Re: [Maria-discuss] Semi-sync replication hangs when changing binlog filename.

[Kristian Nielsen]

Joseph Glanville  writes:

> This fixes the problem for me. How do we go about getting this into a release?

I can push it. It should go into 10.1, I think (this code is not in 10.0).

Thanks for testing!

 - Kristian.

___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help   : https://help.launchpad.net/ListHelp

Re: [Maria-discuss] can we use load data infile from moving a database to a 4 node cluster

[Guillaume Lefranc]

unfortunately no, you can't do that. your cluster needs to be empty.
Maybe have a look at mydumper/myloader that has parallel dump and restore
capabilities?

Regards
GL

On Tue, Aug 16, 2016 at 10:30 AM bbnlaccts acnts 
wrote:

> Thanks for the help Guillaume,
> Can we take a snapshot of a standalone database and run against a cluster
> which already has many other databases running on it ?
>
> Also an u please do point me to a link which talks about this methodology.
>
> Please do let me know
>
> Thanks in advance for helping out.
>
> Arun
>
>
> On Tue, Aug 16, 2016 at 1:51 PM, Guillaume Lefranc <
> guillaume.lefr...@mariadb.com> wrote:
>
>> Hi Arun,
>>
>> Galera requires only InnoDB tables, so the fastest way would be to take
>> an Xtrabackup snapshot of your MariaDB 10 standalone database, restore it
>> on one node of the Galera cluster and then turn on the 3 other nodes.
>>
>> Of course LOAD DATA INFILE is supported by Galera, but it would be slower.
>>
>> On Tue, Aug 16, 2016 at 9:55 AM bbnlaccts acnts 
>> wrote:
>>
>>> Hi,
>>> I am planning on moving a mariadb 10 database to a 4 node galera cluster.
>>> Since the db is kind of big with around 40 gigs capacity and i am
>>> getting very less maintenance window to complete this activity.
>>> Hence planning on using load data infile on alll the big tables.
>>>
>>> Can anybody please do let me know if i am going to see any
>>> drawbacks/bugs in this concept.
>>> Are there any precautionary measures i need to take in order to
>>> accomplish this task.
>>>
>>> Your help is greatly appreciated
>>>
>>> Thanks
>>> Arun
>>> ___
>>> Mailing list: https://launchpad.net/~maria-discuss
>>> Post to : maria-discuss@lists.launchpad.net
>>> Unsubscribe : https://launchpad.net/~maria-discuss
>>> More help   : https://help.launchpad.net/ListHelp
>>>
>> --
>> Guillaume Lefranc
>> Remote DBA Services Manager
>> MariaDB Corporation
>>
>
> --
Guillaume Lefranc
Remote DBA Services Manager
MariaDB Corporation
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help   : https://help.launchpad.net/ListHelp

[Maria-discuss] can we use load data infile from moving a database to a 4 node cluster

[bbnlaccts acnts]

Hi,
I am planning on moving a mariadb 10 database to a 4 node galera cluster.
Since the db is kind of big with around 40 gigs capacity and i am getting
very less maintenance window to complete this activity.
Hence planning on using load data infile on alll the big tables.

Can anybody please do let me know if i am going to see any drawbacks/bugs
in this concept.
Are there any precautionary measures i need to take in order to accomplish
this task.

Your help is greatly appreciated

Thanks
Arun
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help   : https://help.launchpad.net/ListHelp

[Maria-discuss] Regarding Automatic fail-over - Maxscale

[Karthick Subramanian]

Hi All,

Can anyone help me to figure out this. I am planning to use Binlogrouter
for replication (single master and single slave topology). I read from
google forums for maxscale that with binlog router, it'snot possible to use
automatic failover. Did anyone try this.

>From google forum (
https://groups.google.com/forum/#!msg/maxscale/DnFT4eb0BbQ/cHQ3TLEfBAAJ;context-place=forum/maxscale
)
"With the 1.3.0 version of MaxScale, the binlogrouter accepts STOP SLAVE,
START SLAVE and CHANGE MASTER TO ... commands. These can be used to
change the master server but automating this process with the repmgr is
not currently possible. Since the binlogrouter supports the normal slave
management commands, using something like MHA could be possible:
https://code.google.com/p/mysql-master-ha/ "

Could anyone help me to understand whether I can use binlog router along
with automatic failover feature using maxscale. I have read this blog using
shell script we can achieve, but not sure whether with binlog router, it ll
work? (
https://mariadb.com/blog/mariadb-automatic-failover-maxscale-and-mariadb-replication-manager
)

Regards,
Karth
___
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help   : https://help.launchpad.net/ListHelp