Bernhard, All, hi.
> Stewart,
>
> Stewart Smith wrote:
>> I think to really have this work we need a "connect timeout" for
>> client/rpl connections (like we do for mgmapi in ndb) and a way to wait
>> for it to expire (or an error) so we could write it properly.
>
> I fully agree and there is already a bug report for it: #30932.
>
> For telco in general it is essential to have timeout for these things
> and not to hang around waiting ...
>
> There is also other interesting side effects to connected to it: e.g.
> #31024.
I have left my comments on bug#30932 page on that `slave_net_timeout'
sematics dramatically changes since bug#20435/wl#342 heartbeat will be
pushed.
The fear that idling master causes lots of reconnections due to a
small `slave_net_timeout' is vanished.
`slave_net_timeout' is truely can be considered to be a rather
"connect timeout" as it can only expire until the slave successfully
connects (and called the master for sending heartbeats).
>
> So this is a good chance to get something like that at least into the
> ndb 6.X releases :)
>
> /Bernd
>
cheers,
Andrei
