When no users are loaded from backends, instead of counting it as a failure the service is started and the next time a client connects an attempt to load the users from the backends is made.
service.h:Added the 'serviceStartFailed' function which tries to start services with successfully started routers but no successfully started listeners.
mysql_mon.c:Added a call to ServiceStartFailed when servers come available.
readwritesplit.c:route_session_write failed if the last backend on all backends list was not in use. THe situation where not all backends are used by routing session is normal especially if max_slave_connections is not set to 100%. Thus session commands may have failed if user was bit unlucky.
Changed the logic so that the function fails (and session is closed) if routing fails to any such backend which is in use in the session.
Added server references to services instead of using a raw pointers to server instances.
Changed all references to service->database to service->dbref.
tee.c:routeQuery cloned partial MySQL packets to be routed to child router. Problems were due to the use of gwbuf_clone and modutil_MySQL_query which clone/examine only the first buffer from the buffer list which composes GWBUF.
modutil.c: Added function modutil_MySQL_query_len which calculates MySQL packet length and the missing bytecount
tee.c: Added details about the freeing of orphaned sessions.
session.: Added missing cases of session state into the session_state function in debug builds.
tee.c: Created a static, module-wide list of orphaned branch sessions which holds the branch sessions that can't be freed at the
time the parent session is being freed. This list is processed every time a a tee filter session is freeing its memory.
dcb.c:dcb_hashtable_stats:division by zero
hashtable.c:hashtable_get_stats: NULL-pointer reference
service.c:serviceStartPort:set service->users NULL to avoid referring to freed memory
users.c:dcb_usersPrintf: NULL-pointer reference
debugcmd.c:convert_arg: changed return value to 1 in case of error, 0 (==NULL) is valid but it indicates that there are no users loaded.
execute_cmd: fixed command handling
Tee filter now receives two client replies, one from the parent service and
one from the child service. These replies are checked before cloning any more queries
into the child service. If the reply isn't sent fast enough, the child session is closed.
Added clientReply and setUpstream entry points
Tee filter now waits for the branch session to be replied to and if the reply isn't received in five seconds it closes the branch session.
dcb.h:
Added the DCBF_REPLIED flag that is currently only set in dcb_null_write
dcb.c:
The DCBF_REPLIED flag is now set in dcb_null_write
http://bugs.skysql.com/show_bug.cgi?id=665http://bugs.skysql.com/show_bug.cgi?id=664http://bugs.skysql.com/show_bug.cgi?id=649
dcb.c:dcb_final_free: (665):set dcb->session->client pointer to NULL so that it won't be read anymore and other threads won't try to close it.
dcb_final_free:(664):don't free dcb->data, it is either freed in session_alloc if session creation fails or in session_free only.
session.c:if session creation fails, free dcb->data and remove links between client DCB and session.
mysql_backend.c:(665):gw_backend_close:check that session->client isn't NULL and that client DCB's state is still polling before calling dcb_close for it.
mysql_client.c:gw_mysql_do_authentication:if anything fails, and session_alloc won't be called, free dcb->data.
mysql_common.c:gw_send_authentication_to_backend:if session is already closing then return with error.
readwritesplit.h:Removed invalid macros which assumed that ROUTER_CLIENT_SES->rses_backend_ref always pointed to valid and used backend reference and thus included potential risk of NULL-pointer refernce.
mysql_backend.c and mysql_client.c:avoid executing CHK_PROTOCOL(p) after original DCB has been released the memory.
readwritesplit.c:Replaced RSES_CLEINT_DCB macro with a function which returns client DCB for a given router client session.
