The listener iterator hides the details of the listener iteration behind a
small set of functions.
The following for-loop demonstrates the main use-case for the iterator:
LISTENER_ITERATOR iter;
for (SERV_LISTENER *listener = listener_iterator_init(service, &iter);
listener; listener = listener_iterator_next(&iter))
{
/** Do something with the listener */
}
As the listeners are mostly iterated to either check a fact about them or
print their information, the functions cater to that use-case. For this
reason, they should always be allocated off the stack.
The service port list is now iterated in a safe and lock-free manner. This
makes the handling of service ports somewhat simpler since once an item
has been added to the list it will never be removed. It also removes the
need to lock the service when checking whether a service listens on a port
which caused potential deadlocks.
The listeners under the /services/:service/listeners collection are now
fully JSON API compliant resources.
The listeners could also be exposed as a /listeners collection to easily
group all listener type resources in one place. This approach does has
some semantical and practical problems, namely the fact that each listener
has a many-to-one relationship with its service and listeners by
themselves can't exist alone.
The JSON API specification suggests that the API returns the 403 Forbidden
error when the user does an invalid request. The 400 Bad Request isn't the
ideal error for cases where the syntax is correct but the action being
performed is wrong.
Using the JSON Pointer syntax specified in RFC 6901
(https://tools.ietf.org/html/rfc6901) allows for a convenient way to
access values deep in a JSON object.
The JSON objects that are created from the various core MaxScale objects
share a lot of common code. Moving this into a separate files removes the
redundant code.
The sessions resource now follows the JSON API specification:
http://jsonapi.org/
This makes the API relatively easier to use as the specification and the
client libraries to consume this data exist.
The atomic compare-and-swap can be used to implement lock-free
structures. The planned use for this is to remove some of the locking done
in the services when listeners are being manipulated.
Various small changes to part2, as suggested by comments and otherwise.
Mostly renaming, working logic should not change.
Exception: session id changed to 64bit in the container and associated
functions. Another commit will change it to 64bit in the session itself.
MySQL sessions are added to a hasmap when created, removed when closed.
MYSQL_COM_PROCESS_KILL is now detected, the thread_id is read and the kill
command sent to all worker threads to find the correct session. If found, a
fake hangup even is created for the client dcb.
As is, this function is of little use since the client could just disconnect
itself instead. Later on, additional commands of this nature will be added.
The modules, their types and default values are exposed via the
/maxscale/modules resource. Currently, only a list of resources can be
exposed as the externally exposed module object (MXS_MODULE) does not have
the name and type information in it.
The `user`, `password`, `version_string` and `weightby` values should be
allocated as a part of the service structure. This allows them to be
modified at runtime without having to worry about memory allocation
problems.
Although this removes the problem of reallocation, it still does not make
the updating of the strings thread-safe. This can cause invalid values to
be read from the service strings.
The /sessions/ resource was not implemented due to changes in the core
polling mechanics. With the new worker thread messaging system, sessions
can be listed in a safe manner.
The service, filter and monitor resources now have a "parameters" value
which contains a set of all configuration parameters for that object. This
set contains both standard and non-standard parameters.
Also fixed a mistake in the constant name definitions for the monitor
parameters "events" and "script".
The service resource now exposes the parameters that were given to
it. This allows general service parameters to be changed at runtime
through the REST API.
The alteration function is only called when the value of a parameter
changes. This removes some of the redundant logging that was caused by
calling the alteration function with the same values.
A self link to the resource itself provides a convenient way for the
client to request a resource, modify it and call the self link to update
it. This removes some of the burden on the client to keep track of the
resource links by placing these in the resource itself.
The server can now be modified with a PUT request of a modified server
resource. The server resource was reorganized to have the parameters as a
separate entity from the other more general entities of the resource.
The PUT/POST functions return a more appropriate error message when no
request body is provided.
Moved some of the constant names used in server.cc into the config.h
header.
Now all relevant and working core configuration parameters have constants
for their names. This removes some of the risk that comes with the
repetition of the same parameter name in more than place.
The parameter names for monitors and servers now use a set of constant
names. This removes some of the errors caused by spelling mistakes when
the same parameter name is repeated in multiple places.
The service, filter and listener parameters should also be converted to
constants. This allows for a consistent user experience.
The relationships from servers to services and monitors and filters to
services were not implemented. Now each server lists the services and
monitors that use it and each filter lists the services that use the
filter.
This enables the creation of a server and linking of that server to
services and monitors in one atomic operation.
When a resource has a relation to another resource, it should be expressed
as a working link to the resource. By passing the hostname of the server
to the functions, we are able to generate working relation links.
If a resource can be accessed from more than one place, it should be a
link. The previous change that duplicated the content for nested resources
wasn't a very clean solution. With links, the content is easier to browse
using a normal browser.
The admin interface now supports Basic Access authentication. This is not
a secure method of authentication and it should not be used without
unencrypted connections.
Made the admin interface port, authentication, username and password
configurable.
All routers except the binlogrouter now fully implement the JSON
diagnostic entry point. The binlogrouter needs to be handled in a separate
commit as it produces a large amount of diagnostic output.
The modules that implement a diagnostics entry point now return a JSON
type object. This removes the need to format data inside the modules.
The module implementations of these are not yet complete which means that
MaxScale will fail to compile.
All whitespace in object names is now converted to a compacted format with
whitespace replaced with hyphens. If a conversion takes place, a warning
is logged.
Converted some of the tests into C++ as they directly include .cc files.
Monitors can now be printed in JSON format. The REST API resource
`/monitors` accepts GET requests and returns a JSON representation of the
monitors as a response.
The functions don't modify the filters when the JSON is generated. This
means that the parameters can be const.
Minor formatting fixes to service sources.
Both the filters and services can be queried via the REST API now that
these resources can be expressed in JSON format.
As with the other resources, these directly call the functions that
generate the data. This will be done via the inter-thread messaging system
once it's in place.
The REST API now prints individual sessions and servers. It also lists all
servers if no specific server is given.
The functions directly call the printing functions when they should be
using the inter-thread messaging system. When the messaging system is
ready, these functions should be updated.
The admin thread now uses blocking IO. This is not optimal but it
simplifies the code by some amount.
Fixed option processing removing one extra character from key name.
Use correct member variable when checking for the option map end.
