- Hard TTL; the maximum time a value will be used from the cache.
- Soft TLL; the time after which the cache value should be updated
from the server.
So as not to unnecessarily fetch the same value multiple times, when
the soft TTL has been reached, the value will be updated for the first
client, while all other clients will use the stale value until it has
become updated.
With different soft and hard TTLs there is a definite upper bound for
how old a value can be used.
In order for the LRU storage to correctly track the state in the
real storage, we always need to know stale-state of a returned
(or not returned) value.
The return type is now not an a pure enumeration, but a 16-bit
enumeration plus a 16-bit bitmask that can be used for conveying
more information.
- CamelCase for all C++ class and template names
- snake_case for all variables, including member variables and
member functions
- Static member functions start with a capital letter
- Member variables prefixed with m_
- Static member variables prefixed with s_
- Null terminated character string variables prefixed with z
- Pointers prefixed with p
- Smart-pointers prefixed with s
- Where a z, p, or s prefix is present, the following letter is
capitalized, i.e., pEntry and not pentry.
Now all storage configuration values are provided in a single object.
That way it'll be easier to provide more if necessary and also makes
it straightforward to fetch the configuration, which makes it possible,
for instance, to adapt tests according to the way the storage has been
configured.
16MBytes packets are now handled in MULTI result sets as well.
MAXROWS_DISCARDING_RESPONSE state has been removed.
Two new flags control large packets and result discarding
The server weighting parameters couldn't be altered online and they
weren't serialized to disk. Only servers that are created online will have
their weighting parameters persisted to disk.
Any changes to servers that aren't created online need to be manually
added to the configuration file in order for them to take effect after the
restart.
After upgrades, it is usually useful to see which version of MaxScale is
running. By adding a command, we can see the actual version of the running
daemon process instead of the version of the current binary.
The routeQuery() in readconnroute now checks for maintenance mode. If
the server is in maintenance, the session is closed, since this router
has no backend swapping capability.
- Move docs from buffer.c to buffer.h
- Arguments made const-correct.
gwbuf_length
gwbuf_count
gwbuf_copy_data
- Boolean return values changes from int to bool.
gwbuf_add_property
- Meaningless returnvalue changed from int to void.
gwbuf_add_hint
- Unused function removed.
gwbuf_clone_transform
gwbuf_trim
gwbuf_clone cloned only the first buffer of a chain of buffers,
which never can be the desired outcome, while gwbuf_clone_all
cloned all buffers.
Now, gwbuf_clone behaves the way gwbuf_clone_all used to behave
and gwbuf_clone_all has been removed.
Converting the README into Markdown format makes it a lot easier to
comprehent. Also cleaned up the formatting.
The 2.0 branch had a broken Travis configuration. Fixed it and changed
links to point to 2.0 branch.
The luafilter exposes two of the main functions provided by the query
classifier API; the type and operation classification.
The functions can be used by the Lua script with minimal overhead as the
current query being executed is stored only as a pointer. The functions
should only be called inside the `routeQuery` entry point of a Lua script.
The functionality to disable setting of master and slave status values to
Galera nodes was broken by the change to the monitoring algorithm.
The disable_master_role_setting option removed the master node which would
trigger the clearing of the replication status values from the node. A
more direct, and arguably better, way is to check that this option is
disabled before processing the status values for the nodes.
The DECIMAL value type is now properly handled in Avrorouter. It is
processed into an Avro double value when before it was ignored and
replaced with a zero integer.
Previously, server status changes from MaxAdmin would be set immediately
as long as the server lock could be acquired. This meant that it might take
several seconds until the next monitor pass is executed. Usually, this was
fine but in some situations we would want the monitor to run immediately
after the change (MXS-740 and Galera). This patch changes the logic of
setting and clearing status bits to a delayed mode: changes are first applied
to a "status_pending"-variable, and only once the monitor runs will the
setting be applied. To reduce the delay, the monitor now has a flag
which is checked during sleep (between short 0.1s naps). If set, the
sleep is cut short.
If a server is not monitored, the status bits are set directly.
There is a small possibility of a race condition: If a monitor is stopped or
destroyed before the pending change is applied, the change is forgotten.
The configuration processing requires that all parameters for monitors
exist before they are used. This is wrong if we are aiming for a modular
system but is a necessary evil for the time being.
When a Galera cluster loses a member, it will recalculate the
wsrep_local_index values. As the index is zero-based, we can be certain
that in a valid cluster there will always be a node with an index of 0.
If the galeramon can't find a node with an index of 0, it means that
either the cluster hasn't stabilized and there's a pending recalculation
of the index or that there's no connectivity between MaxScale and the node
with the index value 0.
With this change and default settings, active-active MaxScale setups with
Galera clusters should always choose the same node as the master.
