Added depth level 0 for each cluster node, this way the algorithm for
root master selection will the same as in mysql replication:
the root master is server at lowest replication depth with MASTER bit
set
Here for Galera we assume all the servers are at the same level, that’s
0
Rwsplit handles ERRACT_NEW_CONNECTION by clearing backend reference, removing callbacks and associating backend reference with new backend server. If it succeeds and the router session can continue, handleError returns true. Otherwise false. When ever false is returned it means that session must be closed.
Rwsplit now tolerates backend failures in a way that it searches new backends when monitor, backend, or client operation fails due to backend failure.
Added flags to those backend references which have sent something to backend which causes the backend to send results or reply back. Didn't add removal of the flag since there's currently no way to tell whether response from backend contains anything else than session command reply - which aren't counted when BREF_WAITING_RESULT is set and cleared.
dcb.h: define ERRHAND temporarily since changes are still behind that macro
Defined two error handling actions in router.h: ERRACT_NEW_COMMECTION and ERRACT_REPLY_CLIENT.
Failed database is logged at expanding frequence to error and to message log due changes in mysql_mon.c. Added two new members in MONITOR_SERVERS: mon_err_count, and mon_prev_status so that each backend can be treated individually.
Error handling: if mysql_backend.c:dcb_read fails, router's handleError is called instead of closing session.
If mysql_client.c:SESSION_ROUTE_QUERY fails router's handleError is called instead of sending error to client.
readwritesplit.c:select_connect_backend_servers is modified so that in can be called during active router session. When called, it attempts to find one master and maximum number of configured slaves in correct state if necessary.
When handleError needs to replace failed unit it now calls select_connect_backend_servers.
Added a lot of logging to error, trace and message logs which should help the user to handle errors which can't be automatically resolved, like attempt to use nonexisting database.
monitors used mysql_ping without MYSQL_OPT_READ_TIMEOUT which caused read to block. Fixed in mysql and galera monitor.
Added log writing per each status change of each server and repeatedly if server is not running.
Removed SERVER_IS_JOINED checks from rwsplit router.
Maintenance mode prevents a server from having new connections made
to it and stops MaxScale from monitoring a server. It is designed
to allow a server to be taken out of an active cluster for maintenance.
It prevents new connections and allowsg existing connections to
drain away prior to performaing maintenance on the server.
be extracted from the modules.
This gives a way to verify the API that the module provides as well
as the version of that API. The hope is that this will make it possible
for MaxScale to detect out of date plugins and either adapt to use them
or reject loading them.
Also added the ability to set a release state on a per module basis.
This allows for production ready and non-production ready plugins to
be identified.
Removed MARIADB_SRC_PATH
Added MYSQL_ROOT - root directory where MariaDB headers are installed
Added MYSQL_HEADERS - -I notation for three header directories
Added EMBEDDED_LIB - for embedded library directory
Added ERRMSG - for errmsg.sys file
log_manager.cc
Little fixes
log_manager/makefile
Updated header include directive
makefile.inc
Removed DEBUGGER and BACKGR as unnecessary
Added LIB for embedded library file name
query_classifier/makefile
Updated header include directives and embedded library directory
query_classifier/query_classifier.cc
clean up
server/core/Makefile
Updated header include directives and embedded library directory
Added libaio and install of errmsg.sys
server/core/dcb.c
clean up
server/core/gateway.c
Added --language and --skip-innodb to mysql_library_init command-line arguments list
clean up
server/core/load_utils.c
clean up
server/modules/monitor/Makefile
server/modules/routing/readwritesplit/Makefile
Updated header include directives and embedded library directory
utils/skygw_debug.h
http://bugs.skysql.com/show_bug.cgi?id=369
utils/skygw_types.h
utils/skygw_utils.cc
clean up
State update for filewriter was missing and that caused Maxscale to fail if opening of any log file failed.
dcb.c:
Added EAGAIN and EWOULDBLOCK handling to dcb_read. If dcb_close is called for freshly created dcb, dcb is only freed.
gateway.c:
Added file_write_footer and write_footer of which the latter is called at exit time. It simply draws a line to screen.
gw_utils.c:
Some macros for helping comparison between gw_read_gwbuff and dcb_read, which overlap.
poll.c:
Some macros to help enable/disable mutexing in poll_waitevents
service.c:
Check return value of listen and session_alloc and behave accordingly.
mysql_client.c:
If ioctl returned successfully with b==0 it earlier caused closing the client and backend dcbs. Since that doesn't reliably indicate that client has closed socket on its side, Maxscale doesn't close its sockets either.
mysql_common.c:
In gw_receive_backend_auth, if dcb_read returns n==0, it is not considered as an error anymore. The implemented behavior is not yet complete and correct. Result should be successful but the protocol state shouldn't change to MYSQL_IDLE before backend return is received.
In gw_send_authentication_to_backend protocol state was always set to MYSQL_AUTH_RECV even if gw_rwite had failed. Now, return value is read and state is set in caller's context basen on the return value.
skygw_utils.cc:
Removed ss_dassert from skyge_file_init because it prevented from returning meaningful error meassage to the client.:
