hcq/MaxScale
hcq/MaxScale
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch '2.0.1' into develop

Browse Source
This commit is contained in:
Markus Makela 2016-10-03 15:21:04 +03:00
commit fbf5c331f1
16 changed files with 449 additions and 301 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
CMakeLists.txt
View File

@ -218,13 +218,12 @@ if(WITH_MAXSCALE_CNF AND (NOT TARGET_COMPONENT OR "core" STREQUAL "${TARGET_COMP
install_custom_file(server/maxscale.cnf.template ${MAXSCALE_CONFDIR} core)
endif()
# This should be moved to the qc_mysqlembedded directory
# install(PROGRAMS ${ERRMSG} DESTINATION ${MAXSCALE_VARDIR}/lib/maxscale)
install_file(${CMAKE_SOURCE_DIR}/COPYRIGHT core)
install_file(${CMAKE_SOURCE_DIR}/README core)
install_file(${CMAKE_SOURCE_DIR}/LICENSE.TXT core)
install_file(etc/lsyncd_example.conf core)
install_manual(Documentation/maxscale.1 1 core)
install_file(${CMAKE_SOURCE_DIR}/server/maxscale_binlogserver_template.cnf core)
# Install startup scripts and ldconfig files

7
Documentation/Changelog.md
View File

@ -5,14 +5,13 @@
* SSL can be used in the communication between MariaDB MaxScale and the backend servers.
* The number of allowed connections can explicitly be throttled.
* MariaDB MaxScale can continue serving read request even if the master has gone down.
* The security of MaxAdmin has been improved; you can only connect from the
same host MariaDB MaxScale is running on, and the Linux identity is used for
authorization.
* The security of MaxAdmin has been improved; Unix domain sockets can be used in the
communication with MariaDB MaxScale and the Linux identity can be used for authorization.
* MariaDB MaxScale can in real time make binlog events available as raw AVRO or
as JSON objects (beta level functionality).
For more details, please refer to:
* [MariaDB MaxScale 2.0.0 Release Notes](Release-Notes/MaxScale-2.0.0-Release-Notes.md)
* [MariaDB MaxScale 2.0.1 Release Notes](Release-Notes/MaxScale-2.0.1-Release-Notes.md)
## MariaDB MaxScale 1.4
* Authentication now allows table level resolution of grants. MaxScale service

1
Documentation/Getting-Started/Building-MaxScale-from-Source-Code.md
View File

@ -6,6 +6,7 @@ requirements are as follows:
* CMake version 2.8 or later (Packaging requires version 2.8.12 or later)
* GCC version 4.4.7 or later
* libaio
* libcurl
* OpenSSL
* Bison 2.7 or later
* Flex 2.5.35 or later

14
Documentation/Getting-Started/Install-MariaDB-MaxScale-Using-a-Tarball.md
View File

@ -2,6 +2,12 @@
MariaDB MaxScale is also made available as a tarball, which is named like `maxscale-x.y.z.OS.tar.gz` where `x.y.z` is the same as the corresponding version and `OS` identifies the operating system, e.g. `maxscale-2.0.1.centos.7.tar.gz`.
In order to use the tarball, the following libraries are required:
- libcurl
- libaio
- OpenSSL
The tarball has been built with the assumption that it will be installed in `/usr/local`. However, it is possible to install it in any directory, but in that case MariaDB MaxScale must be invoked with a flag.
## Installing as root in `/usr/local`
@ -18,7 +24,7 @@ The required steps are as follows:
$ cd maxscale
$ sudo chown -R maxscale var
Creating the symbolic link is necessary, since MariaDB MaxScale has been built with with the assumption that its base-directory, that is, the directory under which all its sub-directories are found, is `/usr/local/maxscale`.
Creating the symbolic link is necessary, since MariaDB MaxScale has been built with with the assumption that the plugin directory is `/usr/local/maxscale/lib/maxscale`.
The symbolic link also makes it easy to switch between different versions of MariaDB MaxScale that have been installed side by side in `/usr/local`; just make the symbolic link point to another installation.
@ -46,6 +52,12 @@ The `-d` flag causes maxscale _not_ to turn itself into a daemon, which is advis
If you want to place the configuration file somewhere else but in `/etc` you can invoke MariaDB MaxScale with the `--config` flag, for instance, `--config=/usr/local/maxscale/etc/maxscale.cnf`.
Note also that if you want to keep _everything_ under `/usr/local/maxscale` you can invoke MariaDB MaxScale using the flag `--basedir`.
$ sudo bin/maxscale --user=maxscale --basedir=/usr/local/maxscale -d
That will cause MariaDB MaxScale to look for its configuration file in `/usr/local/maxscale/etc` and to store all runtime files under `/usr/local/maxscale/var`.
## Installing in any Directory
Enter a directory where you have the right to create a subdirectory. Then do as follows.

7
Documentation/Release-Notes/MaxScale-2.0.1-Release-Notes.md
View File

@ -139,9 +139,8 @@ Packages can be downloaded [here](https://mariadb.com/resources/downloads).
## Source Code
The source code of MaxScale is tagged at GitHub with a tag, which is identical
with the version of MaxScale. For instance, the tag of version X.Y.Z of MaxScale
is maxscale-X.Y.Z. Further, *master* always refers to the latest released
non-beta version.
The source code of MaxScale is tagged at GitHub with a tag, which is derived
from the version of MaxScale. For instance, the tag of version `X.Y.Z` of MaxScale
is `maxscale-X.Y.Z`.
The source code is available [here](https://github.com/mariadb-corporation/MaxScale).

14
Documentation/Tutorials/Administration-Tutorial.md
View File

@ -6,13 +6,13 @@ Last updated 24th June 2015
The purpose of this tutorial is to introduce the MariaDB MaxScale Administrator to a few of the common administration tasks that need to be performed with MariaDB MaxScale. It is not intended as a reference to all the tasks that may be performed, more this is aimed as an introduction for administrators who are new to MariaDB MaxScale.
[Starting MariaDB MaxScale](#starting)
[Stopping MariaDB MaxScale](#stopping)
[Checking The Status Of The MariaDB MaxScale Services](#checking)
[Persistent Connections](#persistent)
[What Clients Are Connected To MariaDB MaxScale](#clients)
[Rotating the Log File](#rotating)
[Taking A Database Server Out Of Use](#outofuse)
- [Starting MariaDB MaxScale](#starting)
- [Stopping MariaDB MaxScale](#stopping)
- [Checking The Status Of The MariaDB MaxScale Services](#checking)
- [Persistent Connections](#persistent)
- [What Clients Are Connected To MariaDB MaxScale](#clients)
- [Rotating the Log File](#rotating)
- [Taking A Database Server Out Of Use](#outofuse)
<a name="starting"></a>
### Starting MariaDB MaxScale

217
Documentation/Tutorials/Galera-Cluster-Connection-Routing-Tutorial.md
View File

@ -16,152 +16,185 @@ MariaDB MaxScale configuration is held in an ini file that is located in the fil
A global, maxscale, section is included within every MariaDB MaxScale configuration file; this is used to set the values of various MariaDB MaxScale wide parameters, perhaps the most important of these is the number of threads that MariaDB MaxScale will use to execute the code that forwards requests and handles responses for clients.
[maxscale]
threads=4
```
[maxscale]
threads=4
```
Since we are using Galera Cluster and connection routing we want a single to which the client application can connect; MariaDB MaxScale will then route connections to this port onwards to the various nodes within the Galera Cluster. To achieve this within MariaDB MaxScale we need to define a service in the ini file. Create a section for each in your MariaDB MaxScale configuration file and set the type to service, the section name is the names of the service and should be meaningful to the administrator. Names may contain whitespace.
[Galera Service]
type=service
```
[Galera Service]
type=service
```
The router for this section the readconnroute module, also the service should be provided with the list of servers that will be part of the cluster. The server names given here are actually the names of server sections in the configuration file and not the physical hostnames or addresses of the servers.
[Galera Service]
type=service
router=readconnroute
servers=dbserv1, dbserv2, dbserv3
```
[Galera Service]
type=service
router=readconnroute
servers=dbserv1, dbserv2, dbserv3
```
In order to instruct the router to which servers it should route we must add router options to the service. The router options are compared to the status that the monitor collects from the servers and used to restrict the eligible set of servers to which that service may route. In our case we use the option that restricts us to servers that are fully functional members of the Galera cluster which are able to support SQL operations on the cluster. To achieve this we use the router option synced.
[Galera Service]
type=service
router=readconnroute
router_options=synced
servers=dbserv1, dbserv2, dbserv3
```
[Galera Service]
type=service
router=readconnroute
router_options=synced
servers=dbserv1, dbserv2, dbserv3
```
The final step in the service section is to add the username and password that will be used to populate the user data from the database cluster. There are two options for representing the password, either plain text or encrypted passwords may be used. In order to use encrypted passwords a set of keys must be generated that will be used by the encryption and decryption process. To generate the keys use the maxkeys command and pass the name of the secrets file in which the keys are stored.
% maxkeys /var/lib/maxscale/.secrets
%
```
% maxkeys /var/lib/maxscale/.secrets
%
```
Once the keys have been created the maxpasswd command can be used to generate the encrypted password.
% maxpasswd plainpassword
96F99AA1315BDC3604B006F427DD9484
%
```
% maxpasswd plainpassword
96F99AA1315BDC3604B006F427DD9484
%
```
The username and password, either encrypted or plain text, are stored in the service section using the user and passwd parameters.
[Galera Service]
type=service
router=readconnroute
router_options=synced
servers=dbserv1, dbserv2, dbserv3
user=maxscale
passwd=96F99AA1315BDC3604B006F427DD9484
```
[Galera Service]
type=service
router=readconnroute
router_options=synced
servers=dbserv1, dbserv2, dbserv3
user=maxscale
passwd=96F99AA1315BDC3604B006F427DD9484
```
This completes the definitions required by the service, however listening ports must be associated with a service in order to allow network connections. This is done by creating a series of listener sections. These sections again are named for the convenience of the administrator and should be of type listener with an entry labeled service which contains the name of the service to associate the listener with. Each service may have multiple listeners.
[Galera Listener]
type=listener
service=Galera Service
```
[Galera Listener]
type=listener
service=Galera Service
```
A listener must also define the protocol module it will use for the incoming network protocol, currently this should be the MySQLClient protocol for all database listeners. The listener may then supply a network port to listen on and/or a socket within the file system.
[Galera Listener]
type=listener
service=Galera Service
protocol=MySQLClient
port=4306
socket=/tmp/DB.Cluster
```
[Galera Listener]
type=listener
service=Galera Service
protocol=MySQLClient
port=4306
socket=/tmp/DB.Cluster
```
An address parameter may be given if the listener is required to bind to a particular network address when using hosts with multiple network addresses. The default behavior is to listen on all network interfaces.
The next stage is the configuration is to define the server information. This defines how to connect to each of the servers within the cluster, again a section is created for each server, with the type set to server, the network address and port to connect to and the protocol to use to connect to the server. Currently the protocol for all database connections in MySQLBackend.
[dbserv1]
type=server
address=192.168.2.1
port=3306
protocol=MySQLBackend
[dbserv2]
type=server
address=192.168.2.2
port=3306
protocol=MySQLBackend
[dbserv3]
type=server
address=192.168.2.3
port=3306
protocol=MySQLBackend
```
[dbserv1]
type=server
address=192.168.2.1
port=3306
protocol=MySQLBackend
[dbserv2]
type=server
address=192.168.2.2
port=3306
protocol=MySQLBackend
[dbserv3]
type=server
address=192.168.2.3
port=3306
protocol=MySQLBackend
```
In order for MariaDB MaxScale to monitor the servers using the correct monitoring mechanisms a section should be provided that defines the monitor to use and the servers to monitor. Once again a section is created with a symbolic name for the monitor, with the type set to monitor. Parameters are added for the module to use, the list of servers to monitor and the username and password to use when connecting to the the servers with the monitor.
[Galera Monitor]
type=monitor
module=galeramon
servers=dbserv1, dbserv2, dbserv3
user=maxscale
passwd=96F99AA1315BDC3604B006F427DD9484
```
[Galera Monitor]
type=monitor
module=galeramon
servers=dbserv1, dbserv2, dbserv3
user=maxscale
passwd=96F99AA1315BDC3604B006F427DD9484
```
As with the password definition in the server either plain text or encrypted passwords may be used.
The final stage in the configuration is to add the option service which is used by the maxadmin command to connect to MariaDB MaxScale for monitoring and administration purposes. This creates a service section and a listener section.
[CLI]
type=service
router=cli
[CLI Listener]
type=listener
service=CLI
protocol=maxscaled
socket=default
```
[CLI]
type=service
router=cli
[CLI Listener]
type=listener
service=CLI
protocol=maxscaled
socket=default
```
## Starting MariaDB MaxScale
Upon completion of the configuration process MariaDB MaxScale is ready to be started for the first time. This may either be done manually by running the maxscale command or via the service interface.
% maxscale
```
% maxscale
```
or
% service maxscale start
```
% service maxscale start
```
Check the error log in /var/log/maxscale to see if any errors are detected in the configuration file and to confirm MariaDB MaxScale has been started. Also the maxadmin command may be used to confirm that MariaDB MaxScale is running and the services, listeners etc have been correctly configured.
% maxadmin list services
```
% maxadmin list services
Services.
--------------------------+----------------------+--------+---------------
Service Name | Router Module | #Users | Total Sessions
--------------------------+----------------------+--------+---------------
Galera Service | readconnroute | 1 | 1
CLI | cli | 2 | 2
--------------------------+----------------------+--------+---------------
% maxadmin list servers
Servers.
-------------------+-----------------+-------+-------------+-------------------
Server | Address | Port | Connections | Status
-------------------+-----------------+-------+-------------+--------------------
dbserv1 | 192.168.2.1 | 3306 | 0 | Running, Synced, Master
dbserv2 | 192.168.2.2 | 3306 | 0 | Running, Synced, Slave
dbserv3 | 192.168.2.3 | 3306 | 0 | Running, Synced, Slave
-------------------+-----------------+-------+-------------+--------------------
Services.
--------------------------+----------------------+--------+---------------
Service Name | Router Module | #Users | Total Sessions
--------------------------+----------------------+--------+---------------
Galera Service | readconnroute | 1 | 1
CLI | cli | 2 | 2
--------------------------+----------------------+--------+---------------
% maxadmin list servers
Servers.
-------------------+-----------------+-------+-------------+-------------------
Server | Address | Port | Connections | Status
-------------------+-----------------+-------+-------------+--------------------
dbserv1 | 192.168.2.1 | 3306 | 0 | Running, Synced, Master
dbserv2 | 192.168.2.2 | 3306 | 0 | Running, Synced, Slave
dbserv3 | 192.168.2.3 | 3306 | 0 | Running, Synced, Slave
-------------------+-----------------+-------+-------------+--------------------
```
A Galera Cluster is a multi-master clustering technology, however the monitor is able to impose false notions of master and slave roles within a Galera Cluster in order to facilitate the use of Galera as if it were a standard MySQL Replication setup. This is merely an internal MariaDB MaxScale convenience and has no impact on the behavior of the cluster.
You can control which Galera node is the master server by using the _priority_ mechanism of the Galera Monitor module. For more details, read the [Galera Monitor](../Monitors/Galera-Monitor.md) documentation.
% maxadmin list listeners
```
% maxadmin list listeners
Listeners.
---------------------+--------------------+-----------------+-------+--------
Service Name | Protocol Module | Address | Port | State
---------------------+--------------------+-----------------+-------+--------
Galera Service | MySQLClient | * | 4306 | Running
CLI | maxscaled | localhost | 6603 | Running
---------------------+--------------------+-----------------+-------+--------
%
Listeners.
---------------------+--------------------+-----------------+-------+--------
Service Name | Protocol Module | Address | Port | State
---------------------+--------------------+-----------------+-------+--------
Galera Service | MySQLClient | * | 4306 | Running
CLI | maxscaled | localhost | 6603 | Running
---------------------+--------------------+-----------------+-------+--------
%
```
MariaDB MaxScale is now ready to start accepting client connections and routing them to the master or slaves within your cluster. Other configuration options are available that can alter the criteria used for routing, such as using weights to obtain unequal balancing operations. These options may be found in the MariaDB MaxScale Configuration Guide. More detail on the use of maxadmin can be found in the document ["MaxAdmin - The MariaDB MaxScale Administration & Monitoring Client Application"](../Reference/MaxAdmin.md).

2
cmake/package_rpm.cmake
View File

@ -34,3 +34,5 @@ if(EXTRA_PACKAGE_DEPENDENCIES)
endif()
message(STATUS "Generating RPM packages")
# Installing this prevents RPM from deleting the /var/lib/maxscale folder
install(DIRECTORY DESTINATION ${MAXSCALE_VARDIR}/lib/maxscale)

1
query_classifier/qc_mysqlembedded/CMakeLists.txt
View File

@ -15,4 +15,5 @@ if (BUILD_QC_MYSQLEMBEDDED)
set_target_properties(qc_mysqlembedded PROPERTIES LINK_FLAGS -Wl,--version-script=${CMAKE_CURRENT_SOURCE_DIR}/qc_mysqlembedded.map)
#set_target_properties(qc_mysqlembedded PROPERTIES LINK_FLAGS -Wl,-z,defs)
install_module(qc_mysqlembedded libmysqld-parser)
install(PROGRAMS ${ERRMSG} DESTINATION ${MAXSCALE_VARDIR}/lib/maxscale)
endif()

74
query_classifier/qc_sqlite/qc_sqlite.c
View File

@ -21,6 +21,7 @@
#include <platform.h>
#include <query_classifier.h>
#include <skygw_utils.h>
#include <modutil.h>
#include <maxscale/alloc.h>
#include "builtin_functions.h"
@ -130,6 +131,7 @@ static void info_finish(QC_SQLITE_INFO* info);
static void info_free(QC_SQLITE_INFO* info);
static QC_SQLITE_INFO* info_init(QC_SQLITE_INFO* info);
static bool is_submitted_query(const QC_SQLITE_INFO* info, const Parse* pParse);
static void log_invalid_data(GWBUF* query, const char* message);
static bool parse_query(GWBUF* query);
static void parse_query_string(const char* query, size_t len);
static bool query_is_parsed(GWBUF* query);
@ -542,6 +544,33 @@ static bool is_submitted_query(const QC_SQLITE_INFO* info, const Parse* pParse)
return rv;
}
/**
* Logs information about invalid data.
*
* @param query The query that could not be parsed.
* @param message What is being asked for.
*/
static void log_invalid_data(GWBUF* query, const char* message)
{
// At this point the query should be contiguous, but better safe than sorry.
if (GWBUF_LENGTH(query) >= MYSQL_HEADER_LEN + 1)
{
char *sql;
int length;
if (modutil_extract_SQL(query, &sql, &length))
{
if (length > GWBUF_LENGTH(query) - MYSQL_HEADER_LEN - 1)
{
length = GWBUF_LENGTH(query) - MYSQL_HEADER_LEN - 1;
}
MXS_INFO("qc_sqlite: Parsing the query failed, %s: %*s", message, length, sql);
}
}
}
static void append_affected_field(QC_SQLITE_INFO* info, const char* s)
{
size_t len = strlen(s);
@ -805,7 +834,7 @@ static void update_affected_fields_from_select(QC_SQLITE_INFO* info,
update_affected_fields_from_exprlist(info, pSelect->pEList, NULL);
}
if (pSelect->pWhere)
if (pSelect->pWhere)
{
info->has_clause = true;
update_affected_fields(info, 0, pSelect->pWhere, QC_TOKEN_MIDDLE, pSelect->pEList);
@ -839,6 +868,7 @@ static void update_names(QC_SQLITE_INFO* info, const char* zDatabase, const char
{
char* zCopy = MXS_STRDUP(zTable);
MXS_ABORT_IF_NULL(zCopy);
// TODO: Is this call really needed. Check also sqlite3Dequote.
exposed_sqlite3Dequote(zCopy);
enlarge_string_array(1, info->table_names_len, &info->table_names, &info->table_names_capacity);
@ -1136,10 +1166,8 @@ void mxs_sqlite3DeleteFrom(Parse* pParse, SrcList* pTabList, Expr* pWhere, SrcLi
update_affected_fields(info, 0, pWhere, QC_TOKEN_MIDDLE, 0);
}
//TODO: Figure out why the following statements, causes a crash
//TODO: long down the road.
//TODO: exposed_sqlite3SrcListDelete(pParse->db, pTabList);
//TODO: exposed_sqlite3ExprDelete(pParse->db, pWhere);
exposed_sqlite3ExprDelete(pParse->db, pWhere);
exposed_sqlite3SrcListDelete(pParse->db, pTabList);
exposed_sqlite3SrcListDelete(pParse->db, pUsing);
}
@ -2503,9 +2531,9 @@ static uint32_t qc_sqlite_get_type(GWBUF* query)
{
types = info->types;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report query type");
}
}
else
@ -2531,9 +2559,9 @@ static qc_query_op_t qc_sqlite_get_operation(GWBUF* query)
{
op = info->operation;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report query operation");
}
}
else
@ -2563,9 +2591,9 @@ static char* qc_sqlite_get_created_table_name(GWBUF* query)
MXS_ABORT_IF_NULL(created_table_name);
}
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report created tables");
}
}
else
@ -2591,9 +2619,9 @@ static bool qc_sqlite_is_drop_table_query(GWBUF* query)
{
is_drop_table = info->is_drop_table;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report whether query is drop table");
}
}
else
@ -2619,9 +2647,9 @@ static bool qc_sqlite_is_real_query(GWBUF* query)
{
is_real_query = info->is_real_query;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report whether query is a real query");
}
}
else
@ -2663,9 +2691,9 @@ static char** qc_sqlite_get_table_names(GWBUF* query, int* tblsize, bool fullnam
*tblsize = 0;
}
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report what tables are accessed");
}
}
else
@ -2702,9 +2730,9 @@ static bool qc_sqlite_query_has_clause(GWBUF* query)
{
has_clause = info->has_clause;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report whether the query has a where clause");
}
}
else
@ -2730,9 +2758,9 @@ static char* qc_sqlite_get_affected_fields(GWBUF* query)
{
affected_fields = info->affected_fields;
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report what fields are affected");
}
}
else
@ -2769,9 +2797,9 @@ static char** qc_sqlite_get_database_names(GWBUF* query, int* sizep)
database_names = copy_string_array(info->database_names, sizep);
}
}
else
else if (MXS_LOG_PRIORITY_IS_ENABLED(LOG_INFO))
{
MXS_ERROR("qc_sqlite: The query operation was not resolved. Response not valid.");
log_invalid_data(query, "cannot report what databases are accessed");
}
}
else

15
query_classifier/qc_sqlite/sqlite-src-3110100/src/util.c
View File

@ -219,6 +219,21 @@ int sqlite3Dequote(char *z){
}
for(i=1, j=0;; i++){
assert( z[i] );
#ifdef MAXSCALE
if ( z[i]==0 ){
// TODO: This is needed only because exposed_sqlite3Dequote() is called
// TODO: in qc_sqlite.c:update_names(). That call probably is not needed
// TODO: and should be removed, in which case this check could also be
// TODO: removed.
break;
}else if ( z[i]=='\\' ){
z[j++] = '\\';
if ( z[i+1]==quote || z[i+1]=='\\' ){
z[j++] = quote;
i++;
}
} else
#endif
if( z[i]==quote ){
if( z[i+1]==quote ){
z[j++] = quote;

9
server/core/housekeeper.c
View File

@ -276,11 +276,16 @@ hkthread(void *data)
ptr->nextdue = now + ptr->frequency;
taskfn = ptr->task;
taskdata = ptr->data;
// We need to copy type and name, in case hktask_remove is called from
// the callback. Otherwise we will access freed data.
HKTASK_TYPE type = ptr->type;
char name[strlen(ptr->name) + 1];
strcpy(name, ptr->name);
spinlock_release(&tasklock);
(*taskfn)(taskdata);
if (ptr->type == HK_ONESHOT)
if (type == HK_ONESHOT)
{
hktask_remove(ptr->name);
hktask_remove(name);
}
spinlock_acquire(&tasklock);
ptr = tasks;

6
server/modules/authenticator/cdc_plain_auth.c
View File

@ -265,6 +265,12 @@ cdc_auth_set_client_data(CDC_session *client_data,
uint8_t *client_auth_packet,
int client_auth_packet_size)
{
if (client_auth_packet_size % 2 != 0)
{
/** gw_hex2bin expects an even number of bytes */
client_auth_packet_size--;
}
int rval = CDC_STATE_AUTH_ERR;
int decoded_size = client_auth_packet_size / 2;
char decoded_buffer[decoded_size + 1]; // Extra for terminating null

2
server/modules/filter/regexfilter/regexfilter.c
View File

@ -484,7 +484,7 @@ regex_replace(const char *sql, pcre2_code *re, pcre2_match_data *match_data, con
size_t result_size;
/** This should never fail with rc == 0 because we used pcre2_match_data_create_from_pattern() */
if (pcre2_match(re, (PCRE2_SPTR) sql, PCRE2_ZERO_TERMINATED, 0, 0, match_data, NULL))
if (pcre2_match(re, (PCRE2_SPTR) sql, PCRE2_ZERO_TERMINATED, 0, 0, match_data, NULL) > 0)
{
result_size = strlen(sql) + strlen(replace);
result = MXS_MALLOC(result_size);

5
server/modules/routing/avro/avro_client.c
View File

@ -449,7 +449,10 @@ avro_client_process_command(AVRO_INSTANCE *router, AVRO_CLIENT *client, GWBUF *q
const char req_last_gtid[] = "QUERY-LAST-TRANSACTION";
const char req_gtid[] = "QUERY-TRANSACTION";
const size_t req_data_len = sizeof(req_data) - 1;
uint8_t *data = GWBUF_DATA(queue);
size_t buflen = gwbuf_length(queue);
uint8_t data[buflen + 1];
gwbuf_copy_data(queue, 0, buflen, data);
data[buflen] = '\0';
char *command_ptr = strstr((char *)data, req_data);
if (command_ptr != NULL)

371
server/modules/routing/readwritesplit/readwritesplit.c
View File

@ -667,6 +667,15 @@ static void closeSession(ROUTER *instance, void *router_session)
else
{
ss_dassert(!BREF_IS_WAITING_RESULT(bref));
/** This should never be true unless a backend reference is taken
* out of use before clearing the BREF_WAITING_RESULT state */
if (BREF_IS_WAITING_RESULT(bref))
{
MXS_WARNING("A closed backend was expecting a result, this should not be possible. "
"Decrementing active operation counter for this backend.");
bref_clear_state(bref, BREF_WAITING_RESULT);
}
}
}
/** Unlock */
@ -1044,160 +1053,6 @@ lock_failed:
return;
}
/**
* @brief Router error handling routine (API)
*
* Error Handler routine to resolve _backend_ failures. If it succeeds then
* there are enough operative backends available and connected. Otherwise it
* fails, and session is terminated.
*
* @param instance The router instance
* @param router_session The router session
* @param errmsgbuf The error message to reply
* @param backend_dcb The backend DCB
* @param action The action: ERRACT_NEW_CONNECTION or
* ERRACT_REPLY_CLIENT
* @param succp Result of action: true iff router can continue
*
* Even if succp == true connecting to new slave may have failed. succp is to
* tell whether router has enough master/slave connections to continue work.
*/
static void handleError(ROUTER *instance, void *router_session,
GWBUF *errmsgbuf, DCB *problem_dcb,
error_action_t action, bool *succp)
{
SESSION *session;
ROUTER_INSTANCE *inst = (ROUTER_INSTANCE *)instance;
ROUTER_CLIENT_SES *rses = (ROUTER_CLIENT_SES *)router_session;
CHK_DCB(problem_dcb);
/** Don't handle same error twice on same DCB */
if (problem_dcb->dcb_errhandle_called)
{
/** we optimistically assume that previous call succeed */
/*
* The return of true is potentially misleading, but appears to
* be safe with the code as it stands on 9 Sept 2015 - MNB
*/
*succp = true;
return;
}
else
{
problem_dcb->dcb_errhandle_called = true;
}
session = problem_dcb->session;
if (session == NULL || rses == NULL)
{
*succp = false;
}
else if (DCB_ROLE_CLIENT_HANDLER == problem_dcb->dcb_role)
{
*succp = false;
}
else
{
CHK_SESSION(session);
CHK_CLIENT_RSES(rses);
switch (action)
{
case ERRACT_NEW_CONNECTION:
{
if (!rses_begin_locked_router_action(rses))
{
*succp = false;
break;
}
/**
* If master has lost its Master status error can't be
* handled so that session could continue.
*/
if (rses->rses_master_ref && rses->rses_master_ref->bref_dcb == problem_dcb &&
!SERVER_IS_MASTER(rses->rses_master_ref->bref_backend->backend_server))
{
SERVER *srv = rses->rses_master_ref->bref_backend->backend_server;
backend_ref_t *bref;
bref = get_bref_from_dcb(rses, problem_dcb);
if (bref != NULL)
{
CHK_BACKEND_REF(bref);
if (BREF_IS_WAITING_RESULT(bref))
{
bref_clear_state(bref, BREF_WAITING_RESULT);
}
bref_clear_state(bref, BREF_IN_USE);
bref_set_state(bref, BREF_CLOSED);
}
else
{
MXS_ERROR("server %s:%d lost the "
"master status but could not locate the "
"corresponding backend ref.",
srv->name, srv->port);
}
if (rses->rses_config.rw_master_failure_mode != RW_FAIL_INSTANTLY &&
(bref == NULL || !BREF_IS_WAITING_RESULT(bref)))
{
/** The failure of a master is not considered a critical
* failure as partial functionality still remains. Reads
* are allowed as long as slave servers are available
* and writes will cause an error to be returned.
*
* If we were waiting for a response from the master, we
* can't be sure whether it was executed or not. In this
* case the safest thing to do is to close the client
* connection. */
*succp = true;
}
else
{
if (!srv->master_err_is_logged)
{
MXS_ERROR("server %s:%d lost the "
"master status. Readwritesplit "
"service can't locate the master. "
"Client sessions will be closed.",
srv->name, srv->port);
srv->master_err_is_logged = true;
}
*succp = false;
}
}
else
{
/**
* This is called in hope of getting replacement for
* failed slave(s). This call may free rses.
*/
*succp = handle_error_new_connection(inst, &rses, problem_dcb, errmsgbuf);
}
/* Free the lock if rses still exists */
if (rses)
{
rses_end_locked_router_action(rses);
}
break;
}
case ERRACT_REPLY_CLIENT:
{
handle_error_reply_client(session, rses, problem_dcb, errmsgbuf);
*succp = false; /*< no new backend servers were made available */
break;
}
default:
*succp = false;
break;
}
}
dcb_close(problem_dcb);
}
/**
* @brief Get router capabilities (API)
@ -1674,6 +1529,175 @@ static bool rwsplit_process_router_options(ROUTER_INSTANCE *router,
return success;
}
/**
* @brief Router error handling routine (API)
*
* Error Handler routine to resolve _backend_ failures. If it succeeds then
* there are enough operative backends available and connected. Otherwise it
* fails, and session is terminated.
*
* @param instance The router instance
* @param router_session The router session
* @param errmsgbuf The error message to reply
* @param backend_dcb The backend DCB
* @param action The action: ERRACT_NEW_CONNECTION or
* ERRACT_REPLY_CLIENT
* @param succp Result of action: true iff router can continue
*
* Even if succp == true connecting to new slave may have failed. succp is to
* tell whether router has enough master/slave connections to continue work.
*/
static void handleError(ROUTER *instance, void *router_session,
GWBUF *errmsgbuf, DCB *problem_dcb,
error_action_t action, bool *succp)
{
SESSION *session;
ROUTER_INSTANCE *inst = (ROUTER_INSTANCE *)instance;
ROUTER_CLIENT_SES *rses = (ROUTER_CLIENT_SES *)router_session;
CHK_DCB(problem_dcb);
/** Don't handle same error twice on same DCB */
if (problem_dcb->dcb_errhandle_called)
{
/** we optimistically assume that previous call succeed */
/*
* The return of true is potentially misleading, but appears to
* be safe with the code as it stands on 9 Sept 2015 - MNB
*/
*succp = true;
return;
}
else
{
problem_dcb->dcb_errhandle_called = true;
}
session = problem_dcb->session;
bool close_dcb = true;
if (session == NULL || rses == NULL)
{
*succp = false;
}
else if (DCB_ROLE_CLIENT_HANDLER == problem_dcb->dcb_role)
{
*succp = false;
}
else
{
CHK_SESSION(session);
CHK_CLIENT_RSES(rses);
switch (action)
{
case ERRACT_NEW_CONNECTION:
{
if (!rses_begin_locked_router_action(rses))
{
close_dcb = false; /* With the assumption that if the router session is closed,
* then so is the dcb.
*/
*succp = false;
break;
}
/**
* If master has lost its Master status error can't be
* handled so that session could continue.
*/
if (rses->rses_master_ref && rses->rses_master_ref->bref_dcb == problem_dcb &&
!SERVER_IS_MASTER(rses->rses_master_ref->bref_backend->backend_server))
{
SERVER *srv = rses->rses_master_ref->bref_backend->backend_server;
backend_ref_t *bref;
bref = get_bref_from_dcb(rses, problem_dcb);
if (bref != NULL)
{
CHK_BACKEND_REF(bref);
if (BREF_IS_WAITING_RESULT(bref))
{
bref_clear_state(bref, BREF_WAITING_RESULT);
}
bref_clear_state(bref, BREF_IN_USE);
bref_set_state(bref, BREF_CLOSED);
}
else
{
MXS_ERROR("server %s:%d lost the "
"master status but could not locate the "
"corresponding backend ref.",
srv->name, srv->port);
}
if (rses->rses_config.rw_master_failure_mode != RW_FAIL_INSTANTLY &&
(bref == NULL || !BREF_IS_WAITING_RESULT(bref)))
{
/** The failure of a master is not considered a critical
* failure as partial functionality still remains. Reads
* are allowed as long as slave servers are available
* and writes will cause an error to be returned.
*
* If we were waiting for a response from the master, we
* can't be sure whether it was executed or not. In this
* case the safest thing to do is to close the client
* connection. */
*succp = true;
}
else
{
if (!srv->master_err_is_logged)
{
MXS_ERROR("server %s:%d lost the "
"master status. Readwritesplit "
"service can't locate the master. "
"Client sessions will be closed.",
srv->name, srv->port);
srv->master_err_is_logged = true;
}
*succp = false;
}
}
else
{
/**
* This is called in hope of getting replacement for
* failed slave(s). This call may free rses.
*/
*succp = handle_error_new_connection(inst, &rses, problem_dcb, errmsgbuf);
}
dcb_close(problem_dcb);
close_dcb = false;
/* Free the lock if rses still exists */
if (rses)
{
rses_end_locked_router_action(rses);
}
break;
}
case ERRACT_REPLY_CLIENT:
{
handle_error_reply_client(session, rses, problem_dcb, errmsgbuf);
close_dcb = false;
*succp = false; /*< no new backend servers were made available */
break;
}
default:
ss_dassert(!true);
*succp = false;
break;
}
}
if (close_dcb)
{
dcb_close(problem_dcb);
}
}
/**
* @brief Handle an error reply for a client
*
@ -1694,18 +1718,39 @@ static void handle_error_reply_client(SESSION *ses, ROUTER_CLIENT_SES *rses,
client_dcb = ses->client_dcb;
spinlock_release(&ses->ses_lock);
/**
* If bref exists, mark it closed
*/
if ((bref = get_bref_from_dcb(rses, backend_dcb)) != NULL)
if (rses_begin_locked_router_action(rses))
{
CHK_BACKEND_REF(bref);
bref_clear_state(bref, BREF_IN_USE);
bref_set_state(bref, BREF_CLOSED);
if (BREF_IS_WAITING_RESULT(bref))
/**
* If bref exists, mark it closed
*/
if ((bref = get_bref_from_dcb(rses, backend_dcb)) != NULL)
{
bref_clear_state(bref, BREF_WAITING_RESULT);
CHK_BACKEND_REF(bref);
if (BREF_IS_IN_USE(bref))
{
bref_clear_state(bref, BREF_IN_USE);
bref_set_state(bref, BREF_CLOSED);
if (BREF_IS_WAITING_RESULT(bref))
{
bref_clear_state(bref, BREF_WAITING_RESULT);
}
dcb_close(backend_dcb);
}
}
else
{
// All dcbs should be associated with a backend reference.
ss_dassert(!true);
}
rses_end_locked_router_action(rses);
}
else
{
// The session has already been closed, hence the dcb has been
// closed as well.
}
if (sesstate == SESSION_STATE_ROUTER_READY)