maria-discuss team mailing list archive

[Kristian Nielsen <knielsen@xxxxxxxxxxxxxxx>]

Greg <skygreg@xxxxxxxxx> writes:

> I think you're right. I know that DRBD is not the best choice to have
> high-availibility on a MariaDB master, but I have my reasons to make this

Isn't it? I do not have practical experience with MariaDB/MySQL in a HA setup,
but I have always thought that DRBD was one of the best choices, being a
simple and robust solution.

Please do not be confused with my thoughts on SIGKILL and fdatasync(), as I
said it was just academic interest. MariaDB and DRBD are doing the right thing
here, as far as I understand.

> In DRBD, I just set protocol B which should be sufficient, but I can easily
> test with C. I have test some disk options

Well, if I understood you correctly you tested with sync_binlog=1 and verified
that no inconsistencies occured. So this should be fine (I am not aware of the
difference between B and C).

> In MariaDB I set :
> innodb_flush_log_at_trx_commit  = 1
> sync_binlog                     = 1
> innodb_file_per_table = 1

Yes, this is correct.

> I can test things, patchs, but I don't know what ... I didn't found
> WL#4925 patch
> so I can't test it.

We do not currently have any patch or plans for pre-allocation of binlog.

 - Kristian.