- STRERROR_BUFLEN moved to cdefs.h and renamed to MXS_STRERROR_BUFLEN.
Better would be to provide a 'const char* mxs_strerror(int errno)'
that would have a thread specific buffer for the error message.
- MIN and MAX also moved to defs.h as MXS_MIN and MXS_MAX.
- Now only mlist.h of the headers depend upon skygw_utils.h.
When a hole is detected while receiving events an ignorable event is
written in maxscale binlog. Such event is not sent ti slave servers.
Maxbinlogcheck can now check encrypted binlog files without decrypting
them
- All now include maxscale/cdefs.h as the very first file.
- MXS_[BEGIN|END]_DECLS added to all C-headers.
Strictly speaking not necessary for private headers, but
does not hurt either.
- Include guards moved to the very top of the file.
- #pragma once added.
- Headers now to be included as <maxscale/xyz.h>
- First step, no cleanup of headers has been made. Only moving
from one place to another + necessary modifications.
Authenticators now have a similar mechanism to the `router_options`
parameter which enables configurable authentication.
The authenticators also have a new initialize entry point which is similar
to the createInstance entry point of the filters and routers. The value of
`authenticator_options` is passed as a parameter to this function. The
return vaulue of the `initialize` entry point is passed to the `create`
entry point.
Events saved and read have 15 bytes of header XORed with IV.
Partial events are still not handled.
Next implementation will encrypt the whole event instead of 15 bytes XOR
Encryption Context and Encryption Setup structures have been added to
ROUTER_INSTANCE
Replication doesn’t start if binlog file has START_ENCRYPTION_EVENT but
router_option ‘encrypt_binlog’ is Off
The backend reference states should be cleared when a reconnection attempt
is made. Should the creation of a new DCB succeed, the backend should no
longer be closed.
Addition of START_ENCRYPTION_EVENT when encrypt_binlog=1 in Binlog
Server option.
Event is not sent to any slave.
MaxBinlogCheck understands the new event added in MariaDB 10.1.7: the
number of events = 164 as reported by FormatDescriptionEvent
An IGNORABLE event is added into binlog when a gap between two events
is detected.
New routines create and write special events.
Special events are not sent to slaves.
When a backend causes an error and it should be sent to the client, the
backend reference was closed but the waiting results state was not
cleared. This caused a debug assertion to be hit.
The active operation counters are now closed every time a backend referece
is taken out of use. This should fix a few debug assertions that were hit
in tests.
When a client executes commands which do not return results (for example
inserting BLOB data via the C API), readwritesplit expects a result for
each sent packet. This is a somewhat of a false assumption but it clears
itself out when the session is closed normally. If the session is closed
due to an error, the counter is not decremented.
Each sesssion should only increase the number of active operation on a
server by one operation. By checking that the session is not already
executing an operation before incrementing the active operation count the
runtime operation count will be correct.
