| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
|
|
|
|
Jeremy.
|
|
Don't only rely on dptr == NULL.
I stumbled over this one when rewriting some of the dbwrap_ctdb code.
Michael
|
|
regdb_fetch_keys_internal()
Prevents segfaults in some situations.
(For a non existent or empty record, we sometimes rely on the fetch operation
to return dsize==0 and sometimes we rely on dptr==NULL.)
Michael
|
|
for the case that another local process has started a transaction
bewteen releasing the transaction_lock record and starting the
transaction.
Michael
|
|
in db_ctdb_transaction_fetch_start() for error conditions when re-fetching
the transaction_lock record inside the transaction
Michael
|
|
node.
In ctdb_transaction_commit(), when the trans2_commit control fails, there
is a race condition in the 1 second sleep between the local transaction_cancel
and the call to ctdb_replay_transaction(): The database is not locked, and
neither is the transaction_lock record. So another client can start and possibly
complete a new transaction in this gap, but only on the same node: The locking
of the transaction_lock record on a different node which involves migration of
the record to the other node has been disabled by introduction of the
transaction_active flag on the db which closes precisely this gap from the start
of the commit until the call to TRANS2_FINISH or TRANS2_ERROR.
But this mechanism does not cover the case where a process on the same node
tries to start a transaction: There is no obstacle to locking the transaction_lock
record because the record does not need to be migrated.
This commit closes this race condition in ctdb_transaction_fetch_start()
by using the new ctdb_ctrl_transaction_active() call to ask the local
ctdb daemon whether it has a transaction running on the database.
If so, the check is repeated until the running transaction is done.
This does introduce an additional call to the local ctdbd when starting
transactions, but it does close the (hopefully) last race condition.
Michael
|
|
Michael
|
|
CTDB_CONTROL_TRANS2_COMMIT
Michael
|
|
There are two races in concurrent transactions on a single node.
One in starting a transaction and one with replay during commit.
This commit closes the first race by storing the client pid in the
transaction-lock record and comparing the stored pid against its own
pid after releasing the lock and refetching the record inside the
transaction.
Michael
|
|
Michael
|
|
Michael
|
|
This fetches a record from the db and splits out the ctdb header.
Michael
|
|
and use it in db_ctdb_store() and db_ctdb_transaction_store().
Michael
|
|
Michael
|
|
directory with a lot of files.
Jeremy.
|
|
|
|
Without this, we panic in wins_srv_is_dead() and fail to start nmbd with
wins support.
Volker, please check.
Guenther
|
|
It does not cost much and can help a lot when debugging
|
|
|
|
Rather than try and remove the records in the LDB files, make the
provision remove the whole file. This also removes the need to try
and carry forward the old ldb filenames.
Andrew Bartlett
|
|
This splits the code, while keeping the original behaviour. The
provision.py file had become just too long.
Andrew Bartlett
|
|
|
|
|
|
|
|
of the Schema
|
|
Rather than treat the LDAP backend as a special case, treat all
backends the same, with different callbacks.
Andrew Bartlett
|
|
|
|
|
|
|
|
|
|
|
|
While this logic (avoiding to prefix a non-filename with a path) is
important in the code this was copied from (private_dir()), none of
the callers of this function need it.
Andrew Bartlett
|
|
This allows us to change the escaping function without breaking
existing installs. The new escaping function (used for new databases)
is RFC1738 URI encoding, except for the trivial cases without special
characters.
The new databases are also placed in a subdirectory, sam.ldb.d per an
earlier suggestion by metze.
Andrew Bartlett
|
|
Andrew Bartlett
|
|
This is intended to replace our rfc1738_unescape(), and give us an
rfc1738_escape implementation (and hopefully is better tested and more
secure).
Andrew Bartlett
|
|
This avoids trouble when the secrets.ldb is updated with ldbedit but
an smb.conf is not specified.
Andrew Bartlett
|
|
since the initial provision
|
|
|
|
|
|
|
|
|
|
If I read SMB_SUBSYSTEM right then the 2nd argument needs to be the file where
the static_init_rpc (in this case) is defined. This seems to have moved from
server.c to process.c.
Jelmer, please check!
Volker
|
|
|
|
This problem became visible after adding the picky -z defs linker option: On
Solaris libreplace had unresolved symbols, which showed up in the libtalloc
build. PAM_WINBIND_EXTRA_LIBS and WINBIND_NSS_EXTRA_LIBS had been workarounds
to make things work at two placeѕ. These variables have been obsoleted now.
This patch introduces LIBREPLACE_LIBS which contans the linker flags needed for
linking anything using libreplace.
|
|
|
|
Fix bug #4250. Thanks to TAKAHASHI Motonobu <monyo@samba.gr.jp> for reporting!
Karolin
|
|
Fix bug #4252. Thanks to TAKAHASHI Motonobu <monyo@samba.gr.jp> for reporting!
Karolin
|
|
|
