctrl_createdb() is only called by the recovery daemon, so this is
a safe, temporary change.  This is temporary because
ctdb_ctrl_createdb(), create_missing_remote_databases() and
create_missing_local_databases() will all go away soon.

Note that this doesn't cause a change in behaviour.  The main daemon
will still only defer attaches from non-recoverd processes during
recovery.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 98e3d0db2bc5f33217e26fab1dfb4bb91eae534f)

ctdb-protocol: Add control flag CTDB_CTRL_FLAG_ATTACH_RECOVERY

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 17ed0425904a98624284d351ab7617b3e02c0f7b)

ctdb-daemon: Remove unused old client database functions

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit fc23cd1b9cdd1d70067491614b16e616291e8ff2)

ctdb-daemon: Fix database attach deferral logic

Commit 3cc230b5eeca749ab68d19cfda969f72c269f1f6 says:

  Dont allow clients to connect to databases untile we are well past
  and through the initial recovery phase

It is unclear what this commit was attempting to do.  The commit
message implies that more attaches should be deferred but the code
change adds a conjunction that causes less attaches to be deferred.
In particular, no attaches will be deferred after startup is complete.
This seems wrong.

To implement what seems to be stated in the commit message an "or"
needs to be used so that non-recovery daemon attaches are deferred
either when in recovery or before startup is complete.  Making this
change highlights that attaches need to be allowed during the
"startup" event because this is when smbd is started.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit c6c89495fbe9b6f238d10a538eccc92b937a69de)

ctdb-recovery: Refactor banning a node into separate computation

If a node is marked for banning, confirm that it's not become inactive
during the recovery.  If yes, then don't ban the node.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 1c56d6413f86cc15ebac232f39ef1e2a53ae4297)

ctdb-recovery: Don't trust nodemap obtained from local node

It's possible to have a node stopped, but recovery master not yet
updated flags on the local ctdb daemon when recovery is started.  So do
not trust the list of active nodes obtained from the local node.  Query
the connected nodes to calculate the list of active nodes.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit c6a0ff1bed0265e44fd6135d16bfc41919fe5bf5)

ctdb-recovery: Consolidate node state

This avoids passing multiple arguments to async computation.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 6e2f8756f1bce4dfc7fdc435e082f400116e29ec)

ctdb-recovery: Fetched vnnmap is never used, so don't fetch it

New vnnmap is constructed using the information from all the connected
nodes.  So there is no need to fetch the vnnmap from recovery master.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=14294

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 072ff4d12b8f34766120ddae888d772e97bca491)

ctdb-client: Factor out function client_db_tdb()

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 439ef65d290357e513103530183091a9a6fed197)

ctdb-daemon: Implement DB_VACUUM control

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 41a41d5f3e2b8e16e25221e14939dc5962997ac7)

ctdb-vacuum: Only schedule next vacuum event if vacuuuming is scheduled

At the moment vacuuming is always scheduled.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit d462d64cdf001fd5d1cbf2a109df62e087ad0c49)

ctdb-daemon: Factor out code to create vacuuming child

This changes the behaviour for some failures from exiting to simply
attempting to schedule the next run.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 13cedaf0195c6bda3a3820aedb1ee65f36dfc23e)

ctdb-vacuum: Simplify recording of in-progress vacuuming child

There can only be one, so simplify the logic.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 5539edfdbe69d1d5f084d06753cce8ed6e524999)

ctdb-protocol: Add marshalling for control DB_VACUUM

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit 496204feb0e2b6eb2f3d9a74e45596a3e74ad9b1)

ctdb-protocol: Add marshalling for struct ctdb_db_vacuum

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit a896486b62bbcf9915727ba7bfc768fb5383f0c7)

ctdb-protocol: Add new control CTDB_CONTROL_DB_VACUUM

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
(cherry picked from commit b314835341e4028f0770fa7f9a37d2d21448ddfd)

ctdb-vacuum: Avoid processing any more packets

All the vacuum operations if required have an event loop to ensure
completion of pending operations.  Once all the steps are complete,
there is no reason to process any more packets.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit d0cc9edc05b6218a8e20a0a8009cbb9918ff4d02)