log_manager.cc: fixed block buffer overflow. Queries are logged to trace log and long queries exceed the bufsize in length. Those were written beyond allocated memory areas.
mysql_client_server_protocol.h: added mysql_protocol_state_t to indicate whether MySQL protocol object is allocated, usable or freed. Freed means that memory allocations made by the protocol are freed. That is, command history etc.
mysql_backend.c: gw_backend_hangup and gw_error_backend_event used to call error handling function although session was already closing. Added check for session state.
mysql_client.c: route_by_statement lost some packets in case where query was sent in multiple packets.
mysql_common.c: gw_MySQL_get_next_packet failed in packet handling with route_by_statement. When multi-packet query was merged into one, packet type wasn't copied.
protocol_archive_srv_command and mysql_protocol_done didn't have proper locking in place which lead to occasional crashes.
Fix to #473, http://bugs.skysql.com/show_bug.cgi?id=473
Several memory issues. Read the code.
In general one-off hint is only added to gwbuf and freed in gwbuf_free. Stacked hint is copied to stack and to gwbuf. gwbuf is freed after routing but stacked hint is freed either in stop or when session is closed. All this applies to named hint as well except that in addition, it has one more copy in named hint struct which is emptied when session is closed.
session.c:session_setup_filters : fixed memory leak
hintparser.c: added token_free for HINT_TOKENs and fixed a few memory leaks.
mysql_client_server_protocol.h: added mysql_protocol_done which frees memory blocks pointed to by protocol members. Those can't be freed in dcb.c because dcb.c doesn't know about protocol's members.
mysql_backend.c:gw_backend_close: fixed memory leak
mysql_client.c: gw_client_close: fixed memory leak
mysql_common.c: added implementation of mysql_protocol_done
:protocol_archive_srv_command: tried to fix memory leak. Some memory is still leaking according to valgrind. Removed use of uninitialized local variable len.
readwritesplit.c: Fix to bug #469, http://bugs.skysql.com/show_bug.cgi?id=469, rwsplit counts every connection twice in master - counnection counts leak
execute_sescmd_in_backend: fixed a memory leak - visible only in DEBUG=Y build.
readwritesplit/test/makefile: added target for hints tests
session.c:session_setup_filters : fixed memory leak
hintparser.c: added token_free for HINT_TOKENs and fixed a few memory leaks.
mysql_client_server_protocol.h: added mysql_protocol_done which frees memory blocks pointed to by protocol members. Those can't be freed in dcb.c because dcb.c doesn't know about protocol's members.
mysql_backend.c:gw_backend_close: fixed memory leak
mysql_client.c: gw_client_close: fixed memory leak
mysql_common.c: added implementation of mysql_protocol_done
:protocol_archive_srv_command: tried to fix memory leak. Some memory is still leaking according to valgrind. Removed use of uninitialized local variable len.
readwritesplit.c:execute_sescmd_in_backend: fixed a memory leak - visible only in DEBUG=Y build.
session.c:session_setup_filters : fixed memory leak
hintparser.c: added token_free for HINT_TOKENs and fixed a few memory leaks.
mysql_client_server_protocol.h: added mysql_protocol_done which frees memory blocks pointed to by protocol members. Those can't be freed in dcb.c because dcb.c doesn't know about protocol's members.
mysql_backend.c:gw_backend_close: fixed memory leak
mysql_client.c: gw_client_close: fixed memory leak
mysql_common.c: added implementation of mysql_protocol_done
:protocol_archive_srv_command: tried to fix memory leak. Some memory is still leaking according to valgrind. Removed use of uninitialized local variable len.
readwritesplit.c:execute_sescmd_in_backend: fixed a memory leak - visible only in DEBUG=Y build.
Added DCB state check to backend error evend handler. If state isn't DCB_STATE_POLLING anymore, socket is closed with purpose or error occurred but it has already been handled.
Client doesn't hang anymore at least with the documented method.
Fixed bug in session command resul handling. In case where backend sent error message the session command cursor wasn't updated properly.
Added check to bref_clear_state, if bref's waiter counter would go negative, decrement to global operation counter is skipped.
All prepare commands are executed in every backend server currently connected.
All executes are routed to master. If stmt type was recorded in prepare phase in rwsplit router, read-only stmts could be routed to slaves.
COM_STMT_PREPARE gets arbitrary number of response packets from backend database. Since statements are prepared in every backend server and only one multi-packet response can be replied to client, redundant multi-packet responses are discarded. This is done in router. Mechanisms from session command handling are utilized with little changes: router must identify when response consists of multiple packets so that it knows to calculate the number of packets in response and that it is able to discard correct number of packets.
Information to the reply-handling router is provided by backend protocol, which includes a ordered list of commands of commands sent to protocol-owning backend server. A command is stored to protocol struct in mysql_backend.c:gw_MySQLWrite_backend if the statement buffer's type has GWBUF_TYPE_SINGLE_STMT set in mysql_client.c:route_by_statement. GWBUF_TYPE_SINGLE_STATEMENT indicates that there is single statement in the buffer, as opposite to Read Connection router, which accepts streaming input from client.
Conflicts:
server/core/session.c
server/modules/protocol/mysql_backend.c
server/modules/routing/readwritesplit/readwritesplit.c
Resolved.
Addition of user in topfilter report and general tidyup
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.
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.
Supports pulling binlogs from the master, caching to a local file and relaying to a slave.
Only tested with a single slave and a single MaxScale thread.
For each session command a sescmd property is created and added to the end of list of session commands. List is owned by router client session and it includes all session commands from the beginning of router session.
Router maintains an individual session command cursor for each backend. A cursor refers to the first session command which the corresponding backend server haven't yet responded yet.
When response message arrives at any time from backend, first it is checked whether backend's cursor is active. Cursor is active if a session command is routed to backend and the backend haven't responded to it yet. If cursor is active, next it is checked whether the current session command property has been responded by other backend. If both are true, then response message is sent to client as is. If session command response is routed to client already, the arrived response is discarded.
This implementation doesn't guarantee execution order between session commands and queries
if other backend server lags behind in session command execution.
In poll.c : moved processing of EPOLLERR and EPOLLHUP after processing of EPOLLIN and EPOLLOUT.
This ensures that COM_QUIT messages are read and routed forward before signals arrive (from local client/backend).
nection router stream is used and with read/write split router individual statements are passed to router.
Added new function to ROUTER_OBJECT : uint8_t (*getCapabilities)(ROUTER *instance, void* router_session); which is implemented in every route
r.
Added support for multi-statement packets in rwsplit router. In other words, if network packet includes multiple mysql statements, they are separated and passed to router one by one.
Multi-packet statements (those which exceeds network packet boundaries) are _not_ supported yet.
Added support for multi-statement packets.
This is an intermediate commit to save work. Code is not cleaned and there are debug prints and prototypes to be removed.
dcb.c:dcb_write accept also dcb state DCB_STATE_NOPOLLING since it only means that dcb has been removed from epoll set but it is still possible to write to it.
Bug #384http://bugs.skysql.com/show_bug.cgi?id=384
session.h:added new state for SESSION, SESSION_STATE_ROUTER_READY which follows SESSION_STATE_READY. The difference is that ROUTER_READY is set only after router session is successfully created while READY means that session still lacks router.
session.c:set SESSION_STATE_ROUTER_READY when router is created.
mysql_backend.c:gw_read_backend_event, added SESSION_STATE_ROUTER_READY check before router session is closed. Changed chec kso that it doesn't block in infinite loop (although it shouldn't be possible anyway).
mysql_backend.c:gw_error_backend_event, added similar check before session is closed.
If backend authentication fails, wait until session has created router_client_session before reading the session->router_session pointer value. There should only be one case where this situation occurs : when session is being initialized concurrently with authentication. Only consequence is to close router_client_session immediately after session initialization is completed.
Added new argument '-s' which takes additional argument composed of list of logfile identifiers. Logfiles listed with '-s' will be written on main memory instead of disk. In practice, the log file in question will be written in /dev/shm but corresponding symlink is added to log directory. In the case of name conflicts with log files and links, a differentiating sequence number is included in hte name of the file. This, however, is done only when existing file is not writable or is of different type (symlink <> file).
Added new logfile LOGFILE_DEBUG whose contents will be largerly what was included up to date in trace log.
Disabled feature which spreads writes to log files to others because of bug (#338) in the way block buffers are managed.
Changed log manager parameters to match with the current implementation. List of arguments:
"-h - help\n"
"-a <debug prefix> ............(\"skygw_debug\")\n"
"-b <debug suffix> ............(\".log\")\n"
"-c <trace prefix> ............(\"skygw_trace\")\n"
"-d <trace suffix> ............(\".log\")\n"
"-e <message prefix> ............(\"skygw_msg\")\n"
"-f <message suffix> ............(\".log\")\n"
"-g <error prefix> ............(\"skygw_err\")\n"
"-i <error suffix> ............(\".log\")\n"
"-j <log path> ............(\"/tmp\")\n"
"-s <shmem log file ids> ........(no default)\n";
dcb.c
dcb_add_to_zombieslist, add dcb to the front of zombies list instead of inserting to the end of it.
gateway.c
Renamed shutdown_gateway to shutdown_server (Bug #131)
Call skygw_logmanager_init so that trace and debug logs are written to shared memory.
poll.c
dcb.h
Removed some dead code and references to unneeded mutexes.
debugcmd.c
Added enable/disable log command for debug log.
skygw_utils.cc
skygw_file_init now takes optional symlink name as a second argument. Symlink is created to point to the file being created.
resolve_query_type, added GSYSVAR_FUNC type for functions that read system variables and can be executed in Maxscale instead of backend server.
dcb.c
dcb_read, if read returns value <= 0 and if error is EAGAIN/EWOULDBLOCK so there was nothing to read in sthe socket, that is not an error because some other thread may have read the data that was expected to be available.
mysql_backend.c
gw_read_backend_event, used dcb->authlock to ensure that dcb's protocol state is read and modified serially. This removes issues with false authentication failures which may happen when two threads modify and read the state without any control.
mysql_client.c
removed dead code.
readconnroute.c, readwritesplit.c
removed invalid assert which assumed that spinlock can not have larger value than one when itis locked.
poll.c
Removed mutex from epoll_wait.
Removed read and write mutexes from poll_waitevents.
session.c
If session_alloc fails, instead of calling directly free(session), call session_free, which decreases refcounter and only frees session when there are no references left.
Added session_unlink_dcb function which removes link from session and optionally sets the dcb->session pointer to NULL.
readconnection.h, readwritesplit.h
Added check fields to ROUTER_CLIENT_SES strct as well as lock, version number (not used yet) and closed flag.
mysql_backend.c
gw_read_backend_event: if backend_protocol->state was set to MYSQL_AUTH_RECV, function returned, which was unnecessary. If mysql state became MYSQL_AUTH_FAILED, router client session was closed. Removed unnecessary NULL checks because rsession is not allowed to be NULL. Similarly, removed other NULL checks and replaced them with asserts checking that router client session is not NULL at any phase.
mysql_client.c
Removed unused code blocks. Polished log commands. Replaced router client sessions NULL checks with asserts.
mysql_common.c
mysql_send_custom_error: if called with dcb == NULL, return.
readconnroute.c
Replaced malloc with calloc. Added functions rses_begin_router_action and rses_exit_router_action. If router client session is not closed, they take a lock and release it, respectively. Those functions are used for protecting all operations which modify the contents of router client session struct.
readwritesplit.c
Identical changes than in readconnroute.c
skygw_debug.h
Added check number and - macro for ROUTER_CLIENT_SES, and added COM_QUIT to STRPACKETTYPE.
