Yeah - I read about that. I agree that MHA looks like the best one. I
am not experienced with it though (something I need to fix asap!).
I do have experience with the other ones in large environments. And
luckily...they have not failed to me yet!! (fingers crossed!).
2012/10/11, Rick James <rjames@stripped>:
> Yeah, but did you hear about the multi-hour outage of GIT? They depended on
> those tools.
> Until 5.6 is really available, I vote for
> as the best bet for HA, failover, etc.
>> -----Original Message-----
>> From: Manuel Arostegui [mailto:manuel@stripped]
>> Sent: Thursday, October 11, 2012 7:28 AM
>> To: Oriol Martí
>> Cc: replication@stripped
>> Subject: Re: Best solution for a HA without shared storage
>> 2012/10/11 Oriol Martí <omarti@stripped>
>> > Hello,
>> > I've to make a HA mysql, for me the scalability it's not important
>> > because it's not an environment with a lot of queries.
>> > I've thinked about, and I think that the best solution is to mount an
>> > master-slave arquitecture, and if the master goes down, promote the
>> > slave to master. Monitoring this case, after manually we can put the
>> > master with slave status and recover the environment.
>> > What do you think that's the best tool to do this? Pacemaker or Red
>> > Hat Cluster Suite? Does anybody knows some example of configurations?
>> Hello Oriol,
>> There are several ways to it automatically moving just a VIP:
>> - mmm
>> - Pacemaker
>> - Keepalived
>> Whatever solution you finally go for...make sure you handle correctly
>> the read_only option - as typically the slaves are recommended to have
>> it ON.
>> So when promoting a slave automatically (or manually) make sure you or
>> your tool set it to OFF before moving the VIP to the new master.
>> Hope this helps,
Enviado desde mi dispositivo móvil