Mailing List Archive: Adding mysql Resource

Adding mysql Resource - is this normal?

Jan 14, 2014, 11:03 AM

Post #1 of 4 (1671 views)

Hello All,

I've worked with HA and pacemaker for a few years now and today I'm
working with a Fedora 19 cluster that I just created. Following the
excellent /Clusters from Scratch/ by Andrew Beekhof I've created a
couple of resources (IPAddr and Apache) and it works flawlessly - God
how I love a document that spells out how to do something and is
actually readable and correct!

Anyway, I decided to add the mysql resource before I do DRBD stuff, just
to test it out with no shared storage or DRBD. It failed with this error:
*Jan 14 13:50:59 firethorn mysql(MySQL)[10971]: ERROR: Setup problem:
couldn't find command: /usr/bin/safe_mysqld*

I did a find on both servers and sure enough, that command didn't exist.

Doing a systemctl status mysqld.service after starting mysqld manually,
I noticed this:

mysqld.service - MariaDB database server
Loaded: loaded (/usr/lib/systemd/system/mysqld.service; enabled)
Active: active (running) since Tue 2014-01-14 14:57:42 AST; 1s ago
Process: 2177 ExecStartPost=/usr/libexec/mysqld-wait-ready $MAINPID
(code=exited, status=0/SUCCESS)
Process: 2154 ExecStartPre=/usr/libexec/mysqld-prepare-db-dir %n
(code=exited, status=0/SUCCESS)
Main PID: 2176 (*mysqld_safe*)

hmmm, mysqld_safe is there but safe_mysqld is not. So I figured, lets
to a sym link in /usr/bin where mysqld_safe is - > *ln -s mysqld_safe
safe_mysqld*
So now safe_mysqld exists. I re-added my mysql resource and it worked.
No issues.

Is this normal guys? Why is this worded backwards?

Thanks as always,

Mike

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Re: Adding mysql Resource - is this normal? [ In reply to ]

cgasmith at comcast

Jan 14, 2014, 12:29 PM

Post #2 of 4 (1641 views)

Permalink

MySQL was deprecated in fc19+ in favor or MariaDB (fork of) looks like they dropped the ball on the ra, you may be able to just create a symlink, outside of that pound on fedora forum

Sent from my Verizon Wireless 4G LTE DROID

mike <mgbutler@nbnet.nb.ca> wrote:

>Hello All,
>
>I've worked with HA and pacemaker for a few years now and today I'm
>working with a Fedora 19 cluster that I just created. Following the
>excellent /Clusters from Scratch/ by Andrew Beekhof I've created a
>couple of resources (IPAddr and Apache) and it works flawlessly - God
>how I love a document that spells out how to do something and is
>actually readable and correct!
>
>Anyway, I decided to add the mysql resource before I do DRBD stuff, just
>to test it out with no shared storage or DRBD. It failed with this error:
>*Jan 14 13:50:59 firethorn mysql(MySQL)[10971]: ERROR: Setup problem:
>couldn't find command: /usr/bin/safe_mysqld*
>
>I did a find on both servers and sure enough, that command didn't exist.
>
>Doing a systemctl status mysqld.service after starting mysqld manually,
>I noticed this:
>
>mysqld.service - MariaDB database server
> Loaded: loaded (/usr/lib/systemd/system/mysqld.service; enabled)
> Active: active (running) since Tue 2014-01-14 14:57:42 AST; 1s ago
> Process: 2177 ExecStartPost=/usr/libexec/mysqld-wait-ready $MAINPID
>(code=exited, status=0/SUCCESS)
> Process: 2154 ExecStartPre=/usr/libexec/mysqld-prepare-db-dir %n
>(code=exited, status=0/SUCCESS)
> Main PID: 2176 (*mysqld_safe*)
>
>hmmm, mysqld_safe is there but safe_mysqld is not. So I figured, lets
>to a sym link in /usr/bin where mysqld_safe is - > *ln -s mysqld_safe
>safe_mysqld*
>So now safe_mysqld exists. I re-added my mysql resource and it worked.
>No issues.
>
>Is this normal guys? Why is this worded backwards?
>
>Thanks as always,
>
>Mike
>
>_______________________________________________
>Linux-HA mailing list
>Linux-HA@lists.linux-ha.org
>http://lists.linux-ha.org/mailman/listinfo/linux-ha
>See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Re: Adding mysql Resource - is this normal? [ In reply to ]

mgbutler at nbnet

Jan 14, 2014, 12:31 PM

Post #3 of 4 (1637 views)

Permalink

Thanks, that is what I did, create the symlink. It just seemed odd to me
that I'd need to do that.

On 14-01-14 04:29 PM, C.Smith wrote:
> MySQL was deprecated in fc19+ in favor or MariaDB (fork of) looks like they dropped the ball on the ra, you may be able to just create a symlink, outside of that pound on fedora forum
>
> Sent from my Verizon Wireless 4G LTE DROID
>
> mike <mgbutler@nbnet.nb.ca> wrote:
>
>> Hello All,
>>
>> I've worked with HA and pacemaker for a few years now and today I'm
>> working with a Fedora 19 cluster that I just created. Following the
>> excellent /Clusters from Scratch/ by Andrew Beekhof I've created a
>> couple of resources (IPAddr and Apache) and it works flawlessly - God
>> how I love a document that spells out how to do something and is
>> actually readable and correct!
>>
>> Anyway, I decided to add the mysql resource before I do DRBD stuff, just
>> to test it out with no shared storage or DRBD. It failed with this error:
>> *Jan 14 13:50:59 firethorn mysql(MySQL)[10971]: ERROR: Setup problem:
>> couldn't find command: /usr/bin/safe_mysqld*
>>
>> I did a find on both servers and sure enough, that command didn't exist.
>>
>> Doing a systemctl status mysqld.service after starting mysqld manually,
>> I noticed this:
>>
>> mysqld.service - MariaDB database server
>> Loaded: loaded (/usr/lib/systemd/system/mysqld.service; enabled)
>> Active: active (running) since Tue 2014-01-14 14:57:42 AST; 1s ago
>> Process: 2177 ExecStartPost=/usr/libexec/mysqld-wait-ready $MAINPID
>> (code=exited, status=0/SUCCESS)
>> Process: 2154 ExecStartPre=/usr/libexec/mysqld-prepare-db-dir %n
>> (code=exited, status=0/SUCCESS)
>> Main PID: 2176 (*mysqld_safe*)
>>
>> hmmm, mysqld_safe is there but safe_mysqld is not. So I figured, lets
>> to a sym link in /usr/bin where mysqld_safe is - > *ln -s mysqld_safe
>> safe_mysqld*
>> So now safe_mysqld exists. I re-added my mysql resource and it worked.
>> No issues.
>>
>> Is this normal guys? Why is this worded backwards?
>>
>> Thanks as always,
>>
>> Mike
>>
>> _______________________________________________
>> Linux-HA mailing list
>> Linux-HA@lists.linux-ha.org
>> http://lists.linux-ha.org/mailman/listinfo/linux-ha
>> See also: http://linux-ha.org/ReportingProblems
> _______________________________________________
> Linux-HA mailing list
> Linux-HA@lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Re: Adding mysql Resource - is this normal? [ In reply to ]

alain.moulle at bull

Jan 14, 2014, 10:51 PM

Post #4 of 4 (1640 views)

Permalink

Hi Mike,

I don't know why mysqld is missing, but mysqld_safe will not work in
pacemaker, as it is "HA in itself" , meaning that if you
stop the daemon, it will be automatically re-started on the same node,
completely by itself, without any pacemaker order or even configuration.
And this leads obviously to problems under pacemaker as the re-start is
done independantly of any pacemaker orders.

Alain
> Hello All,
>
> I've worked with HA and pacemaker for a few years now and today I'm
> working with a Fedora 19 cluster that I just created. Following the
> excellent /Clusters from Scratch/ by Andrew Beekhof I've created a
> couple of resources (IPAddr and Apache) and it works flawlessly - God
> how I love a document that spells out how to do something and is
> actually readable and correct!
>
> Anyway, I decided to add the mysql resource before I do DRBD stuff,
> just to test it out with no shared storage or DRBD. It failed with
> this error:
> *Jan 14 13:50:59 firethorn mysql(MySQL)[10971]: ERROR: Setup problem:
> couldn't find command: /usr/bin/safe_mysqld*
>
> I did a find on both servers and sure enough, that command didn't exist.
>
> Doing a systemctl status mysqld.service after starting mysqld
> manually, I noticed this:
>
> mysqld.service - MariaDB database server
> Loaded: loaded (/usr/lib/systemd/system/mysqld.service; enabled)
> Active: active (running) since Tue 2014-01-14 14:57:42 AST; 1s ago
> Process: 2177 ExecStartPost=/usr/libexec/mysqld-wait-ready $MAINPID
> (code=exited, status=0/SUCCESS)
> Process: 2154 ExecStartPre=/usr/libexec/mysqld-prepare-db-dir %n
> (code=exited, status=0/SUCCESS)
> Main PID: 2176 (*mysqld_safe*)
>
> hmmm, mysqld_safe is there but safe_mysqld is not. So I figured, lets
> to a sym link in /usr/bin where mysqld_safe is - > *ln -s mysqld_safe
> safe_mysqld*
> So now safe_mysqld exists. I re-added my mysql resource and it worked.
> No issues.
>
> Is this normal guys? Why is this worded backwards?
>
> Thanks as always,
>
> Mike
>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA@lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems