Hello,
Dnia poniedziałek, 22 września 2014 22:02:26 Sławek Kapłoński pisze:
> Hello,
>
> Answears below
>
> ---
> Best regards
> Sławek Kapłoński
> sla...@kaplonski.pl
>
> Dnia poniedziałek, 22 września 2014 13:41:51 Jay Pipes pisze:
> > Hi Peter, Sławek, answers inline...
> >
> > On 09/22/20
-operators@lists.openstack.org
Subject: Re: [Openstack-operators] Openstack and mysql galera with haproxy
Hello,
Answears below
---
Best regards
Sławek Kapłoński
sla...@kaplonski.pl
Dnia poniedziałek, 22 września 2014 13:41:51 Jay Pipes pisze:
> Hi Peter, Sławek, answers inline...
>
> On 09/22/2
Hello,
Answears below
---
Best regards
Sławek Kapłoński
sla...@kaplonski.pl
Dnia poniedziałek, 22 września 2014 13:41:51 Jay Pipes pisze:
> Hi Peter, Sławek, answers inline...
>
> On 09/22/2014 08:12 AM, Peter Boros wrote:
> > Hi,
> >
> > StaleDataError is not given by MySQL, but rather SQLAlc
Hi Peter, Sławek, answers inline...
On 09/22/2014 08:12 AM, Peter Boros wrote:
Hi,
StaleDataError is not given by MySQL, but rather SQLAlchemy. After a
quick look, it seems like SQLAlchemy gets this, if the update updated
different number of rows then it expected. I am not sure what is the
expe
Hi,
StaleDataError is not given by MySQL, but rather SQLAlchemy. After a
quick look, it seems like SQLAlchemy gets this, if the update updated
different number of rows then it expected. I am not sure what is the
expectation based on, perhaps soembody can chime in and we can put
this together. What
Hello,
Thanks for Your explanations. I thought so and now I decrease
"idle_connection_timeout" in neutron and nova. Now when master server is back
to cluster than in less than one minute all conections are again made to this
master node becuase old connections which was made to backup node are
Hi,
Let me answer this and one of your previous questions in one because
they are related.
Earlier you wrote:
> I made such modifications today in my infra and generally it looks
> better now. I don't see deadlocks. But I have one more problem with
> that: generally it works fine when main node i
Hello,
New questions below
---
Best regards
Sławek Kapłoński
sla...@kaplonski.pl
Dnia czwartek, 18 września 2014 09:45:21 Clint Byrum pisze:
> Excerpts from Sławek Kapłoński's message of 2014-09-18 09:29:27 -0700:
> > Hello,
> >
> > Is anyone here using openstack with mysql galera and haproxy?
Hello,
W dniu 2014-09-18 18:45, Clint Byrum napisał(a):
Excerpts from Sławek Kapłoński's message of 2014-09-18 09:29:27 -0700:
Hello,
Is anyone here using openstack with mysql galera and haproxy? Have You
got any
problems with that?
I was today installed such ha infra for database (two mysql
Hello,
I made such modifications today in my infra and generally it looks
better now. I don't see deadlocks. But I have one more problem with
that: generally it works fine when main node is active but in situation
when this node is down, haproxy connect to one of backup nodes. Still
all is ok
One big win is that using xtrabackup (or xtrabackup_v2 if you’re on 5.6) is
that your tables don’t lock momentarily like they do using rsync.
At my previous shop, we had a replicated keystone db across various sites, so
wan segments (gmcast.segment ) became an attractive way to speed replication.
Excerpts from Sławek Kapłoński's message of 2014-09-18 12:34:53 -0700:
> Hello,
>
> Thanks all of You guys for help. I will try to apply that percona's scripts
> and configure it in active/pasive mode.
> About "mysql has gone away" I thought so it is not big problem but it looks
> "ugly" in logs
Hello,
Thanks all of You guys for help. I will try to apply that percona's scripts
and configure it in active/pasive mode.
About "mysql has gone away" I thought so it is not big problem but it looks
"ugly" in logs
Simon - I think that it is not related strictly to neutron and similar
problems
We're using haproxy in front of a 3 node galera cluster, and we see the lost
connection stuff pretty often, but that's normal behaviour, MySQL will tear
down idle connections, SQLA handles reconnecting fine.
As for deadlocks, if you are distributing writes across your nodes, I'd expect
that, we
If you want to use Galera + mysql, make sure you have haproxy set to prefer
a single host, otherwise you will have deadlocking problems. I know it kind
of defeats the purpose of having an all active database setup, but that's
what we've had to do. My other advice would be to use SSDs if you can
aff
Excerpts from Sławek Kapłoński's message of 2014-09-18 09:29:27 -0700:
> Hello,
>
> Is anyone here using openstack with mysql galera and haproxy? Have You got
> any
> problems with that?
> I was today installed such ha infra for database (two mysql servers in galera
> cluster and haproxy on con
Hello,
I'm using MySQL (Percona) and Galera with HAProxy and have previously run
into the same issues you describe.
The first thing to mention is "SELECT ... FOR UPDATE" queries do not work
well with multi-master environments. This was discovered during discussion
at the OpenStack Atlanta Summit.
I’ve seen that error, but not specifically with HAProxy (I was using an F5 at
the time)
IIRC, the problem was with my connection checking.
I assume you’re doing something similar to this:
http://www.percona.com/doc/percona-xtradb-cluster/5.5/howtos/haproxy.html
On Sep 18, 2014, at 9:29 AM, Sławe
Hello,
Is anyone here using openstack with mysql galera and haproxy? Have You got any
problems with that?
I was today installed such ha infra for database (two mysql servers in galera
cluster and haproxy on controller and neutron node, this haproxy is connecting
to one of galera servers with ro
19 matches
Mail list logo
