Re: [CentOS] mysql-connector-odbc i686 package

[Jose Maria Terry Jimenez]

El 30/7/20 a las 4:51, sthustfo escribió:
> Hi all,
>
> I am looking for a i686 package (32 bit) for mysql-connector-odbc. However,
> I could locate only the x86-64 version on the repositories. Not much of a
> DB person but trying to connect to a 64bit MySQL server on remote host.
>
> Any suggestions where I can locate the i686 package? Is there a mariadb
> package that I can use?
>
> Thanks.
> ___
> CentOS mailing list
> CentOS@centos.org
> https://lists.centos.org/mailman/listinfo/centos

Hello

The client architecture has nothing to see with server. Said that, if
you need the conector for 32 bit, try with the one in
https://dev.mysql.com/downloads/connector/odbc/ under RHEL/Oracle Linux.
It **should** work.

Good luck

Best,

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql-connector-odbc i686 package

[Laack, Andrea P]

If you can find the RPMS, you could rebuild for I686.

-Original Message-
From: CentOS  On Behalf Of sthustfo
Sent: Wednesday, July 29, 2020 9:52 PM
To: CentOS mailing list 
Subject: {EXTERNAL} [CentOS] mysql-connector-odbc i686 package


CAUTION:  This email originated outside of BSWH; avoid action unless you know 
the content is safe. Send suspicious emails as attachments to 
badem...@bswhealth.org.

Hi all,

I am looking for a i686 package (32 bit) for mysql-connector-odbc. However, I 
could locate only the x86-64 version on the repositories. Not much of a DB 
person but trying to connect to a 64bit MySQL server on remote host.

Any suggestions where I can locate the i686 package? Is there a mariadb package 
that I can use?

Thanks.
___
CentOS mailing list
CentOS@centos.org
https://urldefense.com/v3/__https://lists.centos.org/mailman/listinfo/centos__;!!JA_k2roV-A!U8Rqszj0J05KQTSPkjTVCaov_95qw-xryk_HYcEQ_qnqFbBw3ZI9bMS4t9x3-vOWfw$
 

**
The information contained in this e-mail may be privileged and/or confidential, 
and protected from disclosure, and no waiver of any attorney-client, work 
product, or other privilege is intended.  If you are the intended recipient, 
further disclosures are prohibited without proper authorization. If you are not 
the intended recipient (or have received this e-mail in error) please notify 
the sender immediately and destroy this e-mail. Any unauthorized copying, 
disclosure or distribution of the material in this e-mail is strictly forbidden 
and possibly a violation of federal or state law and regulations. The sender 
and Baylor Scott & White Health, and its affiliated entities, hereby expressly 
reserve all privileges and confidentiality that might otherwise be waived as a 
result of an erroneous or misdirected e-mail transmission. No employee or agent 
is authorized to conclude any binding agreement on behalf of Baylor Scott & 
White Health, or any affiliated entity, by e-mail without express written 
confirmation by the CEO, the Senior Vice President of Supply Chain Services or 
other duly authorized representative of Baylor Scott & White Health.
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql-connector-odbc i686 package

[sthustfo]

Hi all,

I am looking for a i686 package (32 bit) for mysql-connector-odbc. However,
I could locate only the x86-64 version on the repositories. Not much of a
DB person but trying to connect to a 64bit MySQL server on remote host.

Any suggestions where I can locate the i686 package? Is there a mariadb
package that I can use?

Thanks.
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL on Centos 7 (armv7)

[Robert Moskowitz]

On 06/03/2017 08:52 AM, Johnny Hughes wrote:

On 06/03/2017 08:19 AM, Leon Vergottini wrote:

Dear Community

  


I am trying to install MySQL 5.6 on a Raspberry PI 3B.   Using my normal 
procedure by downloading the yum repo rpm and installing MySQL through that is 
not working.  There is no rpm information for arm7.

  


Can anyone please point me in a direction where I can find a procedure on how 
to install MySQL 5.6 on Centos 7 running on a PI3?

CentOS 7 uses mariadb, not mysql.  The version available in the repo
should be 5.5.52.

If Oracle builds mysql for armv7hl (armhfp), binaries might be available
from them.



Any more questions should be taken to:

https://lists.centos.org/mailman/listinfo/arm-dev

Are you using the armx64 build or armv7hl (32) build?

I use mariadb with the armv7hl build.


___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL on Centos 7 (armv7)

[Johnny Hughes]

On 06/03/2017 08:19 AM, Leon Vergottini wrote:
> Dear Community
> 
>  
> 
> I am trying to install MySQL 5.6 on a Raspberry PI 3B.   Using my normal 
> procedure by downloading the yum repo rpm and installing MySQL through that 
> is not working.  There is no rpm information for arm7.
> 
>  
> 
> Can anyone please point me in a direction where I can find a procedure on how 
> to install MySQL 5.6 on Centos 7 running on a PI3?

CentOS 7 uses mariadb, not mysql.  The version available in the repo
should be 5.5.52.

If Oracle builds mysql for armv7hl (armhfp), binaries might be available
from them.



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL on Centos 7 (armv7)

[Leon Vergottini]

Dear Community

 

I am trying to install MySQL 5.6 on a Raspberry PI 3B.   Using my normal 
procedure by downloading the yum repo rpm and installing MySQL through that is 
not working.  There is no rpm information for arm7.

 

Can anyone please point me in a direction where I can find a procedure on how 
to install MySQL 5.6 on Centos 7 running on a PI3?

 

Kind Regards,

Leon

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql client ssl connection failure

[Devin Reade]

The problem and solution is posted here for posterity.

On a production CentOS 5.11 system, I recently had a cron job
that started failing.  One of the thing that the cron job does
is connect to a MySQL server via TCP using SSL.

Debugging things got me to the point where invoking the mysql
client would result in:

% mysql
ERROR 2026 (HY000): SSL connection error

(Not exactly informative.)  Given that there have been recent openssl
updates, I figured the problem was there, however the lack of any
decent debug output from mysql/openssl made it hard to figure out.
After going down a few false starts, including looking at the SSL
certificate and root CA, I found this:

 

It turns out that the default cipher for the mysql client is
hardcoded, and the one specified is no longer supported in the
server.  The solution is to add to /etc/my.cnf (or ~/.my.cnf)
the following (or the equivalent for your preferred cipher) to
the client section(s):

 ssl-cipher=AES128-SHA


Devin

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Götz Reinicke - IT Koordinator]

Hi,

as that example is for maria db, teh oracle mysql rpm dose not ship with
something like

/etc/systemd/system/mysqld.service.d

Can I just create /etc/systemd/system/mysqld.service.d/limits.conf and
than I'm ready to go?

Thanks for your patience and regards . Götz

Am 29.04.15 um 14:58 schrieb Johan Kooijman:
> Gotz,
> 
> This is due to systemd, it overrules your settings. Add a file to systemd
> config fixes it:
> 
> [root@mysql2 ~]# cat /etc/systemd/system/mariadb.service.d/limits.conf
> [Service]
> LimitNOFILE=1
> LimitMEMLOCK=1



-- 
Götz Reinicke
IT-Koordinator

Tel. +49 7141 969 82 420
E-Mail goetz.reini...@filmakademie.de

Filmakademie Baden-Württemberg GmbH
Akademiehof 10
71638 Ludwigsburg
www.filmakademie.de

Eintragung Amtsgericht Stuttgart HRB 205016

Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
Staatssekretär im Ministerium für Wissenschaft,
Forschung und Kunst Baden-Württemberg

Geschäftsführer: Prof. Thomas Schadt

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Jegadeesh Kumar]

we can set ulimt to limit the number of open files

On Thu, Apr 30, 2015 at 3:22 PM, Götz Reinicke - IT Koordinator <
goetz.reini...@filmakademie.de> wrote:

> +1 Thanks! That was a little bit confusing at first.
>
> Regards . Götz
>
> Am 30.04.15 um 04:21 schrieb carlh04...@gmail.com:
> > Thank you for clarifying this, Johan. Very much appreciated!
> >
> > On Wed, 29 Apr 2015 22:28:00 +0200
> > Johan Kooijman wrote:
> >
> >> Carl,
> >>
> >> By default my.cnf has to obey the OS limits, so in this case the
> >> order is: systemd > /etc/security/limits* > /etc/my*.
> <...>
>
>
> --
> Götz Reinicke
> IT-Koordinator
>
> Tel. +49 7141 969 82 420
> E-Mail goetz.reini...@filmakademie.de
>
> Filmakademie Baden-Württemberg GmbH
> Akademiehof 10
> 71638 Ludwigsburg
> www.filmakademie.de
>
> Eintragung Amtsgericht Stuttgart HRB 205016
>
> Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
> Staatssekretär im Ministerium für Wissenschaft,
> Forschung und Kunst Baden-Württemberg
>
> Geschäftsführer: Prof. Thomas Schadt
>
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>
>
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Götz Reinicke - IT Koordinator]

+1 Thanks! That was a little bit confusing at first.

Regards . Götz

Am 30.04.15 um 04:21 schrieb carlh04...@gmail.com:
> Thank you for clarifying this, Johan. Very much appreciated!
> 
> On Wed, 29 Apr 2015 22:28:00 +0200
> Johan Kooijman wrote:
> 
>> Carl,
>>
>> By default my.cnf has to obey the OS limits, so in this case the
>> order is: systemd > /etc/security/limits* > /etc/my*.
<...>


-- 
Götz Reinicke
IT-Koordinator

Tel. +49 7141 969 82 420
E-Mail goetz.reini...@filmakademie.de

Filmakademie Baden-Württemberg GmbH
Akademiehof 10
71638 Ludwigsburg
www.filmakademie.de

Eintragung Amtsgericht Stuttgart HRB 205016

Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
Staatssekretär im Ministerium für Wissenschaft,
Forschung und Kunst Baden-Württemberg

Geschäftsführer: Prof. Thomas Schadt

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[carlh04426]

Thank you for clarifying this, Johan. Very much appreciated!

On Wed, 29 Apr 2015 22:28:00 +0200
Johan Kooijman wrote:

> Carl,
> 
> By default my.cnf has to obey the OS limits, so in this case the
> order is: systemd > /etc/security/limits* > /etc/my*.
> 
> On Wed, Apr 29, 2015 at 3:22 PM, Carl E. Hartung
>  wrote:
> 
> > Hi Johan,
> >
> > Does systemd also overrule /etc/my.conf?
> >
> > Thx!
> >
> > Carl
> >
> > On Wed, 29 Apr 2015 14:58:52 +0200
> > Johan Kooijman wrote:
> >
> > > Gotz,
> > >
> > > This is due to systemd, it overrules your settings. Add a file to
> > > systemd config fixes it:
> > >
> > > [root@mysql2 ~]#
> > > cat /etc/systemd/system/mariadb.service.d/limits.conf [Service]
> > > LimitNOFILE=1
> > > LimitMEMLOCK=1
> > >
> > >
> > > On Wed, Apr 29, 2015 at 8:31 AM, Götz Reinicke - IT Koordinator <
> > > goetz.reini...@filmakademie.de> wrote:
> > >
> > > > Hi,
> > > >
> > > > may be somewon has a working solution and information on that:
> > > >
> > > > I installed the most recent mysql community on a server and do
> > > > get a lot of "errno: 24 - Too many open files".
> > > >
> > > > There are suggestions to increase the open_files_limit,
> > > > change/add that to /etc/security/limits.conf and modify the
> > > > systemd script by hand.
> > > >
> > > > Depending on how you start mysql (restart, or at systemboot
> > > > time ...) the limit get set or not.
> > > >
> > > > Question: Is that may be a known bug or something?
> > > >
> > > > And thats mit a question regarding mariadb or mysql.
> > > >
> > > > Regards . Götz
> > > > --
> > > > Götz Reinicke
> > > > IT-Koordinator
> > > >
> > > > Tel. +49 7141 969 82 420
> > > > E-Mail goetz.reini...@filmakademie.de
> > > >
> > > > Filmakademie Baden-Württemberg GmbH
> > > > Akademiehof 10
> > > > 71638 Ludwigsburg
> > > > www.filmakademie.de
> > > >
> > > > Eintragung Amtsgericht Stuttgart HRB 205016
> > > >
> > > > Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
> > > > Staatssekretär im Ministerium für Wissenschaft,
> > > > Forschung und Kunst Baden-Württemberg
> > > >
> > > > Geschäftsführer: Prof. Thomas Schadt
> > > >
> > > >
> > > > ___
> > > > CentOS mailing list
> > > > CentOS@centos.org
> > > > http://lists.centos.org/mailman/listinfo/centos
> > > >
> > > >
> > >
> > >
> >
> > ___
> > CentOS mailing list
> > CentOS@centos.org
> > http://lists.centos.org/mailman/listinfo/centos
> >
> 
> 
> 

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Johan Kooijman]

Carl,

By default my.cnf has to obey the OS limits, so in this case the order is:
systemd > /etc/security/limits* > /etc/my*.

On Wed, Apr 29, 2015 at 3:22 PM, Carl E. Hartung 
wrote:

> Hi Johan,
>
> Does systemd also overrule /etc/my.conf?
>
> Thx!
>
> Carl
>
> On Wed, 29 Apr 2015 14:58:52 +0200
> Johan Kooijman wrote:
>
> > Gotz,
> >
> > This is due to systemd, it overrules your settings. Add a file to
> > systemd config fixes it:
> >
> > [root@mysql2 ~]# cat /etc/systemd/system/mariadb.service.d/limits.conf
> > [Service]
> > LimitNOFILE=1
> > LimitMEMLOCK=1
> >
> >
> > On Wed, Apr 29, 2015 at 8:31 AM, Götz Reinicke - IT Koordinator <
> > goetz.reini...@filmakademie.de> wrote:
> >
> > > Hi,
> > >
> > > may be somewon has a working solution and information on that:
> > >
> > > I installed the most recent mysql community on a server and do get
> > > a lot of "errno: 24 - Too many open files".
> > >
> > > There are suggestions to increase the open_files_limit, change/add
> > > that to /etc/security/limits.conf and modify the systemd script by
> > > hand.
> > >
> > > Depending on how you start mysql (restart, or at systemboot
> > > time ...) the limit get set or not.
> > >
> > > Question: Is that may be a known bug or something?
> > >
> > > And thats mit a question regarding mariadb or mysql.
> > >
> > > Regards . Götz
> > > --
> > > Götz Reinicke
> > > IT-Koordinator
> > >
> > > Tel. +49 7141 969 82 420
> > > E-Mail goetz.reini...@filmakademie.de
> > >
> > > Filmakademie Baden-Württemberg GmbH
> > > Akademiehof 10
> > > 71638 Ludwigsburg
> > > www.filmakademie.de
> > >
> > > Eintragung Amtsgericht Stuttgart HRB 205016
> > >
> > > Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
> > > Staatssekretär im Ministerium für Wissenschaft,
> > > Forschung und Kunst Baden-Württemberg
> > >
> > > Geschäftsführer: Prof. Thomas Schadt
> > >
> > >
> > > ___
> > > CentOS mailing list
> > > CentOS@centos.org
> > > http://lists.centos.org/mailman/listinfo/centos
> > >
> > >
> >
> >
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
Met vriendelijke groeten / With kind regards,
Johan Kooijman
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Carl E. Hartung]

Hi Johan,

Does systemd also overrule /etc/my.conf?

Thx!

Carl

On Wed, 29 Apr 2015 14:58:52 +0200
Johan Kooijman wrote:

> Gotz,
> 
> This is due to systemd, it overrules your settings. Add a file to
> systemd config fixes it:
> 
> [root@mysql2 ~]# cat /etc/systemd/system/mariadb.service.d/limits.conf
> [Service]
> LimitNOFILE=1
> LimitMEMLOCK=1
> 
> 
> On Wed, Apr 29, 2015 at 8:31 AM, Götz Reinicke - IT Koordinator <
> goetz.reini...@filmakademie.de> wrote:
> 
> > Hi,
> >
> > may be somewon has a working solution and information on that:
> >
> > I installed the most recent mysql community on a server and do get
> > a lot of "errno: 24 - Too many open files".
> >
> > There are suggestions to increase the open_files_limit, change/add
> > that to /etc/security/limits.conf and modify the systemd script by
> > hand.
> >
> > Depending on how you start mysql (restart, or at systemboot
> > time ...) the limit get set or not.
> >
> > Question: Is that may be a known bug or something?
> >
> > And thats mit a question regarding mariadb or mysql.
> >
> > Regards . Götz
> > --
> > Götz Reinicke
> > IT-Koordinator
> >
> > Tel. +49 7141 969 82 420
> > E-Mail goetz.reini...@filmakademie.de
> >
> > Filmakademie Baden-Württemberg GmbH
> > Akademiehof 10
> > 71638 Ludwigsburg
> > www.filmakademie.de
> >
> > Eintragung Amtsgericht Stuttgart HRB 205016
> >
> > Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
> > Staatssekretär im Ministerium für Wissenschaft,
> > Forschung und Kunst Baden-Württemberg
> >
> > Geschäftsführer: Prof. Thomas Schadt
> >
> >
> > ___
> > CentOS mailing list
> > CentOS@centos.org
> > http://lists.centos.org/mailman/listinfo/centos
> >
> >
> 
> 

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Johan Kooijman]

Gotz,

This is due to systemd, it overrules your settings. Add a file to systemd
config fixes it:

[root@mysql2 ~]# cat /etc/systemd/system/mariadb.service.d/limits.conf
[Service]
LimitNOFILE=1
LimitMEMLOCK=1


On Wed, Apr 29, 2015 at 8:31 AM, Götz Reinicke - IT Koordinator <
goetz.reini...@filmakademie.de> wrote:

> Hi,
>
> may be somewon has a working solution and information on that:
>
> I installed the most recent mysql community on a server and do get a lot
> of "errno: 24 - Too many open files".
>
> There are suggestions to increase the open_files_limit, change/add that
> to /etc/security/limits.conf and modify the systemd script by hand.
>
> Depending on how you start mysql (restart, or at systemboot time ...)
> the limit get set or not.
>
> Question: Is that may be a known bug or something?
>
> And thats mit a question regarding mariadb or mysql.
>
> Regards . Götz
> --
> Götz Reinicke
> IT-Koordinator
>
> Tel. +49 7141 969 82 420
> E-Mail goetz.reini...@filmakademie.de
>
> Filmakademie Baden-Württemberg GmbH
> Akademiehof 10
> 71638 Ludwigsburg
> www.filmakademie.de
>
> Eintragung Amtsgericht Stuttgart HRB 205016
>
> Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
> Staatssekretär im Ministerium für Wissenschaft,
> Forschung und Kunst Baden-Württemberg
>
> Geschäftsführer: Prof. Thomas Schadt
>
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>
>


-- 
Met vriendelijke groeten / With kind regards,
Johan Kooijman
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Todor Petkov]

On 29/04/2015 09:31 AM, Götz Reinicke - IT Koordinator wrote:

Hi,

may be somewon has a working solution and information on that:

I installed the most recent mysql community on a server and do get a 
lot

of "errno: 24 - Too many open files".

There are suggestions to increase the open_files_limit, change/add that
to /etc/security/limits.conf and modify the systemd script by hand.

Depending on how you start mysql (restart, or at systemboot time ...)
the limit get set or not.

Question: Is that may be a known bug or something?

And thats mit a question regarding mariadb or mysql.

Regards . Götz
--
Götz Reinicke
IT-Koordinator

Tel. +49 7141 969 82 420
E-Mail goetz.reini...@filmakademie.de

Filmakademie Baden-Württemberg GmbH
Akademiehof 10
71638 Ludwigsburg
www.filmakademie.de

Eintragung Amtsgericht Stuttgart HRB 205016

Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
Staatssekretär im Ministerium für Wissenschaft,
Forschung und Kunst Baden-Württemberg

Geschäftsführer: Prof. Thomas Schadt


Hi,

can you do "lsof|grep mysql" and show us the output?

Regards,


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Mysql 5.6, Centos 7 and errno: 24 - Too many open files

[Götz Reinicke - IT Koordinator]

Hi,

may be somewon has a working solution and information on that:

I installed the most recent mysql community on a server and do get a lot
of "errno: 24 - Too many open files".

There are suggestions to increase the open_files_limit, change/add that
to /etc/security/limits.conf and modify the systemd script by hand.

Depending on how you start mysql (restart, or at systemboot time ...)
the limit get set or not.

Question: Is that may be a known bug or something?

And thats mit a question regarding mariadb or mysql.

Regards . Götz
-- 
Götz Reinicke
IT-Koordinator

Tel. +49 7141 969 82 420
E-Mail goetz.reini...@filmakademie.de

Filmakademie Baden-Württemberg GmbH
Akademiehof 10
71638 Ludwigsburg
www.filmakademie.de

Eintragung Amtsgericht Stuttgart HRB 205016

Vorsitzender des Aufsichtsrats: Jürgen Walter MdL
Staatssekretär im Ministerium für Wissenschaft,
Forschung und Kunst Baden-Württemberg

Geschäftsführer: Prof. Thomas Schadt

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Paul Jones]

On 30/03/15 05:22, Tim Dunphy wrote:


mysql> FLUSH PRIVILEGES;




Yup! That was it. Thanks for the reminder! :)

Tim


>From the mySQL man pages:

If you modify the grant tables indirectly using account-management statements 
such as GRANT, 
REVOKE, SET 
PASSWORD, or RENAME 
USER, the server 
notices these changes and loads the grant tables into memory again immediately.

So I'm confused as to why FLUSH PRIVILEGES is necessary, because this suggests 
the GRANT command is doing a flush anyway.

I ask because I get this behaviour occasionally with adding mySQL 
users/permissions too, and I hate blithely running a command 'because it works' 
if it's not actually the correct thing to do (though admittedly it does work).

Because we're creating a user as well as assigning permissions maybe? So a 
GRANT to an existing user wouldn't require the FLUSH PRIVILEGES?

Appreciate this is a mySQL rather than CentOS question, but it bugs me on 
CentOS machines, so that's something!

Paul

--


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Fernando Cassia]

On Mon, Mar 30, 2015 at 3:16 AM, Fernando Cassia  wrote:

> You should keep in mind the security-related changes coming in MySQL 5..7,
> if you ever choose to upgrade.


Sorry, I erased this link by mistake while composing my reply. Here it goes.

http://mysqlopt.blogspot.com/2015/02/mysql-575-m15-out-of-box-security.html

FC
-- 
During times of Universal Deceit, telling the truth becomes a revolutionary
act
 George Orwell
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Fernando Cassia]

On Mon, Mar 30, 2015 at 12:27 AM, Tim Dunphy  wrote:

>  I've been having some trouble creating a mysql user that can connect to
> the database from localhost. It's always been a straight forward thing to
> do in the past, so its time for a sanity check, if you guys don't mind.
>

Hi Tim,

You should keep in mind the security-related changes coming in MySQL 5..7,
if you ever choose to upgrade.
https://dev.mysql.com/doc/refman/5.7/en/mysql-nutshell.html
http://mysqlserverteam.com/whats-new-in-mysql-5-7-so-far/

Repos
https://dev.mysql.com/downloads/repo/yum/

Best,
FC
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Hal Wigoda]

[root@ops:~] #mysql --user=proftpd --password=testpattern  -h localhost


On Sun, Mar 29, 2015 at 10:27 PM, Tim Dunphy  wrote:

> Hey all,
>
>  I've been having some trouble creating a mysql user that can connect to
> the database from localhost. It's always been a straight forward thing to
> do in the past, so its time for a sanity check, if you guys don't mind.
>
> Ok, so here's the actual command with actual simplified password that I'm
> using. It's on localhost so I don't think it's a security threat unless
> someone gets access to the box itself.
>
> mysql> grant all privileges on ftp.* to 'proftpd'@'localhost' identified
> by
> 'testpattern';
> Query OK, 0 rows affected (0.35 sec)
>
> Here's what it looks like when you select it from the mysql database:
>
> mysql> select User,Host,Password from user where User like 'proftpd';
> +-+---+---+
> | User| Host  | Password  |
> +-+---+---+
> | proftpd | localhost | *2EE931CA39652F1ED359A3A36961511B387E74A9 |
> +-+---+---+
> 1 row in set (0.00 sec)
>
> And here's my attempt to connect with the password shown. Which is
> something I don't usually do, but am doing now to demonstrate what's going
> on:
>
> [root@ops:~] #mysql -uproftpd -ptestpattern  -h localhost
> ERROR 1045 (28000): Access denied for user 'proftpd'@'localhost' (using
> password: YES)
>
> OK, so as I've said this should work!
>
> The database I'm trying to give the user access to does also exist:
>
> mysql> show databases like 'ftp';
> ++
> | Database (ftp) |
> ++
> | ftp|
> ++
> 1 row in set (0.34 sec)
>
>
> I checked the error log for mysql and didn't find any clues there:
>
> [root@ops:~] #grep log /etc/my.cnf
> log-error=/var/log/mysqld.log
>
> [root@ops:~] #tail /var/log/mysqld.log
> InnoDB: Restoring possible half-written data pages from the doublewrite
> InnoDB: buffer...
> 150329 13:30:34  InnoDB: Waiting for the background threads to start
> 150329 13:30:35 InnoDB: 5.5.42 started; log sequence number 6071094973
> 150329 13:30:35 [Note] Server hostname (bind-address): '0.0.0.0'; port:
> 3306
> 150329 13:30:35 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
> 150329 13:30:35 [Note] Server socket created on IP: '0.0.0.0'.
> 150329 13:30:35 [Note] Event Scheduler: Loaded 0 events
> 150329 13:30:35 [Note] /usr/libexec/mysqld: ready for connections.
> Version: '5.5.42'  socket: '/var/lib/mysql/mysql.sock'  port: 3306  MySQL
> Community Server (GPL) by Remi
>
> Does anybody out there have any idea why this isn't working?
>
> Thanks
> Tim
>
> --
> GPG me!!
>
> gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
-
Hal Wigoda
Chicago
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Tim Dunphy]

>
> mysql> FLUSH PRIVILEGES;


Yup! That was it. Thanks for the reminder! :)

Tim

On Mon, Mar 30, 2015 at 12:15 AM, Steven Tardy  wrote:

>
> > mysql> grant all privileges on ftp.* to 'proftpd'@'localhost'
> identified by
> > 'testpattern';
> > Query OK, 0 rows affected (0.35 sec)
>
> mysql> FLUSH PRIVILEGES;
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
GPG me!!

gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql can't connect from localhost -strange behavior

[Steven Tardy]

> mysql> grant all privileges on ftp.* to 'proftpd'@'localhost' identified by
> 'testpattern';
> Query OK, 0 rows affected (0.35 sec)

mysql> FLUSH PRIVILEGES;
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql can't connect from localhost -strange behavior

[Tim Dunphy]

Hey all,

 I've been having some trouble creating a mysql user that can connect to
the database from localhost. It's always been a straight forward thing to
do in the past, so its time for a sanity check, if you guys don't mind.

Ok, so here's the actual command with actual simplified password that I'm
using. It's on localhost so I don't think it's a security threat unless
someone gets access to the box itself.

mysql> grant all privileges on ftp.* to 'proftpd'@'localhost' identified by
'testpattern';
Query OK, 0 rows affected (0.35 sec)

Here's what it looks like when you select it from the mysql database:

mysql> select User,Host,Password from user where User like 'proftpd';
+-+---+---+
| User| Host  | Password  |
+-+---+---+
| proftpd | localhost | *2EE931CA39652F1ED359A3A36961511B387E74A9 |
+-+---+---+
1 row in set (0.00 sec)

And here's my attempt to connect with the password shown. Which is
something I don't usually do, but am doing now to demonstrate what's going
on:

[root@ops:~] #mysql -uproftpd -ptestpattern  -h localhost
ERROR 1045 (28000): Access denied for user 'proftpd'@'localhost' (using
password: YES)

OK, so as I've said this should work!

The database I'm trying to give the user access to does also exist:

mysql> show databases like 'ftp';
++
| Database (ftp) |
++
| ftp|
++
1 row in set (0.34 sec)


I checked the error log for mysql and didn't find any clues there:

[root@ops:~] #grep log /etc/my.cnf
log-error=/var/log/mysqld.log

[root@ops:~] #tail /var/log/mysqld.log
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
150329 13:30:34  InnoDB: Waiting for the background threads to start
150329 13:30:35 InnoDB: 5.5.42 started; log sequence number 6071094973
150329 13:30:35 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
150329 13:30:35 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
150329 13:30:35 [Note] Server socket created on IP: '0.0.0.0'.
150329 13:30:35 [Note] Event Scheduler: Loaded 0 events
150329 13:30:35 [Note] /usr/libexec/mysqld: ready for connections.
Version: '5.5.42'  socket: '/var/lib/mysql/mysql.sock'  port: 3306  MySQL
Community Server (GPL) by Remi

Does anybody out there have any idea why this isn't working?

Thanks
Tim

-- 
GPG me!!

gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[James B. Byrne]

On Thu, March 12, 2015 11:49, Tim Dunphy wrote:
>
> Ok, yeah I can understand that. I'll correct it. Still need a way to
> get SSL enabled however. Any suggestions there?

The method we use is to create an application specific directory under
/etc/pki and place its certificates and keys in there.  This is how
several packages from RHEL/EPEL do things:

drwxr-xr-x. 2 root root 4096 Mar   1 11:57 /etc/pki/appname

-rw-r--r-- 1 root appname  2561 Mar  2 15:09 83.der
-rw-r- 1 root appname  3243 Mar  2 10:46 83.key
-rw-r--r-- 1 root appname  9708 Mar  2 10:46 83.pem


HTH

-- 
***  E-Mail is NOT a SECURE channel  ***
James B. Byrnemailto:byrn...@harte-lyne.ca
Harte & Lyne Limited  http://www.harte-lyne.ca
9 Brockley Drive  vox: +1 905 561 1241
Hamilton, Ontario fax: +1 905 561 0757
Canada  L8E 3C3

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[Tim Dunphy]

Hey Alberto,

Perfect! Thanks for your response. Moving the certs and keys to an
alternate location worked exactly right.

Master:

MariaDB [(none)]> show variables like '%ssl%';
+---+--+
| Variable_name | Value|
+---+--+
| have_openssl  | YES  |
| have_ssl  | YES  |
| ssl_ca| /opt/mysql/ca.crt|
| ssl_capath|  |
| ssl_cert  | /opt/mysql/mysql.crt |
| ssl_cipher|  |
| ssl_key   | /opt/mysql/mysql.key |
+---+--+
7 rows in set (0.01 sec)

Slave:

mysql> show variables like '%ssl%';
+---++
| Variable_name | Value  |
+---++
| have_openssl  | YES|
| have_ssl  | YES|
| ssl_ca| /opt/mysql/ca.crt  |
| ssl_capath||
| ssl_cert  | /opt/mysql/mysql-slave.crt |
| ssl_cipher||
| ssl_key   | /opt/mysql/mysql-slave.key |
+---++
7 rows in set (0.00 sec)

At least now SSL is recognized by the systems.

mysql> show slave status \G
*** 1. row ***
   Slave_IO_State: Waiting for master to send event
  Master_Host: web2.somewhere.com
  Master_User: jf_slave
  Master_Port: 3306
Connect_Retry: 60
  Master_Log_File: mysql-bin.06
  Read_Master_Log_Pos: 27664
   Relay_Log_File: mysqld-relay-bin.02
Relay_Log_Pos: 391
Relay_Master_Log_File: mysql-bin.06
 Slave_IO_Running: Yes
Slave_SQL_Running: Yes
  Replicate_Do_DB: tesdb
  Replicate_Ignore_DB:
   Replicate_Do_Table:
   Replicate_Ignore_Table:
  Replicate_Wild_Do_Table:
  Replicate_Wild_Ignore_Table:
   Last_Errno: 0
   Last_Error:
 Skip_Counter: 0

*   Exec_Master_Log_Pos: 27664  Relay_Log_Space: 548*
  Until_Condition: None
   Until_Log_File:
Until_Log_Pos: 0
   Master_SSL_Allowed: Yes
   Master_SSL_CA_File: /opt/mysql/ca.crt
   Master_SSL_CA_Path:
  Master_SSL_Cert: /opt/mysql/mysql-slave.crt
Master_SSL_Cipher:
   Master_SSL_Key: /opt/mysql/mysql-slave.key
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
   Last_SQL_Errno: 0
   Last_SQL_Error:
  Replicate_Ignore_Server_Ids:
 Master_Server_Id: 1
1 row in set (0.00 sec)


Thanks so much for all your help! This was very sanity-saving. :)

Best!
Tim


On Thu, Mar 12, 2015 at 10:42 AM, Alberto Rivera Laporte <
arlapo...@gmail.com> wrote:

> On Thu, Mar 12, 2015 at 8:57 AM Tim Dunphy  wrote:
>
> > Hey everybody,
> >
> >  I'm trying to get mysql master/slave replication to work under SSL. I've
> > created the certs for both the slave and the master. I've configured the
> > master and slave my.cnf. And it does appear that replication is actually
> > working.
> >
> > Master is actually MariaDB (version 5.5.41-MariaDB-log, and the slave is
> > MySQL (version 5.5.41-log).
> >
> > But there are two issues I'd like to resolve. One is that SSL appears to
> be
> > disabled.
> >
> > If I look at both the master and the slave and do a 'show variables'
> > command, I can see that it's recognizing the certs. But the
> 'have_openssl'
> > and 'have_ssl' variables are showing as DISABLED.
> >
> > Watch, on the master:
> >
> > MariaDB [(none)]> show variables like '%ssl%';
> > +---++
> > | Variable_name | Value  |
> > +---++
> > | have_openssl  | DISABLED   |
> > | have_ssl  | DISABLED   |
> > | ssl_ca| /etc/pki/CA/certs/ca.crt   |
> > | ssl_capath||
> > | ssl_cert  | /etc/pki/tls/certs/mysql.crt   |
> > | ssl_cipher||
> > | ssl_key   | /etc/pki/tls/private/mysql.key |
> > +---++
> > 7 rows in set (0.01 sec)
> >
> > On the slave:
> >
> > mysql> show variables like '%ssl%';
> > +---+--+
> > | Variable_name | Value|
> > +---+--+
> > | have_openssl  | DISABLED |
> > | have_ssl  | DISABLED |
> > | ssl_ca| /etc/pki/CA/certs/ca.crt |
> > | ssl_capath|  

Re: [CentOS] mysql replication - problems

[Alberto Rivera Laporte]

On Thu, Mar 12, 2015 at 10:49 AM Tim Dunphy  wrote:

> >
> > No: /etc/pki/CA should NOT be group writeable. Ditto for
> > /etc/pki/tls/cernts and private
>
>
> Ok, yeah I can understand that. I'll correct it. Still need a way to get
> SSL enabled however. Any suggestions there?
>
>
I totally misread your configuration options and locations on your original
post, my apologies.  You indeed had what should be considered correct
locations for the ssl cert and key files.  So if it still not functioning I
would defer you to the MysQL mailing list / support channels to see if they
can assist you in figuring out any further.

Best of luck once again.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[Alberto Rivera Laporte]

On Thu, Mar 12, 2015 at 10:49 AM Tim Dunphy  wrote:

> >
> > No: /etc/pki/CA should NOT be group writeable. Ditto for
> > /etc/pki/tls/cernts and private
>



I agree - Sorry I did not mean to imply that the directory permissions on
/etc/pki/CA should be modified.  However it was mentioned it as a probable
issue for his ssl configuration on mysql/mariadb not .


>
> Ok, yeah I can understand that. I'll correct it. Still need a way to get
> SSL enabled however. Any suggestions there?
>
> Thanks
> Tim
>
>
Here's a test I did on a vanilla Centos 7 with mariadb from the stock
Centos repo, first with the locations you chose on your non-working
scenario, and second with an alternate location chosen for the ssl key and
cert files.


First example:

-- /etc/my.cnf.d/server.cnf

[mysqld]
ssl-key=/etc/pki/CA/private/test-key.pem
ssl-cert=/etc/pki/CA/certs/test-cert.pem


Here were the errors on startup from having the ssl files in that location
which subsequently end up ssl being disabled, much like you are
experiencing.

-- /var/log/mariadb/mariadb.log

150312 13:37:51  InnoDB: Waiting for the background threads to start
150312 13:37:52 Percona XtraDB (http://www.percona.com) 5.5.40-MariaDB-36.1
started; log sequence number 0
150312 13:37:52 [Note] Plugin 'FEEDBACK' is disabled.
SSL error: Unable to get private key from '/etc/pki/CA/private/test-key.pem'
150312 13:37:52 [Warning] Failed to setup SSL
150312 13:37:52 [Warning] SSL error: Unable to get private key
150312 13:37:52 [Note] Server socket created on IP: '0.0.0.0'.
150312 13:37:52 [Note] Event Scheduler: Loaded 0 events
150312 13:37:52 [Note] /usr/libexec/mysqld: ready for connections.
Version: '5.5.41-MariaDB'  socket: '/var/lib/mysql/mysql.sock'  port: 3306
MariaDB Server


MariaDB [(none)]> show variables like '%ssl%';
+---+--+
| Variable_name | Value|
+---+--+
| have_openssl  | DISABLED |
| have_ssl  | DISABLED



Second example:

Here is the results of the ssl key and cert files being moved to another
specific location where the mysql user can read those files and restarting
the service:

-- /etc/my.cnf.d/server.cnf

[mysqld]
ssl-key=/etc/mysql/test-key.pem
ssl-cert=/etc/mysql/test-cert.pem



-- /var/log/mariadb/mariadb.log
150312 13:48:19  InnoDB: Waiting for the background threads to start
150312 13:48:20 Percona XtraDB (http://www.percona.com) 5.5.40-MariaDB-36.1
started; log sequence number 1597945
150312 13:48:20 [Note] Plugin 'FEEDBACK' is disabled.
150312 13:48:20 [Note] Server socket created on IP: '0.0.0.0'.
150312 13:48:20 [Note] Event Scheduler: Loaded 0 events
150312 13:48:20 [Note] /usr/libexec/mysqld: ready for connections.
Version: '5.5.41-MariaDB'  socket: '/var/lib/mysql/mysql.sock'  port: 3306
MariaDB Server



MariaDB [(none)]> show variables like '%ssl%';
+---+--+
| Variable_name | Value|
+---+--+
| have_openssl  | YES  |
| have_ssl  | YES  |




If you want to maintain consistency and store the certificates in
/etc/pki/* these should be the locations where they can be stored:

#certs, including intermediates
/etc/pki/tls/certs

# keys
/etc/pki/tls/private


Here were the result of a third test:

-- /etc/my.cnf.d/server.cnf

[mysqld]

ssl-key=/etc/pki/tls/private/test-key.pem
ssl-cert=/etc/pki/tls/certs/test-cert.pem



MariaDB [(none)]> show variables like '%ssl%';
+---+---+
| Variable_name | Value |
+---+---+
| have_openssl  | YES   |
| have_ssl  | YES   |
| ssl_ca|   |
| ssl_capath|   |
| ssl_cert  | /etc/pki/tls/certs/test-cert.pem  |
| ssl_cipher|   |
| ssl_key   | /etc/pki/tls/private/test-key.pem |
+---+---+


Best,
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[Valeri Galtsev]

On Thu, March 12, 2015 10:40 am, m.r...@5-cent.us wrote:
> Tim Dunphy wrote:
>>>
>>> The mysqld process runs as the mysql user.  It's parent which is the
>>> mysqld_safe runs as the root user.   That being said the mysql user
>>> needs to have at least read permission to the locations where the ssl
> files
>>> are located.   By default on Centos the /etc/pki/CA/private directory
>>> has
>>> its directory permissions to only allow the root user.  If the mysql
>>> user
>>> cannot read all ssl files SSL will not work.
> 
>> Thanks for your reply! That answer actually makes complete sense. Ok, so
>> here is what I tried, so far without success. I gave the mysql group
>> ownership of all related directories. And changed group permissions so
>> that group can access them:
>>
>> [root@web2:/etc] #ls -ld /etc/pki/CA
>> drwxrwxr-x. 6 root mysql 4096 Jan 20 15:58 /etc/pki/CA
>> [root@web2:/etc] #ls -ld /etc/pki/tls/{private,certs}
>> drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/certs
>> drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/private
>>
>>  Restarted the mariadb service. And when I took another look at the SSL
>> variable, it's still showing that SSL is not enabled:
> 
> Some of those will *not* work. For example, you will has ssh issues
> yourself is ~/.ssh is *anything* other than 700.
>
> No: /etc/pki/CA should NOT be group writeable. Ditto for
> /etc/pki/tls/cernts and private.
>

I have my doubts about permissions on /etc/pki/tls/private and on private
key inside it as well. Somebody hopefully will correct me as I don't know
how it is implemented in mysql/mariadb, but I assume sanity. And sanity
suggests that the first process (mysqld_safe) that runs as root reads
private key (and likely certificate), then passes private key to the child
process(es) which runs as regular user that is not able to read private
key, but gets it from parent proces. My assumption comes from what apache
is doing (only apache used droppriv).

I would (roll perms/ownership) back to default, and try to check locally
using openssl whether daemon is using ssl/cert/key, maybe start mysql
daemon in debugger to see what is going on with reading private key. I
would also think of other reasons why your instance of mysql (or mariadb)
could not be able to use _your_ key and cert, see, e.g.:

http://forums.mysql.com/read.php?11,400856,401127

(your case may be different, I would just try think wider, but maybe
debugger will give you the direct lead).

Valeri


Valeri Galtsev
Sr System Administrator
Department of Astronomy and Astrophysics
Kavli Institute for Cosmological Physics
University of Chicago
Phone: 773-702-4247

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[Tim Dunphy]

>
> No: /etc/pki/CA should NOT be group writeable. Ditto for
> /etc/pki/tls/cernts and private


Ok, yeah I can understand that. I'll correct it. Still need a way to get
SSL enabled however. Any suggestions there?

Thanks
Tim

On Thu, Mar 12, 2015 at 11:40 AM,  wrote:

> Tim Dunphy wrote:
> >>
> >> The mysqld process runs as the mysql user.  It's parent which is the
> >> mysqld_safe runs as the root user.   That being said the mysql user
> >> needs to have at least read permission to the locations where the ssl
> files
> >> are located.   By default on Centos the /etc/pki/CA/private directory
> has
> >> its directory permissions to only allow the root user.  If the mysql
> user
> >> cannot read all ssl files SSL will not work.
> 
> > Thanks for your reply! That answer actually makes complete sense. Ok, so
> > here is what I tried, so far without success. I gave the mysql group
> > ownership of all related directories. And changed group permissions so
> > that group can access them:
> >
> > [root@web2:/etc] #ls -ld /etc/pki/CA
> > drwxrwxr-x. 6 root mysql 4096 Jan 20 15:58 /etc/pki/CA
> > [root@web2:/etc] #ls -ld /etc/pki/tls/{private,certs}
> > drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/certs
> > drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/private
> >
> >  Restarted the mariadb service. And when I took another look at the SSL
> > variable, it's still showing that SSL is not enabled:
> 
> Some of those will *not* work. For example, you will has ssh issues
> yourself is ~/.ssh is *anything* other than 700.
>
> No: /etc/pki/CA should NOT be group writeable. Ditto for
> /etc/pki/tls/cernts and private.
>
>mark
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
GPG me!!

gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[m . roth]

Tim Dunphy wrote:
>>
>> The mysqld process runs as the mysql user.  It's parent which is the
>> mysqld_safe runs as the root user.   That being said the mysql user
>> needs to have at least read permission to the locations where the ssl
files
>> are located.   By default on Centos the /etc/pki/CA/private directory has
>> its directory permissions to only allow the root user.  If the mysql user
>> cannot read all ssl files SSL will not work.

> Thanks for your reply! That answer actually makes complete sense. Ok, so
> here is what I tried, so far without success. I gave the mysql group
> ownership of all related directories. And changed group permissions so
> that group can access them:
>
> [root@web2:/etc] #ls -ld /etc/pki/CA
> drwxrwxr-x. 6 root mysql 4096 Jan 20 15:58 /etc/pki/CA
> [root@web2:/etc] #ls -ld /etc/pki/tls/{private,certs}
> drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/certs
> drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/private
>
>  Restarted the mariadb service. And when I took another look at the SSL
> variable, it's still showing that SSL is not enabled:

Some of those will *not* work. For example, you will has ssh issues
yourself is ~/.ssh is *anything* other than 700.

No: /etc/pki/CA should NOT be group writeable. Ditto for
/etc/pki/tls/cernts and private.

   mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql replication - problems

[Tim Dunphy]

>
> The mysqld process runs as the mysql user.  It's parent which is the
> mysqld_safe runs as the root user.   That being said the mysql user needs
> to have at least read permission to the locations where the ssl files are
>


> located.   By default on Centos the /etc/pki/CA/private directory has its
> directory permissions to only allow the root user.  If the mysql user
> cannot read all ssl files SSL will not work.



> 2.  Regarding your replication specific user not being able to connect to
> the master.   It may not work until SSL is fully working since you
> specifically stated to require and SSL connection.   So the symptom of this
> might be resolved when SSL is fixed.


Thanks for your reply! That answer actually makes complete sense. Ok, so
here is what I tried, so far without success. I gave the mysql group
ownership of all related directories. And changed group permissions so that
group can access them:

[root@web2:/etc] #ls -ld /etc/pki/CA
drwxrwxr-x. 6 root mysql 4096 Jan 20 15:58 /etc/pki/CA
[root@web2:/etc] #ls -ld /etc/pki/tls/{private,certs}
drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/certs
drwxrwxr-x. 2 root mysql 4096 Mar 11 22:57 /etc/pki/tls/private

 Restarted the mariadb service. And when I took another look at the SSL
variable, it's still showing that SSL is not enabled:

MariaDB [(none)]> show variables like '%ssl%';
+---++
| Variable_name | Value  |
+---++
| have_openssl  | DISABLED   |
| have_ssl  | DISABLED   |
| ssl_ca| /etc/pki/CA/certs/ca.crt   |
| ssl_capath||
| ssl_cert  | /etc/pki/tls/certs/mysql.crt   |
| ssl_cipher||
| ssl_key   | /etc/pki/tls/private/mysql.key |
+---++
7 rows in set (0.00 sec)

Do you think I'm going about this in the right way? Is there anything else
I can try to resolve this?

Thanks
Tim

On Thu, Mar 12, 2015 at 10:42 AM, Alberto Rivera Laporte <
arlapo...@gmail.com> wrote:

> On Thu, Mar 12, 2015 at 8:57 AM Tim Dunphy  wrote:
>
> > Hey everybody,
> >
> >  I'm trying to get mysql master/slave replication to work under SSL. I've
> > created the certs for both the slave and the master. I've configured the
> > master and slave my.cnf. And it does appear that replication is actually
> > working.
> >
> > Master is actually MariaDB (version 5.5.41-MariaDB-log, and the slave is
> > MySQL (version 5.5.41-log).
> >
> > But there are two issues I'd like to resolve. One is that SSL appears to
> be
> > disabled.
> >
> > If I look at both the master and the slave and do a 'show variables'
> > command, I can see that it's recognizing the certs. But the
> 'have_openssl'
> > and 'have_ssl' variables are showing as DISABLED.
> >
> > Watch, on the master:
> >
> > MariaDB [(none)]> show variables like '%ssl%';
> > +---++
> > | Variable_name | Value  |
> > +---++
> > | have_openssl  | DISABLED   |
> > | have_ssl  | DISABLED   |
> > | ssl_ca| /etc/pki/CA/certs/ca.crt   |
> > | ssl_capath||
> > | ssl_cert  | /etc/pki/tls/certs/mysql.crt   |
> > | ssl_cipher||
> > | ssl_key   | /etc/pki/tls/private/mysql.key |
> > +---++
> > 7 rows in set (0.01 sec)
> >
> > On the slave:
> >
> > mysql> show variables like '%ssl%';
> > +---+--+
> > | Variable_name | Value|
> > +---+--+
> > | have_openssl  | DISABLED |
> > | have_ssl  | DISABLED |
> > | ssl_ca| /etc/pki/CA/certs/ca.crt |
> > | ssl_capath|  |
> > | ssl_cert  | /etc/pki/tls/certs/mysql-slave.crt   |
> > | ssl_cipher|  |
> > | ssl_key   | /etc/pki/tls/private/mysql-slave.key |
> > +---+--+
> > 7 rows in set (0.00 sec)
> >
> > And yet I clearly have SSL enabled in both configurations.
> >
> > In the master mysql configuration I have:
> >
> > [root@web2:~] #cat /etc/my.cnf
> > [mysqld]
> > datadir=/var/lib/mysql
> > socket=/var/lib/mysql/mysql.sock
> > symbolic-links=0
> > *ssl*
> > *ssl-ca=/etc/pki/CA/certs/ca.crt*
> > *ssl-cert=/etc/pki/tls/certs/mysql.crt*
> > *ssl-key=/etc/pki/tls/private/mysql.key*
> > server-id   = 1
> > log_bin = /var/log/mariadb/mysql-bin.log
> > expire_logs_days= 10
> > max_binlog_size = 100M
> > binlog_do_db= jokefire
> >
>

Re: [CentOS] mysql replication - problems

[Alberto Rivera Laporte]

On Thu, Mar 12, 2015 at 8:57 AM Tim Dunphy  wrote:

> Hey everybody,
>
>  I'm trying to get mysql master/slave replication to work under SSL. I've
> created the certs for both the slave and the master. I've configured the
> master and slave my.cnf. And it does appear that replication is actually
> working.
>
> Master is actually MariaDB (version 5.5.41-MariaDB-log, and the slave is
> MySQL (version 5.5.41-log).
>
> But there are two issues I'd like to resolve. One is that SSL appears to be
> disabled.
>
> If I look at both the master and the slave and do a 'show variables'
> command, I can see that it's recognizing the certs. But the 'have_openssl'
> and 'have_ssl' variables are showing as DISABLED.
>
> Watch, on the master:
>
> MariaDB [(none)]> show variables like '%ssl%';
> +---++
> | Variable_name | Value  |
> +---++
> | have_openssl  | DISABLED   |
> | have_ssl  | DISABLED   |
> | ssl_ca| /etc/pki/CA/certs/ca.crt   |
> | ssl_capath||
> | ssl_cert  | /etc/pki/tls/certs/mysql.crt   |
> | ssl_cipher||
> | ssl_key   | /etc/pki/tls/private/mysql.key |
> +---++
> 7 rows in set (0.01 sec)
>
> On the slave:
>
> mysql> show variables like '%ssl%';
> +---+--+
> | Variable_name | Value|
> +---+--+
> | have_openssl  | DISABLED |
> | have_ssl  | DISABLED |
> | ssl_ca| /etc/pki/CA/certs/ca.crt |
> | ssl_capath|  |
> | ssl_cert  | /etc/pki/tls/certs/mysql-slave.crt   |
> | ssl_cipher|  |
> | ssl_key   | /etc/pki/tls/private/mysql-slave.key |
> +---+--+
> 7 rows in set (0.00 sec)
>
> And yet I clearly have SSL enabled in both configurations.
>
> In the master mysql configuration I have:
>
> [root@web2:~] #cat /etc/my.cnf
> [mysqld]
> datadir=/var/lib/mysql
> socket=/var/lib/mysql/mysql.sock
> symbolic-links=0
> *ssl*
> *ssl-ca=/etc/pki/CA/certs/ca.crt*
> *ssl-cert=/etc/pki/tls/certs/mysql.crt*
> *ssl-key=/etc/pki/tls/private/mysql.key*
> server-id   = 1
> log_bin = /var/log/mariadb/mysql-bin.log
> expire_logs_days= 10
> max_binlog_size = 100M
> binlog_do_db= jokefire
>
> [mysqld_safe]
> log-error=/var/log/mariadb/mariadb.log
> pid-file=/var/run/mariadb/mariadb.pid
>
> On the mysql slave:
>
> [root@ops:~] #cat /etc/my.cnf
> [mysqld]
> # Settings user and group are ignored when systemd is used (fedora >= 15).
> # If you need to run mysqld under different user or group,
> # customize your systemd unit file for mysqld according to the
> # instructions in http://fedoraproject.org/wiki/Systemd
> user=mysql 
> *ssl*
> *server-id=2*
>
> *replicate-do-db=jokefire*
> *ssl-ca=/etc/pki/CA/certs/ca.crt*
> *ssl-cert=/etc/pki/tls/certs/mysql-slave.crt*
> *ssl-key=/etc/pki/tls/private/mysql-slave.key*
> thread_cache_size = 4
>
> datadir=/var/lib/mysql
> socket=/var/lib/mysql/mysql.sock
> symbolic-links=0
> ;plugin-load=rpl_semi_sync_master=semisync_master.so
> ;plugin-load=rpl_semi_sync_slave=semisync_slave.so
> ;rpl_semi_sync_master_enabled=1
> ;rpl_semi_sync_master_timeout=10
> ;rpl_semi_sync_slave_enabled=1
> ;performance_schema
> query_cache_size = 8MB
> innodb_buffer_pool_size = 199M
> general_log_file=/var/log/mysql/mysql.log
> general_log=1
> log-error=/var/log/mysql/mysql_error_log
> log-slow-queries=/var/log/mysql/mysql_slow_log
> wait_timeout = 86400
>
> [mysqld_safe]
> general_log_file=/var/log/mysql/mysql.log
> general_log=1
> log-error=/var/log/mysql/mysql_error_log
> log-slow-queries=/var/log/mysql/mysql_slow_log
> pid-file=/var/run/mysqld/mysqld.pid
> innodb_buffer_pool_size = 199M
> wait_timeout = 28800
> interactive_timeout = 28800
> master-connect-retry=60
>
> So my first question is, why is SSL not enabled in either database? I
> restarted the service on both machines before taking a look at the
> variables.
>
> The next problem I'm having is that I can't seem to get the replication
> user to connect. I had to use an account with more privileges (grant all)
> in order to connect from the slave to the master.
>
> I used this grant on the master to try and setup the replication user:
>
> GRANT REPLICATION SLAVE ON *.* TO 'jf_slave'@'ops.somewhere.com'
> IDENTIFIED
> BY 'secret' REQUIRE SSL;
>
> Then back on the slave I used this command to connect the slave to the
> master:
>
> mysql> CHANGE MASTER TO MASTER_HOST='web2.somewhere.com',
> MASTER_USER='jf_slave', MASTER_PASSWORD='secret',
> M

[CentOS] mysql replication - problems

[Tim Dunphy]

Hey everybody,

 I'm trying to get mysql master/slave replication to work under SSL. I've
created the certs for both the slave and the master. I've configured the
master and slave my.cnf. And it does appear that replication is actually
working.

Master is actually MariaDB (version 5.5.41-MariaDB-log, and the slave is
MySQL (version 5.5.41-log).

But there are two issues I'd like to resolve. One is that SSL appears to be
disabled.

If I look at both the master and the slave and do a 'show variables'
command, I can see that it's recognizing the certs. But the 'have_openssl'
and 'have_ssl' variables are showing as DISABLED.

Watch, on the master:

MariaDB [(none)]> show variables like '%ssl%';
+---++
| Variable_name | Value  |
+---++
| have_openssl  | DISABLED   |
| have_ssl  | DISABLED   |
| ssl_ca| /etc/pki/CA/certs/ca.crt   |
| ssl_capath||
| ssl_cert  | /etc/pki/tls/certs/mysql.crt   |
| ssl_cipher||
| ssl_key   | /etc/pki/tls/private/mysql.key |
+---++
7 rows in set (0.01 sec)

On the slave:

mysql> show variables like '%ssl%';
+---+--+
| Variable_name | Value|
+---+--+
| have_openssl  | DISABLED |
| have_ssl  | DISABLED |
| ssl_ca| /etc/pki/CA/certs/ca.crt |
| ssl_capath|  |
| ssl_cert  | /etc/pki/tls/certs/mysql-slave.crt   |
| ssl_cipher|  |
| ssl_key   | /etc/pki/tls/private/mysql-slave.key |
+---+--+
7 rows in set (0.00 sec)

And yet I clearly have SSL enabled in both configurations.

In the master mysql configuration I have:

[root@web2:~] #cat /etc/my.cnf
[mysqld]
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
symbolic-links=0
*ssl*
*ssl-ca=/etc/pki/CA/certs/ca.crt*
*ssl-cert=/etc/pki/tls/certs/mysql.crt*
*ssl-key=/etc/pki/tls/private/mysql.key*
server-id   = 1
log_bin = /var/log/mariadb/mysql-bin.log
expire_logs_days= 10
max_binlog_size = 100M
binlog_do_db= jokefire

[mysqld_safe]
log-error=/var/log/mariadb/mariadb.log
pid-file=/var/run/mariadb/mariadb.pid

On the mysql slave:

[root@ops:~] #cat /etc/my.cnf
[mysqld]
# Settings user and group are ignored when systemd is used (fedora >= 15).
# If you need to run mysqld under different user or group,
# customize your systemd unit file for mysqld according to the
# instructions in http://fedoraproject.org/wiki/Systemd
user=mysql
*ssl*
*server-id=2*

*replicate-do-db=jokefire*
*ssl-ca=/etc/pki/CA/certs/ca.crt*
*ssl-cert=/etc/pki/tls/certs/mysql-slave.crt*
*ssl-key=/etc/pki/tls/private/mysql-slave.key*
thread_cache_size = 4

datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
symbolic-links=0
;plugin-load=rpl_semi_sync_master=semisync_master.so
;plugin-load=rpl_semi_sync_slave=semisync_slave.so
;rpl_semi_sync_master_enabled=1
;rpl_semi_sync_master_timeout=10
;rpl_semi_sync_slave_enabled=1
;performance_schema
query_cache_size = 8MB
innodb_buffer_pool_size = 199M
general_log_file=/var/log/mysql/mysql.log
general_log=1
log-error=/var/log/mysql/mysql_error_log
log-slow-queries=/var/log/mysql/mysql_slow_log
wait_timeout = 86400

[mysqld_safe]
general_log_file=/var/log/mysql/mysql.log
general_log=1
log-error=/var/log/mysql/mysql_error_log
log-slow-queries=/var/log/mysql/mysql_slow_log
pid-file=/var/run/mysqld/mysqld.pid
innodb_buffer_pool_size = 199M
wait_timeout = 28800
interactive_timeout = 28800
master-connect-retry=60

So my first question is, why is SSL not enabled in either database? I
restarted the service on both machines before taking a look at the
variables.

The next problem I'm having is that I can't seem to get the replication
user to connect. I had to use an account with more privileges (grant all)
in order to connect from the slave to the master.

I used this grant on the master to try and setup the replication user:

GRANT REPLICATION SLAVE ON *.* TO 'jf_slave'@'ops.somewhere.com' IDENTIFIED
BY 'secret' REQUIRE SSL;

Then back on the slave I used this command to connect the slave to the
master:

mysql> CHANGE MASTER TO MASTER_HOST='web2.somewhere.com',
MASTER_USER='jf_slave', MASTER_PASSWORD='secret',
MASTER_LOG_FILE='mysql-bin.02', MASTER_LOG_POS=34697, MASTER_SSL=1,
MASTER_SSL_CA = '/etc/pki/CA/certs/ca.crt', MASTER_SSL_CERT =
'/etc/pki/tls/certs/mysql.crt', MASTER_SSL_KEY =
'/etc/pki/tls/private/mysql.key';

And when I start up the slave I see that there's a problem connecting from
the slave to the master:

mysql> show slave status \G
*

Re: [CentOS] MySQL - replication - how to restore master?

[David Beveridge]

It's quite simple really.
Just make your master a slave of your slave.
It's called Multi Master.

see http://mysql-mmm.org/



On Wed, Aug 13, 2014 at 9:53 PM, John Horne 
wrote:

> Hello,
>
> We have MySQL running as a master which is replicating to a single slave
> server. We are, however, considering what is required when a 'disaster'
> of some sort happens to either server. By disaster, this could be some
> event which requires the entire server to be rebuilt, and which would
> usually include restoring from nightly backups directories such as
> '/var/lib/mysql' and '/var/log/mysql' (as set in our my.cnf file). It
> could also refer to an event which only affects the mysql service, but
> requires us to stop the mysql master service. This may involve
> reinstalling the mysql package, and, again, restoring the
> '/var/lib/mysql' and '/var/log/mysql' directories.
>
> In the case of losing the slave server, we have found instructions for
> rebuilding the slave database and restarting replication using a
> mysqldump backup taken from the master server. We have tested this and
> it works fine.
>
> However, I am having trouble finding out what to do should we lose the
> master server. Typically mysqldump backups of the master are done
> overnight, so a failure during the day would mean that the slave is
> ahead of the master backup. So this poses two questions:
>
> 1) If the master fails, and we perform (at that time) a mysqldump of the
> slave, we could import the data into the master, but what commands do we
> need to tell the master (and slave?) to start replication based on the
> imported data? As far as I can gather the master replication data is
> held in the '/var/log/mysql' directory (in our case) in the bin log
> files, and these would typically be restored after a disaster.
>
> 2) If the master fails and we import the overnight backup data, what
> commands do we then need to issue on the master and slave to restart
> replication from the imported data? In particular, on the master do we
> just delete the bin log files and let replication start afresh? And on
> the slave, which at that time would be ahead of the master, how do we
> sort out the replication? Do we drop the existing database and import
> the backup data into the slave as well, so that both the master and
> slave start with the same data?
>
>
>
>
> Thanks,
>
> John.
>
> --
> John Horne   Tel: +44 (0)1752 587287
> Plymouth University, UK
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL - replication - how to restore master?

[Denniston, Todd A CIV NAVSURFWARCENDIV Crane]

> -Original Message-
> From: John Horne [mailto:john.ho...@plymouth.ac.uk]
> Sent: Wednesday, August 13, 2014 7:53 AM
> To: CentOS list
> Subject: [CentOS] MySQL - replication - how to restore master?
> 
> Hello,
> 

> However, I am having trouble finding out what to do should we lose the
> master server. Typically mysqldump backups of the master are done
> overnight, so a failure during the day would mean that the slave is
> ahead of the master backup. So this poses two questions:
> 


Note: I have not been running MySQL or HA systems for a very good while now.  
Salt appropriately.

Perhaps the following URL can help:
http://dev.mysql.com/doc/refman/5.0/en/replication-solutions-switch.html

I believe  useful sets of key words for the searching the web would be:
mysql master slave promotion
and
mysqldump  slave

http://dev.mysql.com/doc/refman/5.0/en/replication-solutions-backups-mysqldump.html

Even when this disclaimer is not here:
I am not a contracting officer. I do not have authority to make or modify the 
terms of any contract.


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL - replication - how to restore master?

[John Horne]

Hello,

We have MySQL running as a master which is replicating to a single slave
server. We are, however, considering what is required when a 'disaster'
of some sort happens to either server. By disaster, this could be some
event which requires the entire server to be rebuilt, and which would
usually include restoring from nightly backups directories such as
'/var/lib/mysql' and '/var/log/mysql' (as set in our my.cnf file). It
could also refer to an event which only affects the mysql service, but
requires us to stop the mysql master service. This may involve
reinstalling the mysql package, and, again, restoring the
'/var/lib/mysql' and '/var/log/mysql' directories.

In the case of losing the slave server, we have found instructions for
rebuilding the slave database and restarting replication using a
mysqldump backup taken from the master server. We have tested this and
it works fine.

However, I am having trouble finding out what to do should we lose the
master server. Typically mysqldump backups of the master are done
overnight, so a failure during the day would mean that the slave is
ahead of the master backup. So this poses two questions:

1) If the master fails, and we perform (at that time) a mysqldump of the
slave, we could import the data into the master, but what commands do we
need to tell the master (and slave?) to start replication based on the
imported data? As far as I can gather the master replication data is
held in the '/var/log/mysql' directory (in our case) in the bin log
files, and these would typically be restored after a disaster.

2) If the master fails and we import the overnight backup data, what
commands do we then need to issue on the master and slave to restart
replication from the imported data? In particular, on the master do we
just delete the bin log files and let replication start afresh? And on
the slave, which at that time would be ahead of the master, how do we
sort out the replication? Do we drop the existing database and import
the backup data into the slave as well, so that both the master and
slave start with the same data?




Thanks,

John.

-- 
John Horne   Tel: +44 (0)1752 587287
Plymouth University, UK

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Ljubomir Ljubojevic]

On 02/13/2014 09:21 PM, Ljubomir Ljubojevic wrote:
> On 02/13/2014 09:18 PM, Ljubomir Ljubojevic wrote:
>> On 02/13/2014 08:49 PM, Les Mikesell wrote:
>>> On Thu, Feb 13, 2014 at 1:06 PM, Ljubomir Ljubojevic  
>>> wrote:

 I have a strange problem.

 I installed cacti, and set it up.

 When I click on the Plugins Management, I noticed Error
 "FATAL: Database or Table does not exist"

 Further testing revealed that at the moment MySQL can not create
 temporary files in /tmp. It is same for ANY database!

 SELinux is disabled.
 I logged to mysql as root and error is same.

 Command used is:
 CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;


 Error is:
 ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)

 What could be the problem, any idea how to proceed?
>>>
>>> http://bugs.mysql.com/bug.php?id=60574
>>> The error may be on the source of the copy (bak_menu.frm) not in tmp.
>>>
>>
>> This is interesting.
>>
>> I tested on cacti database and on 2 Joomla databases with same results.
>> But there might be something there.
>>
>> Database changed
>> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE jos_menu;
>> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_1.frm' (errno: 9)
>> mysql> CREATE TEMPORARY TABLE barney (SELECT * FROM jos_menu WHERE 1=0);
>> Query OK, 0 rows affected (0.01 sec)
>> Records: 0  Duplicates: 0  Warnings: 0
>>
>> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugintest LIKE jos_menu;
>> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_3.frm' (errno: 9)
>> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS barney LIKE jos_menu;
>> Query OK, 0 rows affected, 1 warning (0.00 sec)
>>
>> Changed name of table to create produced result, but with warning.
>>
>
> It looks like "IF NOT EXISTS" might be creating problems.
>

Nope, that is not it, seams "LIKE" is the problem.


-- 
Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Ljubomir Ljubojevic]

On 02/13/2014 09:18 PM, Ljubomir Ljubojevic wrote:
> On 02/13/2014 08:49 PM, Les Mikesell wrote:
>> On Thu, Feb 13, 2014 at 1:06 PM, Ljubomir Ljubojevic  wrote:
>>>
>>> I have a strange problem.
>>>
>>> I installed cacti, and set it up.
>>>
>>> When I click on the Plugins Management, I noticed Error
>>> "FATAL: Database or Table does not exist"
>>>
>>> Further testing revealed that at the moment MySQL can not create
>>> temporary files in /tmp. It is same for ANY database!
>>>
>>> SELinux is disabled.
>>> I logged to mysql as root and error is same.
>>>
>>> Command used is:
>>> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;
>>>
>>>
>>> Error is:
>>> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)
>>>
>>> What could be the problem, any idea how to proceed?
>>
>> http://bugs.mysql.com/bug.php?id=60574
>> The error may be on the source of the copy (bak_menu.frm) not in tmp.
>>
>
> This is interesting.
>
> I tested on cacti database and on 2 Joomla databases with same results.
> But there might be something there.
>
> Database changed
> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE jos_menu;
> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_1.frm' (errno: 9)
> mysql> CREATE TEMPORARY TABLE barney (SELECT * FROM jos_menu WHERE 1=0);
> Query OK, 0 rows affected (0.01 sec)
> Records: 0  Duplicates: 0  Warnings: 0
>
> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugintest LIKE jos_menu;
> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_3.frm' (errno: 9)
> mysql> CREATE TEMPORARY TABLE IF NOT EXISTS barney LIKE jos_menu;
> Query OK, 0 rows affected, 1 warning (0.00 sec)
>
> Changed name of table to create produced result, but with warning.
>

It looks like "IF NOT EXISTS" might be creating problems.



-- 
Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Ljubomir Ljubojevic]

On 02/13/2014 08:49 PM, Les Mikesell wrote:
> On Thu, Feb 13, 2014 at 1:06 PM, Ljubomir Ljubojevic  wrote:
>>
>> I have a strange problem.
>>
>> I installed cacti, and set it up.
>>
>> When I click on the Plugins Management, I noticed Error
>> "FATAL: Database or Table does not exist"
>>
>> Further testing revealed that at the moment MySQL can not create
>> temporary files in /tmp. It is same for ANY database!
>>
>> SELinux is disabled.
>> I logged to mysql as root and error is same.
>>
>> Command used is:
>> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;
>>
>>
>> Error is:
>> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)
>>
>> What could be the problem, any idea how to proceed?
>
> http://bugs.mysql.com/bug.php?id=60574
> The error may be on the source of the copy (bak_menu.frm) not in tmp.
>

This is interesting.

I tested on cacti database and on 2 Joomla databases with same results. 
But there might be something there.

Database changed
mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE jos_menu;
ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_1.frm' (errno: 9)
mysql> CREATE TEMPORARY TABLE barney (SELECT * FROM jos_menu WHERE 1=0);
Query OK, 0 rows affected (0.01 sec)
Records: 0  Duplicates: 0  Warnings: 0

mysql> CREATE TEMPORARY TABLE IF NOT EXISTS plugintest LIKE jos_menu;
ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_3.frm' (errno: 9)
mysql> CREATE TEMPORARY TABLE IF NOT EXISTS barney LIKE jos_menu;
Query OK, 0 rows affected, 1 warning (0.00 sec)

Changed name of table to create produced result, but with warning.



-- 
Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Ljubomir Ljubojevic]

On 02/13/2014 08:17 PM, Clint Dilks wrote:
> Hi,
>
> Have you verified that permissions on /tmp appear correct ?
>
> An example from one of my systems
>
>
> # ls -lsahd /tmp
> 4.0K drwxrwxrwt. 3 root root 4.0K Feb 14 07:01 /tm
>
> So permissions 1777
>

Already checked.

[root@chiron ~]# ls -lsahd /tmp
4.0K drwxrwxrwt. 17 root root 4.0K Feb 13 20:55 /tmp


> Perhaps test creating a file in /tmp as root and as a non root user and let
> us know the results that way we may be able to help further.
>

There is enough space, and I successfully created files as root, mysql 
and plnet user (domain owner).

[root@chiron ~]# df -h
Filesystem 
   Size  Used Avail Use% Mounted on
/dev/mapper/vg_chiron-LV_CHIRON_ROOT   25G   15G  8.3G  65% /
tmpfs 
   1.3G  676K  1.3G   1% /dev/shm
/dev/md0  485M  197M  263M  43% /boot
/dev/mapper/vg_chiron-LG_CHIRON_EXTRA 1.8T  1.6T  143G  92% /extra


>
> On Fri, Feb 14, 2014 at 8:06 AM, Ljubomir Ljubojevic wrote:
>
>> Hi.
>>
>> I have a strange problem.
>>
>> I installed cacti, and set it up.
>>
>> When I click on the Plugins Management, I noticed Error
>> "FATAL: Database or Table does not exist"
>>
>> Further testing revealed that at the moment MySQL can not create
>> temporary files in /tmp. It is same for ANY database!
>>
>> SELinux is disabled.
>> I logged to mysql as root and error is same.
>>
>> Command used is:
>> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;
>>
>>
>> Error is:
>> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)
>>
>> What could be the problem, any idea how to proceed?
>>
>> I do not think I edited any MySQL setting.
>>


I forgot to mention that I use Virtualmin GPL for ~10 domains, but up 
until now I have not seen any problems.



-- 
Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Les Mikesell]

On Thu, Feb 13, 2014 at 1:06 PM, Ljubomir Ljubojevic  wrote:
>
> I have a strange problem.
>
> I installed cacti, and set it up.
>
> When I click on the Plugins Management, I noticed Error
> "FATAL: Database or Table does not exist"
>
> Further testing revealed that at the moment MySQL can not create
> temporary files in /tmp. It is same for ANY database!
>
> SELinux is disabled.
> I logged to mysql as root and error is same.
>
> Command used is:
> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;
>
>
> Error is:
> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)
>
> What could be the problem, any idea how to proceed?

http://bugs.mysql.com/bug.php?id=60574
The error may be on the source of the copy (bak_menu.frm) not in tmp.

-- 
   Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Clint Dilks]

Hi,

Have you verified that permissions on /tmp appear correct ?

An example from one of my systems


# ls -lsahd /tmp
4.0K drwxrwxrwt. 3 root root 4.0K Feb 14 07:01 /tm

So permissions 1777

Perhaps test creating a file in /tmp as root and as a non root user and let
us know the results that way we may be able to help further.




On Fri, Feb 14, 2014 at 8:06 AM, Ljubomir Ljubojevic wrote:

> Hi.
>
> I have a strange problem.
>
> I installed cacti, and set it up.
>
> When I click on the Plugins Management, I noticed Error
> "FATAL: Database or Table does not exist"
>
> Further testing revealed that at the moment MySQL can not create
> temporary files in /tmp. It is same for ANY database!
>
> SELinux is disabled.
> I logged to mysql as root and error is same.
>
> Command used is:
> CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;
>
>
> Error is:
> ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)
>
> What could be the problem, any idea how to proceed?
>
> I do not think I edited any MySQL setting.
>
> --
> Ljubomir Ljubojevic
> (Love is in the Air)
> PL Computers
> Serbia, Europe
>
> StarOS, Mikrotik and CentOS/RHEL/Linux consultant
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL Can't create file '/tmp/#sql2e0_98_0.frm' (errno: 9) problem

[Ljubomir Ljubojevic]

Hi.

I have a strange problem.

I installed cacti, and set it up.

When I click on the Plugins Management, I noticed Error
"FATAL: Database or Table does not exist"

Further testing revealed that at the moment MySQL can not create 
temporary files in /tmp. It is same for ANY database!

SELinux is disabled.
I logged to mysql as root and error is same.

Command used is:
CREATE TEMPORARY TABLE IF NOT EXISTS plugin_temp_table LIKE bak_menu;


Error is:
ERROR 1004 (HY000): Can't create file '/tmp/#sql21f8_f3_0.frm' (errno: 9)

What could be the problem, any idea how to proceed?

I do not think I edited any MySQL setting.

-- 
Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL stale socket

[Luigi Rosa]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Paul Heinlein said the following on 20/12/2013 16:50:
> On Fri, 20 Dec 2013, Luigi Rosa wrote:
> 
>> I can confirm that this is a bug added to the latest releases og MySQL.
>> 
>> I tested it only in VM (both VMware and XEN).
>> 
>> How to reproduce.
>> 
>> * start a CentOS VM with MySQL * abruptly turn the VM off (not shutdown,
>> of course) * restart VM * MySQL starts but the connections via
>> /var/lib/mysql/mysql.sock do not work
> 
> I've seen this too, and not just on VMs.

A friend just told me that he saw this issue recently on a Debian server, so
is definitively a MySQL problem.



Ciao,
luigi

- -- 
/
+--[Luigi Rosa]--
\

Customer: I have Windows Vista on my computer.
Tech: Yes.
Customer: My computer isn't working right.
Tech: Yes. You already told me that.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.14 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlK0bLcACgkQ3kWu7Tfl6ZQsuACfcYHQuEgGThvBuTwCDITqqULS
IVsAoJwtbkLBxvahWDgqdVfuySAh41vl
=FyA0
-END PGP SIGNATURE-
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL stale socket

[Paul Heinlein]

On Fri, 20 Dec 2013, Luigi Rosa wrote:


I can confirm that this is a bug added to the latest releases og MySQL.

I tested it only in VM (both VMware and XEN).

How to reproduce.

* start a CentOS VM with MySQL
* abruptly turn the VM off (not shutdown, of course)
* restart VM
* MySQL starts but the connections via /var/lib/mysql/mysql.sock do not work


I've seen this too, and not just on VMs.

--
Paul Heinlein
heinl...@madboa.com
45°38' N, 122°6' W___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL stale socket

[Luigi Rosa]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I can confirm that this is a bug added to the latest releases og MySQL.

I tested it only in VM (both VMware and XEN).

How to reproduce.

* start a CentOS VM with MySQL
* abruptly turn the VM off (not shutdown, of course)
* restart VM
* MySQL starts but the connections via /var/lib/mysql/mysql.sock do not work




Jake Shipton said the following on 16/12/2013 11:12:
> On Mon, 16 Dec 2013 10:51:54 +0100 Luigi Rosa  wrote:
> 
>> 
>> I have a server with a faulty power supply that resets the server every
>> 4/5 hours.
>> 
>> While I wait for the supplier to change the PS, I am facing an issue with
>> MySQL.
>> 
>> When the system comes back on, mysqld script starts the daemon, but the
>> connections via /var/lib/mysql/mysql.sock don't work because MySQL
>> doesn't delete and recreate the stale socket.
>> 
>> Apart from editing /etc/init.d/mysqld to force the deletion of a stale
>> socket are there any other way to solve this issue?
>> 
>> 
>> Thank you un advance.
>> 
>> 
>> Ciao, luigi
>> 
>> 
>> ___ CentOS mailing list 
>> CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
> 
> Hi,
> 
> As a *temporary* solution you could do the following:
> 
> # chkconfig mysqld off # nano /etc/rc.local Add at bottom: rm
> /var/lib/mysql/mysql.sock service mysqld start Now Save.
> 
> Next time the machine boots it should remove the sock file, and then start
> mysqld.
> 
> Once your PSU is fixed, remove those lines and run:
> 
> # chkconfig mysqld on
> 
> Job done! :-D
> 
> Hope this helps.
> 
> Kind Regards, Jake Shipton (JakeMS) GPG Key: 0xE3C31D8F GPG Fingerprint:
> 7515 CC63 19BD 06F9 400A DE8A 1D0B A5CF E3C3 1D8F 
> ___ CentOS mailing list 
> CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
> 



Ciao,
luigi

- -- 
/
+--[Luigi Rosa]--
\

ANOTHER GLITCH IN THE CALL
  We don't need no indirection
  We don't need no flow control
  No data typing or declarations
  Did you leave those lists alone
  Hey! Hacker! Leave those lists alone!
  All in all, it's just a pure-Lisp function call
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.14 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEUEARECAAYFAlKz7nMACgkQ3kWu7Tfl6ZRW6QCY/x0h/O6cl42hmUO3kcLstzBC
JgCfcMizFR6axWBbzsGnnNNG8RmWx20=
=h+P3
-END PGP SIGNATURE-
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL stale socket

[Jake Shipton]

On Mon, 16 Dec 2013 10:51:54 +0100
Luigi Rosa  wrote:

> 
> I have a server with a faulty power supply that resets the server
> every 4/5 hours.
> 
> While I wait for the supplier to change the PS, I am facing an issue
> with MySQL.
> 
> When the system comes back on, mysqld script starts the daemon, but
> the connections via /var/lib/mysql/mysql.sock don't work because
> MySQL doesn't delete and recreate the stale socket.
> 
> Apart from editing /etc/init.d/mysqld to force the deletion of a
> stale socket are there any other way to solve this issue?
> 
> 
> Thank you un advance.
> 
> 
> Ciao,
> luigi
> 
> 
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos

Hi, 

As a *temporary* solution you could do the following:

# chkconfig mysqld off
# nano /etc/rc.local
 Add at bottom: 
  rm /var/lib/mysql/mysql.sock
  service mysqld start
 Now Save.

Next time the machine boots it should remove the sock file, and then
start mysqld.

Once your PSU is fixed, remove those lines and run:

# chkconfig mysqld on

Job done! :-D

Hope this helps.

Kind Regards, 
Jake Shipton (JakeMS)
GPG Key: 0xE3C31D8F
GPG Fingerprint: 7515 CC63 19BD 06F9 400A DE8A 1D0B A5CF E3C3 1D8F
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL stale socket

[Luigi Rosa]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I have a server with a faulty power supply that resets the server every 4/5
hours.

While I wait for the supplier to change the PS, I am facing an issue with MySQL.

When the system comes back on, mysqld script starts the daemon, but the
connections via /var/lib/mysql/mysql.sock don't work because MySQL doesn't
delete and recreate the stale socket.

Apart from editing /etc/init.d/mysqld to force the deletion of a stale socket
are there any other way to solve this issue?


Thank you un advance.


Ciao,
luigi

- -- 
/
+--[Luigi Rosa]--
\

Microsoft may issue tablets by Christmas. No word yet on whether
they'll contain potassium cyanide or saxitoxin.
--fark.com
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.14 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlKuzTUACgkQ3kWu7Tfl6ZTV8wCfdFrw937bBokoy3MPcIRy3Qd+
fWMAn1/fEyC6LBZp/0kHHzByqg8nFRP3
=jV5r
-END PGP SIGNATURE-
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL stale socket

[Luigi Rosa]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I have a server with a faulty power supply that resets the server every 4/5
hours.

While I wait for the supplier to change the PS, I am facing an issue with MySQL.

When the system comes back on, mysqld script starts the daemon, but the
connections via /var/lib/mysql/mysql.sock don't work because MySQL doesn't
delete and recreate the stale socket.

Apart from editing /etc/init.d/mysqld to force the deletion of a stale socket
are there any other way to solve this issue?


Thank you un advance.


Ciao,
luigi

- -- 
/
+--[Luigi Rosa]--
\

Microsoft may issue tablets by Christmas. No word yet on whether
they'll contain potassium cyanide or saxitoxin.
--fark.com
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.14 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlKuzTMACgkQ3kWu7Tfl6ZRW6gCfaU56Jcayg7TfZU9b0vTkTJXs
m7EAoM7ROhaRk+x+3C0+ygi/Xk+2PKke
=BWy2
-END PGP SIGNATURE-
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[John R Pierce]

On 11/15/2013 10:19 AM, Les Mikesell wrote:
> On Fri, Nov 15, 2013 at 12:16 PM, Reindl Harald  
> wrote:
>> >
>> >in other words you do not know anything

Reindl is banned from this list for serial rudeness, but replies 
privately to list postings with his 'wisdom'.  please don't copy his 
private responses back to the list.



-- 
john r pierce  37N 122W
somewhere on the middle of the left coast

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 12:32 PM, Reindl Harald  wrote:
>
>> Yes, if you do things that work in general, you don't have to know all
>> of the internals of every specific version of every specific program
>> that you might ever touch.
>
> than better be quiet instead contradict and correct people with
> generic wisdom which are on-topic and have the specific knowledge
> and experience over many years

Fine, the next time I want to move MySQL 4.0-4.5 on Apple OSX *Power
PC* to MySQL 5.1/5.5 on Linux x86_64, I'll ask your advice.

-- 
   Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 12:16 PM, Reindl Harald  wrote:
>
> in other words you do not know anything about the mysql-internals
> as you statet more than once, have no expierience and come up with
> generic theory to defend your statements

Yes, if you do things that work in general, you don't have to know all
of the internals of every specific version of every specific program
that you might ever touch.

-- 
  Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 12:00 PM, Reindl Harald  wrote:
>
> and why did you strip the part where i googled for you the endianess
> of MacOSX on PPC because i clearly statet mysql-data dirs used here
> in production went from MySQL 3.x on Windows i686 over MySQL 4.0-4.5
> on Apple OSX *Power PC* to MySQL 5.1/5.5 on Linux x86_64?

Because that has nothing to do with the generic issue of moving binary
data from one CPU type to another.  If you got lucky and something
happened to work for you once, great.  It's not something people
should count on - especially for programs that provide a tool for
portability and consistency.

-- 
   Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 10:58 AM, Reindl Harald  wrote:
>
>> I don't know specifically about Mysql but pretty much anything that stores 
>> binary
>> representations to disk needs to be converted.
>
> with this logic you would need to convert MS office files
> containing embedded images too

No, image formats have standardized binary representations so they are
portable.  The problem is with applications that write their own
binary values and data structures to disk without conversion.   I
don't know much about msyql internals, but I wouldn't expect it to go
to the trouble of normalizing things like index files just in case you
wanted to move those disks to a sparc instead of doing a dump/load.

-- 
  Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 10:01 AM, Reindl Harald  wrote:
>
> abd what has the CPU type to do with the mysql datadir?

Really???  You've never used processors with different bit/byte order
than Intel's?  I don't know specifically about Mysql but pretty much
anything that stores binary representations to disk needs to be
converted.

> speaking with 10 years expierience and some hundret databases originally
> created with MySQl 3.x on Windows, 4.x-5.0 running on MacOSX and since
> MySQL 5.1 until now with 5.5 as well as MariaDB 5.5 running on Linux
> without a single time payling around with dumps

So no sparc's or the like in the picture?   Not sure which way powerpc macs ran.

-- 
   Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Leon Fauster]

Am 15.11.2013 um 16:55 schrieb Les Mikesell :
> On Fri, Nov 15, 2013 at 6:18 AM, Timothy Murphy  wrote:
>> Last week I had a hard disk failure on my CentOS server.
>> I managed to re-install CentOS on a new disk.
>> I have the old mysql databases from /var/lib/mysql .
>> Can I just move them to my new disk?
>> 
>> Any help or suggestions gratefully received.
> 
> What you are 'supposed' to do is take the backup that you make
> regularly with 'mysqldump' and restore it with the mysql command line
> program.But if you are running on the same cpu type and mysql
> version, moving the disk or copying the directory contents will work
> (along with the things already mentioned about /etc/my.cnf and
> selinux).



provided that the old system had shutdown the mysql process 
regularly. Means, the integrity of /var/lib/mysql is warranted. 

--
LF






___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Les Mikesell]

On Fri, Nov 15, 2013 at 6:18 AM, Timothy Murphy  wrote:
> Last week I had a hard disk failure on my CentOS server.
> I managed to re-install CentOS on a new disk.
> I have the old mysql databases from /var/lib/mysql .
> Can I just move them to my new disk?
>
> Any help or suggestions gratefully received.

What you are 'supposed' to do is take the backup that you make
regularly with 'mysqldump' and restore it with the mysql command line
program.But if you are running on the same cpu type and mysql
version, moving the disk or copying the directory contents will work
(along with the things already mentioned about /etc/my.cnf and
selinux).

-- 
  Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[m . roth]

Marios Zindilis wrote:
> On Fri, Nov 15, 2013 at 2:18 PM, Timothy Murphy 
> wrote:
>
>> Last week I had a hard disk failure on my CentOS server.
>> I managed to re-install CentOS on a new disk.
>> I have the old mysql databases from /var/lib/mysql .
>> Can I just move them to my new disk?
>>
>> Any help or suggestions gratefully received.

> MySQL is usually confined in /var/lib/mysql, so if you move the
> /etc/my.cnf
> and the /var/lib/mysql from the old server to the new one, you should be
> able to start the server normally, otherwise you should get some useful
> errors.
>
> A typical error in such case is MySQL files ownership/permissions.
>
Marios, please don't top post.

And speaking of ownership/permissions... watch out for selinux contexts.
You may want to touch /.autorelabel and reboot, and walk away for a
while

 mark

  mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql databases from old system?

[Marios Zindilis]

MySQL is usually confined in /var/lib/mysql, so if you move the /etc/my.cnf
and the /var/lib/mysql from the old server to the new one, you should be
able to start the server normally, otherwise you should get some useful
errors.

A typical error in such case is MySQL files ownership/permissions.


On Fri, Nov 15, 2013 at 2:18 PM, Timothy Murphy  wrote:

> Last week I had a hard disk failure on my CentOS server.
> I managed to re-install CentOS on a new disk.
> I have the old mysql databases from /var/lib/mysql .
> Can I just move them to my new disk?
>
> Any help or suggestions gratefully received.
>
> --
> Timothy Murphy
> e-mail: gayleard /at/ eircom.net
> School of Mathematics, Trinity College, Dublin 2, Ireland
>
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
Marios Zindilis
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Mysql databases from old system?

[Timothy Murphy]

Last week I had a hard disk failure on my CentOS server.
I managed to re-install CentOS on a new disk.
I have the old mysql databases from /var/lib/mysql .
Can I just move them to my new disk?

Any help or suggestions gratefully received.

-- 
Timothy Murphy  
e-mail: gayleard /at/ eircom.net
School of Mathematics, Trinity College, Dublin 2, Ireland


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Johnny Hughes]

On 10/30/2013 05:22 AM, Leon Fauster wrote:
> Am 29.10.2013 um 22:56 schrieb Reindl Harald :
>>> php-mysql is linked against mysql-5.0-libs and talking to 
>>> a mysql-5.5 server. Should this cause an unpredictable behavior?
>> no - how do you come to this idea?
> In the last decade i saw very strange results while updating 
> or doing similar things. the question above is more a theoretical 
> thought despite the rational knowledge that in this example 
> (communication-layer) the compatibility is preserved.

If you look at the Software collections info for EL6, they specifically
discuss that issue:

http://dev.centos.org/centos/6/SCL/docs/#sect-Installation_and_Usage-Migrating_from_MySQL_5.1_to_MySQL_5.5

(see section 2.6.3)

So, Red Hat has specifically tested using the 5.1 (for EL6 and 5.0 for
EL5) client library with the 5.5 database and has certified it as
working enough that they support it.

Does that mean it is flawless ... likely not.  But it is as intended.



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Leon Fauster]

Am 30.10.2013 um 17:31 schrieb Matthew B. Brookover :
> 
> On Wed, 2013-10-30 at 11:22 +0100, Leon Fauster wrote:
>> Am 29.10.2013 um 22:56 schrieb Reindl Harald :
>>> 
>>> no - how do you come to this idea?
>> 
>> In the last decade i saw very strange results while updating 
>> or doing similar things. the question above is more a theoretical 
>> thought despite the rational knowledge that in this example 
>> (communication-layer) the compatibility is preserved.
> 
> I have always tried to keep the client libraries in sync with the
> database in my own code.  I was surprised that the upstream builders had
> mixed the 5.0 client with the 5.5 data base.  I know of cases where
> library developers will provide for compatibility with various versions
> of protocols.  It looks to me like this is one of those cases.
> 
> We are using Perl and PHP with MySQL, and both seem to work with the 5.0
> client library on the 5.5 database.
> 
> Besides the usual tests like create a new web page, delete a web page,
> edit a page, etc, we pointed a spider that is part of burp suite at the
> web site on a development system.  The web site is several thousand
> pages, some are PHP with parts that are pulled from an upgraded mysql
> 5.5 data base and others are flat files.  Burp suite has been running
> for several hours and every thing looks good so far.


interesting, thanks for sharing. 


> I think this weekend, we will start moving production servers from
> CentOS 5.9 to 5.10 and from MySQL 5.0 to 5.5.



--
LF


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Matthew B. Brookover]

On Wed, 2013-10-30 at 11:22 +0100, Leon Fauster wrote:
> Am 29.10.2013 um 22:56 schrieb Reindl Harald :
> >> php-mysql is linked against mysql-5.0-libs and talking to 
> >> a mysql-5.5 server. Should this cause an unpredictable behavior?
> > 
> > no - how do you come to this idea?
> 
> In the last decade i saw very strange results while updating 
> or doing similar things. the question above is more a theoretical 
> thought despite the rational knowledge that in this example 
> (communication-layer) the compatibility is preserved.

I have always tried to keep the client libraries in sync with the
database in my own code.  I was surprised that the upstream builders had
mixed the 5.0 client with the 5.5 data base.  I know of cases where
library developers will provide for compatibility with various versions
of protocols.  It looks to me like this is one of those cases.

We are using Perl and PHP with MySQL, and both seem to work with the 5.0
client library on the 5.5 database.

Besides the usual tests like create a new web page, delete a web page,
edit a page, etc, we pointed a spider that is part of burp suite at the
web site on a development system.  The web site is several thousand
pages, some are PHP with parts that are pulled from an upgraded mysql
5.5 data base and others are flat files.  Burp suite has been running
for several hours and every thing looks good so far.

I think this weekend, we will start moving production servers from
CentOS 5.9 to 5.10 and from MySQL 5.0 to 5.5.

Matt


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Leon Fauster]

Am 29.10.2013 um 22:56 schrieb Reindl Harald :
>> php-mysql is linked against mysql-5.0-libs and talking to 
>> a mysql-5.5 server. Should this cause an unpredictable behavior?
> 
> no - how do you come to this idea?

In the last decade i saw very strange results while updating 
or doing similar things. the question above is more a theoretical 
thought despite the rational knowledge that in this example 
(communication-layer) the compatibility is preserved.

--
LF  


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Eero Volotinen]

ges and will leave
> > mysql-5.0-5.el5_9 in place for now.  FYI, I searched through the rest of
> > the installed packages, the only one that has the client library is
> > mysql-5.0.95-5.el5_9.
>
>
> php-mysql is linked against mysql-5.0-libs and talking to
> a mysql-5.5 server. Should this cause an unpredictable behavior?
>

No.

--
Eero
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Leon Fauster]

Am 28.10.2013 um 22:30 schrieb Matthew B. Brookover :
> On Mon, 2013-10-28 at 22:18 +0200, Eero Volotinen wrote:
>> 2013/10/28 Matthew B. Brookover 
>> 
>>> I upgraded a web server from CentOS 5.9 to 5.10.  The instructions also
>>> indicate that MySQL 5.0 is no longer supported so I followed the
>>> instructions to upgrade to 5.5.  Everything seems to be working so I
>>> tried to remove MySQL 5.0 and MySQL 5.1.  It seems there are still
>>> 
>>> As I understand it, there is no longer any support for
>>> mysql-5.0.95-5.el5_9, but it looks like it is still in use.  Is there
>>> any plan to move Perl DBD, mod_auth_mysql, ODBC, SASL, Dovecot, and PHP
>>> from the 5.0 MySQL drivers to the 5.5 drivers?
>>> 
>>> Is there something that I missed in the upgrade process?
>>> 
>> 
>> Yes, you cannot remove mysql-5.0. see redhat knowledge base for more
>> information.
> 
> Thanks Eero, Good to know.
> 
> I did remove the rest of the 5.0 and 5.1 packages and will leave
> mysql-5.0-5.el5_9 in place for now.  FYI, I searched through the rest of
> the installed packages, the only one that has the client library is
> mysql-5.0.95-5.el5_9.


php-mysql is linked against mysql-5.0-libs and talking to 
a mysql-5.5 server. Should this cause an unpredictable behavior?

--
LF


 

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Matthew B. Brookover]

On Mon, 2013-10-28 at 22:18 +0200, Eero Volotinen wrote:
> 2013/10/28 Matthew B. Brookover 
> 
> > I upgraded a web server from CentOS 5.9 to 5.10.  The instructions also
> > indicate that MySQL 5.0 is no longer supported so I followed the
> > instructions to upgrade to 5.5.  Everything seems to be working so I
> > tried to remove MySQL 5.0 and MySQL 5.1.  It seems there are still
> >
> > As I understand it, there is no longer any support for
> > mysql-5.0.95-5.el5_9, but it looks like it is still in use.  Is there
> > any plan to move Perl DBD, mod_auth_mysql, ODBC, SASL, Dovecot, and PHP
> > from the 5.0 MySQL drivers to the 5.5 drivers?
> >
> > Is there something that I missed in the upgrade process?
> >
> 
> Yes, you cannot remove mysql-5.0. see redhat knowledge base for more
> information.

Thanks Eero, Good to know.

I did remove the rest of the 5.0 and 5.1 packages and will leave
mysql-5.0-5.el5_9 in place for now.  FYI, I searched through the rest of
the installed packages, the only one that has the client library is
mysql-5.0.95-5.el5_9.

[root@w3 root]# rpm -qa | grep mysql | while read a; do echo $a; rpm -ql $a | 
grep libmysqlclient | sed 's/^/   /'; done
mysql55-mysql-devel-5.5.32-3.el5
mysql55-mysql-server-5.5.32-3.el5
mysql55-runtime-1-12.el5
mod_auth_mysql-3.0.0-3.2.el5_3
mysql-connector-odbc-3.51.26r1127-2.el5
mysql-5.0.95-5.el5_9
/usr/lib/mysql/libmysqlclient.so.15
/usr/lib/mysql/libmysqlclient.so.15.0.0
/usr/lib/mysql/libmysqlclient_r.so.15
/usr/lib/mysql/libmysqlclient_r.so.15.0.0
mysql55-mysql-5.5.32-3.el5
mysql55-mysql-bench-5.5.32-3.el5
php53-mysql-5.3.3-21.el5
mysql55-mysql-libs-5.5.32-3.el5
mysql55-1-12.el5
mysql-bench-5.0.95-5.el5_9
mysql55-build-1-12.el5
mysql55-mysql-test-5.5.32-3.el5
[root@w3 root]# 


Matt

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Eero Volotinen]

2013/10/28 Matthew B. Brookover 

> I upgraded a web server from CentOS 5.9 to 5.10.  The instructions also
> indicate that MySQL 5.0 is no longer supported so I followed the
> instructions to upgrade to 5.5.  Everything seems to be working so I
> tried to remove MySQL 5.0 and MySQL 5.1.  It seems there are still
>
> As I understand it, there is no longer any support for
> mysql-5.0.95-5.el5_9, but it looks like it is still in use.  Is there
> any plan to move Perl DBD, mod_auth_mysql, ODBC, SASL, Dovecot, and PHP
> from the 5.0 MySQL drivers to the 5.5 drivers?
>
> Is there something that I missed in the upgrade process?
>

Yes, you cannot remove mysql-5.0. see redhat knowledge base for more
information.

--
Eero,RHCE
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL 5.5, PHP and other tools still use 5.0 client drivers

[Matthew B. Brookover]

I upgraded a web server from CentOS 5.9 to 5.10.  The instructions also
indicate that MySQL 5.0 is no longer supported so I followed the
instructions to upgrade to 5.5.  Everything seems to be working so I
tried to remove MySQL 5.0 and MySQL 5.1.  It seems there are still
dependencies on MySQL 5.0:

[root@w3 ~]# rpm -e mysql51-mysql-libs-5.1.70-1.el5 mysql-server-5.0.95-5.el5_9 
mysql51-build-1-9.el5 mysql51-mysql-5.1.70-1.el5 
mysql51-mysql-server-5.1.70-1.el5 mysql51-mysql-test-5.1.70-1.el5 
mysql-devel-5.0.95-5.el5_9 mysql-5.0.95-5.el5_9 mysql-test-5.0.95-5.el5_9 
mysql51-runtime-1-9.el5 mysql51-1-9.el5 mysql51-mysql-bench-5.1.70-1.el5 
mysql51-mysql-libs-5.1.70-1.el5 mysql51-build-1-9.el5 
mysql51-mysql-5.1.70-1.el5 mysql51-mysql-server-5.1.70-1.el5 
mysql51-mysql-test-5.1.70-1.el5 mysql51-runtime-1-9.el5 mysql51-1-9.el5 
mysql51-mysql-bench-5.1.70-1.el5
error: Failed dependencies:
libmysqlclient.so.15 is needed by (installed) perl-DBD-MySQL-3.0007-2.el5.i386
libmysqlclient.so.15 is needed by (installed) 
mod_auth_mysql-3.0.0-3.2.el5_3.i386
libmysqlclient.so.15 is needed by (installed) 
mysql-connector-odbc-3.51.26r1127-2.el5.i386
libmysqlclient.so.15 is needed by (installed) 
cyrus-sasl-sql-2.1.22-7.el5_8.1.i386
libmysqlclient.so.15 is needed by (installed) dovecot-1.0.7-8.el5_9.1.i386
libmysqlclient.so.15 is needed by (installed) php53-mysql-5.3.3-21.el5.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
perl-DBD-MySQL-3.0007-2.el5.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
mod_auth_mysql-3.0.0-3.2.el5_3.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
mysql-connector-odbc-3.51.26r1127-2.el5.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
cyrus-sasl-sql-2.1.22-7.el5_8.1.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
dovecot-1.0.7-8.el5_9.1.i386
libmysqlclient.so.15(libmysqlclient_15) is needed by (installed) 
php53-mysql-5.3.3-21.el5.i386
libmysqlclient_r.so.15 is needed by (installed) 
mysql-connector-odbc-3.51.26r1127-2.el5.i386
libmysqlclient_r.so.15 is needed by (installed) 
MySQL-python-1.2.3-0.1.c1.el5.i386
libmysqlclient_r.so.15(libmysqlclient_15) is needed by (installed) 
mysql-connector-odbc-3.51.26r1127-2.el5.i386
libmysqlclient_r.so.15(libmysqlclient_15) is needed by (installed) 
MySQL-python-1.2.3-0.1.c1.el5.i386
mysql = 5.0.95-5.el5_9 is needed by (installed) mysql-bench-5.0.95-5.el5_9.i386
[root@w3 ~]# find /usr/lib -name libmysqlclient.so.15 -ls
721332 0 lrwxrwxrwx 1 root root 24 Jan 23 2013 
/usr/lib/mysql/libmysqlclient.so.15 -> libmysqlclient.so.15.0.0
[root@w3 ~]# find /usr/lib -name libmysqlclient.so.15.0.0 -ls
721556 1424 -rwxr-xr-x 1 root root 1452764 Jan 22 2013 
/usr/lib/mysql/libmysqlclient.so.15.0.0
[root@w3 ~]# rpm -qf /usr/lib/mysql/libmysqlclient.so.15.0.0
mysql-5.0.95-5.el5_9
[root@w3 ~]#

I started up httpd, mysql, etc, and used lsof to see which library was
in use:
[root@w3 csm-vhosts]# lsof -p 4552  | grep mysql
httpd   4552 apache  memREG  253,143032   593822 
/usr/lib/php/modules/mysql.so
httpd   4552 apache  memREG  253,1   113324   593824 
/usr/lib/php/modules/mysqli.so
httpd   4552 apache  memREG  253,124460   593835 
/usr/lib/php/modules/pdo_mysql.so
httpd   4552 apache  memREG  253,1  1452764   721556 
/usr/lib/mysql/libmysqlclient.so.15.0.0
[root@w3 csm-vhosts]# 

As I understand it, there is no longer any support for
mysql-5.0.95-5.el5_9, but it looks like it is still in use.  Is there
any plan to move Perl DBD, mod_auth_mysql, ODBC, SASL, Dovecot, and PHP
from the 5.0 MySQL drivers to the 5.5 drivers?

Is there something that I missed in the upgrade process?

thanks

Matt
mbroo...@mines.edu

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.1.69 at CentOS 6.4 doesn't know enable-named-pipe?

[sjt5atra]

http://dev.mysql.com/doc/refman/5.1/en/server-options.html#option_mysqld_skip-networking

On Sep 13, 2013, at 6:44 AM, Alexander Farber  
wrote:

> Hello,
> 
> I've only had experience with PostgreSQL sofar,
> but have now to install MySQL (and WordPress)
> on a CentOS 6.4 /64 bit server.
> 
> I have installed the mysql-5.1.69-1.el6_4.x86_64
> package and executed the following commands:
> 
> # chkconfig mysqld on
> # service mysqld start
> # /usr/bin/mysqladmin -u root password 'x'
> # /usr/bin/mysql_secure_installation
> 
> Then I've noticed that mysqld_safe process
> is listening at 0.0.0.0 and decided to change that -
> so that my WordPress installation only uses
> domain sockets (or unix pipes? not sure).
> 
> So I've modified the /etc/my.cnf to:
> 
> [mysqld]
> datadir=/var/lib/mysql
> socket=/var/lib/mysql/mysql.sock
> user=mysql
> # Disabling symbolic-links is recommended to prevent assorted security risks
> symbolic-links=0
> bind-address = localhost
> skip-networking
> enable-named-pipe
> 
> [mysqld_safe]
> log-error=/var/log/mysqld.log
> pid-file=/var/run/mysqld/mysqld.pid
> 
> But now MySQL refuses to start:
> 
> # service mysqld restart
> Stopping mysqld:   [  OK  ]
> MySQL Daemon failed to start.
> Starting mysqld:   [FAILED]
> 
> The  /var/log/mysqld.log contains:
> 
> mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
>  InnoDB: Initializing buffer pool, size = 8.0M
>  InnoDB: Completed initialization of buffer pool
>  InnoDB: Started; log sequence number 0 44233
> [ERROR] /usr/libexec/mysqld: unknown option '--enable-named-pipe'
> [ERROR] Aborting
>  InnoDB: Starting shutdown...
>  InnoDB: Shutdown completed; log sequence number 0 44233
> [Note] /usr/libexec/mysqld: Shutdown complete
> mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended
> 
> I've searched Google and grepped
> /usr/share/mysql/*.cnf for that directive,
> but haven't found any hints there.
> 
> Any hints please?
> 
> Regards
> Alex
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.1.69 at CentOS 6.4 doesn't know enable-named-pipe?

[Alexander Farber]

Thank you, that was it.

I didn't realize "mysqld-nt" means "Windows only"
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL 5.1.69 at CentOS 6.4 doesn't know enable-named-pipe?

[Marios Zindilis]

Docs says it's Windows specific:

http://dev.mysql.com/doc/refman/5.1/en/server-options.html#option_mysqld_enable-named-pipe


On Fri, Sep 13, 2013 at 1:44 PM, Alexander Farber <
alexander.far...@gmail.com> wrote:

> Hello,
>
> I've only had experience with PostgreSQL sofar,
> but have now to install MySQL (and WordPress)
> on a CentOS 6.4 /64 bit server.
>
> I have installed the mysql-5.1.69-1.el6_4.x86_64
> package and executed the following commands:
>
> # chkconfig mysqld on
> # service mysqld start
> # /usr/bin/mysqladmin -u root password 'x'
> # /usr/bin/mysql_secure_installation
>
> Then I've noticed that mysqld_safe process
> is listening at 0.0.0.0 and decided to change that -
> so that my WordPress installation only uses
> domain sockets (or unix pipes? not sure).
>
> So I've modified the /etc/my.cnf to:
>
> [mysqld]
> datadir=/var/lib/mysql
> socket=/var/lib/mysql/mysql.sock
> user=mysql
> # Disabling symbolic-links is recommended to prevent assorted security
> risks
> symbolic-links=0
> bind-address = localhost
> skip-networking
> enable-named-pipe
>
> [mysqld_safe]
> log-error=/var/log/mysqld.log
> pid-file=/var/run/mysqld/mysqld.pid
>
> But now MySQL refuses to start:
>
> # service mysqld restart
> Stopping mysqld:   [  OK  ]
> MySQL Daemon failed to start.
> Starting mysqld:   [FAILED]
>
> The  /var/log/mysqld.log contains:
>
>  mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
>   InnoDB: Initializing buffer pool, size = 8.0M
>   InnoDB: Completed initialization of buffer pool
>   InnoDB: Started; log sequence number 0 44233
>  [ERROR] /usr/libexec/mysqld: unknown option '--enable-named-pipe'
>  [ERROR] Aborting
>   InnoDB: Starting shutdown...
>   InnoDB: Shutdown completed; log sequence number 0 44233
>  [Note] /usr/libexec/mysqld: Shutdown complete
>  mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended
>
> I've searched Google and grepped
> /usr/share/mysql/*.cnf for that directive,
> but haven't found any hints there.
>
> Any hints please?
>
> Regards
> Alex
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
Marios Zindilis
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL 5.1.69 at CentOS 6.4 doesn't know enable-named-pipe?

[Alexander Farber]

Hello,

I've only had experience with PostgreSQL sofar,
but have now to install MySQL (and WordPress)
on a CentOS 6.4 /64 bit server.

I have installed the mysql-5.1.69-1.el6_4.x86_64
package and executed the following commands:

# chkconfig mysqld on
# service mysqld start
# /usr/bin/mysqladmin -u root password 'x'
# /usr/bin/mysql_secure_installation

Then I've noticed that mysqld_safe process
is listening at 0.0.0.0 and decided to change that -
so that my WordPress installation only uses
domain sockets (or unix pipes? not sure).

So I've modified the /etc/my.cnf to:

[mysqld]
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
user=mysql
# Disabling symbolic-links is recommended to prevent assorted security risks
symbolic-links=0
bind-address = localhost
skip-networking
enable-named-pipe

[mysqld_safe]
log-error=/var/log/mysqld.log
pid-file=/var/run/mysqld/mysqld.pid

But now MySQL refuses to start:

# service mysqld restart
Stopping mysqld:   [  OK  ]
MySQL Daemon failed to start.
Starting mysqld:   [FAILED]

The  /var/log/mysqld.log contains:

 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
  InnoDB: Initializing buffer pool, size = 8.0M
  InnoDB: Completed initialization of buffer pool
  InnoDB: Started; log sequence number 0 44233
 [ERROR] /usr/libexec/mysqld: unknown option '--enable-named-pipe'
 [ERROR] Aborting
  InnoDB: Starting shutdown...
  InnoDB: Shutdown completed; log sequence number 0 44233
 [Note] /usr/libexec/mysqld: Shutdown complete
 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended

I've searched Google and grepped
/usr/share/mysql/*.cnf for that directive,
but haven't found any hints there.

Any hints please?

Regards
Alex
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mysql cluster vs. mysql-libs

[Karanbir Singh]

On 06/06/2012 12:45 AM, Dennis Jacobfeuerborn wrote:
> Hi,
> it seems even a minimal install of Centos 6.2 now requires mysql-libs as a
> dependency for postfix and cronie. When I try to install the mysql cluster
> rpms from mysql.com I get a collision between the files.
> Since this is the first time I get into contact with mysql cluster I wonder
> if somebody has an idea how to resolve this or knows alternative packages
> that don't conflict with mysql-libs?

mysql.com should ship a compat rpm to handle this - its a bug at their
end if they dont.


-- 
Karanbir Singh
+44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh
ICQ: 2522219| Yahoo IM: z00dax  | Gtalk: z00dax
GnuPG Key : http://www.karan.org/publickey.asc
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Mysql cluster vs. mysql-libs

[Dennis Jacobfeuerborn]

Hi,
it seems even a minimal install of Centos 6.2 now requires mysql-libs as a
dependency for postfix and cronie. When I try to install the mysql cluster
rpms from mysql.com I get a collision between the files.
Since this is the first time I get into contact with mysql cluster I wonder
if somebody has an idea how to resolve this or knows alternative packages
that don't conflict with mysql-libs?

Regards,
  Dennis
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql

[Steve Clark]

On 06/05/2012 11:13 AM, Leonard den Ottolander wrote:
> See for yourself:
>
> # grep mysql /etc/shadow
> mysql:!!:15404::
Ah - good,  from man 5 shadow...

   encrypted password
Refer to crypt(3) for details on how this string is interpreted.

If the password field contains some string that is not a valid 
result of crypt(3), for
instance ! or *, the user will not be able to use a unix password 
to log in (but the user
may log in the system by other means).

This field may be empty, in which case no passwords are required to 
authenticate as the
specified login name. However, some applications which read the 
/etc/shadow file may
decide not to permit any access at all if the password field is 
empty.

A password field which starts with a exclamation mark means that 
the password is locked.
The remaining characters on the line represent the password field 
before the password was
locked.

-- 
Stephen Clark
*NetWolves*
Director of Technology
Phone: 813-579-3200
Fax: 813-882-0209
Email: steve.cl...@netwolves.com
http://www.netwolves.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql

[Leonard den Ottolander]

Hello Steve,

On Tue, 2012-06-05 at 10:57 -0400, Steve Clark wrote:
> I see that installing mysql in Centos 6.x creates a mysql user
> with a login shell of /bin/bash.  Is a default password also installed?
> 
> I certainly hope not, but it makes me nervous.

See for yourself:

# grep mysql /etc/shadow
mysql:!!:15404::

Regards,
Leonard.

-- 
mount -t life -o ro /dev/dna /genetic/research


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql

[Steve Clark]

Hello List,

I see that installing mysql in Centos 6.x creates a mysql user
with a login shell of /bin/bash.  Is a default password also installed?

I certainly hope not, but it makes me nervous.

-- 
Stephen Clark
*NetWolves*
Director of Technology
Phone: 813-579-3200
Fax: 813-882-0209
Email: steve.cl...@netwolves.com
http://www.netwolves.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql secure installation and multi mysqld

[Bob Hoffman]

On 5/17/2012 5:30 AM, Leon Jacobs wrote:
> On Thu, May 17, 2012 at 5:44 AM, Bob Hoffman  wrote:
>> is it possible to make each one of those instances a replicate/slave of
>> a different master
>> (thus a backup mysql server in a way?)
> I just recently finished playing with this. I ended up brining up a
> dedicated VM for this "backup mysql server". I then created 2 new init
> scripts (based off the mysqld one) and modified both to point to their
> respective my.cnf's (with auto startup via chkconfig etc etc). Both
> have separate data directories, run off 2 separate ports (and socket
> files) - which are fire walled anyways as I have no requirement for
> this to be open on the slaves and replicate 2 separate masters :)
>
> L.
>
Okay, so it looks doable.
My plan is to add the instances inside of the my.cnf and then use 
mysqld_multi to run them as slaves
to separate website masters...

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql secure installation and multi mysqld

[Leon Jacobs]

On Thu, May 17, 2012 at 5:44 AM, Bob Hoffman  wrote:
> is it possible to make each one of those instances a replicate/slave of
> a different master
> (thus a backup mysql server in a way?)

I just recently finished playing with this. I ended up brining up a
dedicated VM for this "backup mysql server". I then created 2 new init
scripts (based off the mysqld one) and modified both to point to their
respective my.cnf's (with auto startup via chkconfig etc etc). Both
have separate data directories, run off 2 separate ports (and socket
files) - which are fire walled anyways as I have no requirement for
this to be open on the slaves and replicate 2 separate masters :)

L.

-- 
Sent using electronic mail
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql secure installation and multi mysqld

[Bob Hoffman]

If you adjust the my.cnf file to make separate instances of mysql.

How would you go through the mysql secure installation? Is is possible 
or must it all be done manually
for each one?

is it possible to make each one of those instances a replicate/slave of 
a different master
(thus a backup mysql server in a way?)

working on this now, but having issues getting it to work.
(centos 6)
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mysql-bin files - filling up the space.

[Giovanni Tirloni]

On Thu, May 3, 2012 at 11:23 AM, Prabhpal S. Mavi
 wrote:
>
>
> Dear All Greetings,
>
> i am seeking help from guys with mysql knowledge. i can see lot of these
> files in mysql directory. And they are eventually filling up the space on
> the server. what these files are? some exists with very old time stamps.
> such as February 2012. can these be deleted?
>
> This Number Starts From ---mysql-bin.01
>
> -rw-rw  1 mysql  mysql   159M Apr 25 12:24 mysql-bin.000197
> -rw-rw  1 mysql  mysql     5M Apr 26 00:00 mysql-bin.000198
> -rw-rw  1 mysql  mysql   8.1M Apr 27 00:00 mysql-bin.000200
>
> -rw-rw  1 mysql  mysql   125M MAY 27 20:29 mysql-bin.000230
>

The first hit here has a very detailed description of that:
https://www.google.com.br/search?q=mysql-bin

In case your results are different:
http://www.cyberciti.biz/faq/what-is-mysql-binary-log/

--
Giovanni
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] mysql-bin files - filling up the space.

[Prabhpal S. Mavi]

Dear All Greetings,

i am seeking help from guys with mysql knowledge. i can see lot of these
files in mysql directory. And they are eventually filling up the space on
the server. what these files are? some exists with very old time stamps.
such as February 2012. can these be deleted?

This Number Starts From ---mysql-bin.01

-rw-rw  1 mysql  mysql   159M Apr 25 12:24 mysql-bin.000197
-rw-rw  1 mysql  mysql 5M Apr 26 00:00 mysql-bin.000198
-rw-rw  1 mysql  mysql   8.1M Apr 27 00:00 mysql-bin.000200

-rw-rw  1 mysql  mysql   125M MAY 27 20:29 mysql-bin.000230



Prabh S. Mavi



___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL query profiling

[Johnny Hughes]

On 03/22/2012 11:18 AM, Andy Taylor wrote:
> Hi,
>
> I recently upgraded to the latest MySQL 5.0.95 package as part of the 
> CentOS 5.8 upgrade. I use MySQL query profiling on a regular basis for 
> performance testing. However, if I try to use any of the profiling 
> functionality I get this error:
>
> ERROR 1289 (HY000): The 'SHOW PROFILE' feature is disabled; you need 
> MySQL built with 'enable-profiling' to have it working
>
> Oddly enough, profiling is enabled in the MySQL package bundled with 
> CentOS 6.2, and was enabled in the package before on CentOS 5 (5.0.77 I 
> think)
>
> I'm assuming this is probably an upstream decision, but does anyone 
> know of any way around this other than compiling my own RPM/using a 
> different RPM from another repository? Query profiling is essential 
> functionality in my opinion, so I'm surprised it hasn't been enabled in 
> the package.
>
> Thanks,
>
> Andy

After verifying this issue, we have made a bugs entry for CentOS:
http://bugs.centos.org/view.php?id=5615

and upstream:
https://bugzilla.redhat.com/show_bug.cgi?id=806365

We can build a version that has this enabled if there is a big need or a
long delay from upstream.




signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL query profiling

[Leonard den Ottolander]

Hello Johnny, Andy,

On Thu, 2012-03-22 at 12:01 -0500, Johnny Hughes wrote:
> On 03/22/2012 11:46 AM, Johnny Hughes wrote:
> OK ... I have verified that it is now turned off and that it was on in
> 5.0.77 
> 
> from 5.0.95 build log:
> <..>
> checking whether query profiling should be included... no
> <..> (tests)
> profiling  [ skipped ]   Test requires: 'have_profiling'
> 
> and from 5.0.77 build log:
> (there is nothing about checking if it is to be included ... )
> <..> (tests)
> profiling  [ pass ]109
> 
> So, profiling was indeed in 5.0.77 and it is indeed not in 5.0.95
> 
> The question is, what does RHEL do ... anyone out there who can test the
> behavior in RHEL?

Probably the same as it seems to be an upstream (as in MySQL) issue:
http://bugs.mysql.com/bug.php?id=48344

Regards,
Leonard.

-- 
mount -t life -o ro /dev/dna /genetic/research


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL query profiling

[Karanbir Singh]

On 03/22/2012 05:01 PM, Johnny Hughes wrote:
> 
> So, profiling was indeed in 5.0.77 and it is indeed not in 5.0.95

The default in configure for 5.0.77 was to have it turned on, whereas in
5.0.95 the default has been to turn it off :


+{ $as_echo "$as_me:$LINENO: checking whether query profiling should be
included" >&5
+$as_echo_n "checking whether query profiling should be included... " >&6; }
 # Check whether --enable-profiling was given.
 if test "${enable_profiling+set}" = set; then
   enableval=$enable_profiling;  ENABLED_PROFILING=$enableval
 else
-   ENABLED_PROFILING=yes
+   ENABLED_PROFILING=no
 fi

Since there is no --enable-profiling in the .spec file, its going to go
with the change in behaviour of the configure script.

> The question is, what does RHEL do ... anyone out there who can test the
> behavior in RHEL?

I'm guessing its the same there as well. But it would be interesting to
get confirmation ( and therefore a bug report on bugs.centos.org and
cross references to bugzilla.redhat.com )


-- 
Karanbir Singh
+44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh
ICQ: 2522219| Yahoo IM: z00dax  | Gtalk: z00dax
GnuPG Key : http://www.karan.org/publickey.asc
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL query profiling

[Johnny Hughes]

On 03/22/2012 11:46 AM, Johnny Hughes wrote:
> On 03/22/2012 11:32 AM, Johnny Hughes wrote:
>> On 03/22/2012 11:18 AM, Andy Taylor wrote:
>>> Hi,
>>>
>>> I recently upgraded to the latest MySQL 5.0.95 package as part of the 
>>> CentOS 5.8 upgrade. I use MySQL query profiling on a regular basis for 
>>> performance testing. However, if I try to use any of the profiling 
>>> functionality I get this error:
>>>
>>> ERROR 1289 (HY000): The 'SHOW PROFILE' feature is disabled; you need 
>>> MySQL built with 'enable-profiling' to have it working
>>>
>>> Oddly enough, profiling is enabled in the MySQL package bundled with 
>>> CentOS 6.2, and was enabled in the package before on CentOS 5 (5.0.77 I 
>>> think)
>>>
>>> I'm assuming this is probably an upstream decision, but does anyone 
>>> know of any way around this other than compiling my own RPM/using a 
>>> different RPM from another repository? Query profiling is essential 
>>> functionality in my opinion, so I'm surprised it hasn't been enabled in 
>>> the package.
>>>
>> I will double check the spec file and see if they have turn it off on
>> purpose and post what I find out.
> This is the difference beween the spec files:
> =
> --- mysql.spec(revision 4056)
> +++ mysql.spec(revision 4522)
> @@ -1,11 +1,11 @@
>  Name: mysql
> -Version: 5.0.77
> -Release: 4%{?dist}.6
> +Version: 5.0.95
> +Release: 1%{?dist}.1
>  Summary: MySQL client programs and shared libraries
>  Group: Applications/Databases
>  URL: http://www.mysql.com
>  # exceptions allow client libraries to be linked with most open source SW,
> -# not only GPL code.
> +# not only GPL code.  See README.mysql-license
>  License: GPLv2 with exceptions
>  
>  # Regression tests take a long time, you can skip 'em with this
> @@ -16,6 +16,7 @@
>  Source3: my.cnf
>  Source4: scriptstub.c
>  Source5: my_config.h
> +Source7: README.mysql-license
>  # Working around perl dependency checking bug in rpm FTTB. Remove later.
>  Source999: filter-requires-mysql.sh
>  Patch1: mysql-libdir.patch
> @@ -28,36 +29,19 @@
>  Patch8: mysql-install-test.patch
>  Patch9: mysql-bdb-link.patch
>  Patch10: mysql-strmov.patch
> -Patch11: mysql-html-bug.patch
> +Patch12: mysql-s390-truncate.patch
>  Patch13: mysql-no-dbug.patch
>  Patch15: mysql-stack-guard.patch
>  Patch16: mysql-expired-certs.patch
> -Patch17: mysql-format-string.patch
> -Patch18: mysql-name-const.patch
> -Patch19: mysql-cve-2009-4019.patch
> -Patch20: mysql-cve-2009-4028.patch
> -Patch21: mysql-cve-2009-4030.patch
>  Patch22: mysql-cve-2010-1626.patch
> -Patch23: mysql-cve-2010-1848_1850.patch
> -Patch24: mysql-cve-2010-3677.patch
>  Patch25: mysql-cve-2010-3680.patch
>  Patch26: mysql-cve-2010-3681.patch
> -Patch27: mysql-cve-2010-3682.patch
> -Patch28: mysql-cve-2010-3833.patch
> -Patch29: mysql-cve-2010-3835.patch
> -Patch30: mysql-cve-2010-3836.patch
> -Patch31: mysql-cve-2010-3837.patch
> -Patch32: mysql-cve-2010-3838.patch
>  Patch33: mysql-cve-2010-3839.patch
> -Patch34: mysql-cve-2010-3840.patch
> -Patch35: mysql-large-blobs.patch
> -Patch36: mysql-explain-crash.patch
> -Patch37: mysql-bug42037.patch
>  Patch38: mysql-more-blobs.patch
>  
>  BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root
> -Prereq: /sbin/ldconfig, /sbin/install-info, grep, fileutils, chkconfig
> -BuildRequires: gperf, perl, readline-devel, openssl-devel
> +Requires: /sbin/ldconfig, /sbin/install-info, grep, fileutils, chkconfig
> +BuildRequires: perl, readline-devel, openssl-devel
>  BuildRequires: gcc-c++, ncurses-devel, zlib-devel
>  BuildRequires: libtool automake autoconf gawk
>  # make test requires time
> @@ -81,7 +65,7 @@
>  
>  Summary: The MySQL server and related files
>  Group: Applications/Databases
> -Prereq: /sbin/chkconfig, /usr/sbin/useradd
> +Requires: /sbin/chkconfig, /usr/sbin/useradd
>  Requires: %{name} = %{version}-%{release}
>  Requires: sh-utils
>  # mysqlhotcopy needs DBI/DBD support
> @@ -145,31 +129,14 @@
>  %patch8 -p1
>  %patch9 -p1
>  %patch10 -p1
> -%patch11 -p1
> +%patch12 -p1
>  %patch13 -p1
>  %patch15 -p1
>  %patch16 -p1
> -%patch17 -p1
> -%patch18 -p1
> -%patch19 -p1
> -%patch20 -p1
> -%patch21 -p1
>  %patch22 -p1
> -%patch23 -p1
> -%patch24 -p1
>  %patch25 -p1
>  %patch26 -p1
> -%patch27 -p1
> -%patch28 -p1
> -%patch29 -p1
> -%patch30 -p1
> -%patch31 -p1
> -%patch32 -p1
>  %patch33 -p1
> -%patch34 -p1
> -%patch35 -p1
> -%patch36 -p1
> -%patch37 -p1
>  %patch38 -p1
>  
>  libtoolize --force
> @@ -312,6 +279,9 @@
>  mkdir -p $RPM_BUILD_ROOT/etc/ld.so.conf.d
>  echo "%{_libdir}/mysql" >
> $RPM_BUILD_ROOT/etc/ld.so.conf.d/%{name}-%{_arch}.conf
>  
> +# copy additional docs into build tree so %%doc will find them
> +cp %{SOURCE7} README.mysql-license
> +
>  %clean
>  rm -rf $RPM_BUILD_ROOT
>  
> @@ -354,7 +324,7 @@
>  
>  %files
>  %defattr(-,root,root)
> -%doc README COPYING EXCEPTIONS-CLIENT
> +%doc README COPYING README.mysql-license
>  
>  %{_bindir}/msql2mysql

Re: [CentOS] MySQL query profiling

[Johnny Hughes]

On 03/22/2012 11:32 AM, Johnny Hughes wrote:
> On 03/22/2012 11:18 AM, Andy Taylor wrote:
>> Hi,
>>
>> I recently upgraded to the latest MySQL 5.0.95 package as part of the 
>> CentOS 5.8 upgrade. I use MySQL query profiling on a regular basis for 
>> performance testing. However, if I try to use any of the profiling 
>> functionality I get this error:
>>
>> ERROR 1289 (HY000): The 'SHOW PROFILE' feature is disabled; you need 
>> MySQL built with 'enable-profiling' to have it working
>>
>> Oddly enough, profiling is enabled in the MySQL package bundled with 
>> CentOS 6.2, and was enabled in the package before on CentOS 5 (5.0.77 I 
>> think)
>>
>> I'm assuming this is probably an upstream decision, but does anyone 
>> know of any way around this other than compiling my own RPM/using a 
>> different RPM from another repository? Query profiling is essential 
>> functionality in my opinion, so I'm surprised it hasn't been enabled in 
>> the package.
>>
> I will double check the spec file and see if they have turn it off on
> purpose and post what I find out.

This is the difference beween the spec files:
=
--- mysql.spec(revision 4056)
+++ mysql.spec(revision 4522)
@@ -1,11 +1,11 @@
 Name: mysql
-Version: 5.0.77
-Release: 4%{?dist}.6
+Version: 5.0.95
+Release: 1%{?dist}.1
 Summary: MySQL client programs and shared libraries
 Group: Applications/Databases
 URL: http://www.mysql.com
 # exceptions allow client libraries to be linked with most open source SW,
-# not only GPL code.
+# not only GPL code.  See README.mysql-license
 License: GPLv2 with exceptions
 
 # Regression tests take a long time, you can skip 'em with this
@@ -16,6 +16,7 @@
 Source3: my.cnf
 Source4: scriptstub.c
 Source5: my_config.h
+Source7: README.mysql-license
 # Working around perl dependency checking bug in rpm FTTB. Remove later.
 Source999: filter-requires-mysql.sh
 Patch1: mysql-libdir.patch
@@ -28,36 +29,19 @@
 Patch8: mysql-install-test.patch
 Patch9: mysql-bdb-link.patch
 Patch10: mysql-strmov.patch
-Patch11: mysql-html-bug.patch
+Patch12: mysql-s390-truncate.patch
 Patch13: mysql-no-dbug.patch
 Patch15: mysql-stack-guard.patch
 Patch16: mysql-expired-certs.patch
-Patch17: mysql-format-string.patch
-Patch18: mysql-name-const.patch
-Patch19: mysql-cve-2009-4019.patch
-Patch20: mysql-cve-2009-4028.patch
-Patch21: mysql-cve-2009-4030.patch
 Patch22: mysql-cve-2010-1626.patch
-Patch23: mysql-cve-2010-1848_1850.patch
-Patch24: mysql-cve-2010-3677.patch
 Patch25: mysql-cve-2010-3680.patch
 Patch26: mysql-cve-2010-3681.patch
-Patch27: mysql-cve-2010-3682.patch
-Patch28: mysql-cve-2010-3833.patch
-Patch29: mysql-cve-2010-3835.patch
-Patch30: mysql-cve-2010-3836.patch
-Patch31: mysql-cve-2010-3837.patch
-Patch32: mysql-cve-2010-3838.patch
 Patch33: mysql-cve-2010-3839.patch
-Patch34: mysql-cve-2010-3840.patch
-Patch35: mysql-large-blobs.patch
-Patch36: mysql-explain-crash.patch
-Patch37: mysql-bug42037.patch
 Patch38: mysql-more-blobs.patch
 
 BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root
-Prereq: /sbin/ldconfig, /sbin/install-info, grep, fileutils, chkconfig
-BuildRequires: gperf, perl, readline-devel, openssl-devel
+Requires: /sbin/ldconfig, /sbin/install-info, grep, fileutils, chkconfig
+BuildRequires: perl, readline-devel, openssl-devel
 BuildRequires: gcc-c++, ncurses-devel, zlib-devel
 BuildRequires: libtool automake autoconf gawk
 # make test requires time
@@ -81,7 +65,7 @@
 
 Summary: The MySQL server and related files
 Group: Applications/Databases
-Prereq: /sbin/chkconfig, /usr/sbin/useradd
+Requires: /sbin/chkconfig, /usr/sbin/useradd
 Requires: %{name} = %{version}-%{release}
 Requires: sh-utils
 # mysqlhotcopy needs DBI/DBD support
@@ -145,31 +129,14 @@
 %patch8 -p1
 %patch9 -p1
 %patch10 -p1
-%patch11 -p1
+%patch12 -p1
 %patch13 -p1
 %patch15 -p1
 %patch16 -p1
-%patch17 -p1
-%patch18 -p1
-%patch19 -p1
-%patch20 -p1
-%patch21 -p1
 %patch22 -p1
-%patch23 -p1
-%patch24 -p1
 %patch25 -p1
 %patch26 -p1
-%patch27 -p1
-%patch28 -p1
-%patch29 -p1
-%patch30 -p1
-%patch31 -p1
-%patch32 -p1
 %patch33 -p1
-%patch34 -p1
-%patch35 -p1
-%patch36 -p1
-%patch37 -p1
 %patch38 -p1
 
 libtoolize --force
@@ -312,6 +279,9 @@
 mkdir -p $RPM_BUILD_ROOT/etc/ld.so.conf.d
 echo "%{_libdir}/mysql" >
$RPM_BUILD_ROOT/etc/ld.so.conf.d/%{name}-%{_arch}.conf
 
+# copy additional docs into build tree so %%doc will find them
+cp %{SOURCE7} README.mysql-license
+
 %clean
 rm -rf $RPM_BUILD_ROOT
 
@@ -354,7 +324,7 @@
 
 %files
 %defattr(-,root,root)
-%doc README COPYING EXCEPTIONS-CLIENT
+%doc README COPYING README.mysql-license
 
 %{_bindir}/msql2mysql
 %{_bindir}/mysql
@@ -519,6 +489,11 @@
 %{_mandir}/man1/mysql_client_test.1*
 
 %changelog
+* Sat Feb  4 2012 Tom Lane  5.0.95-1.el5_7.1
+- Update to 5.0.95, to get the last upstream bugfixes in this release
series
+  including numerous CVEs announced in January 2012
+Resolves: #787140
+
 * Thu Apr 28 2011 Tom Lane  5.0.77

Re: [CentOS] MySQL query profiling

[Johnny Hughes]

On 03/22/2012 11:18 AM, Andy Taylor wrote:
> Hi,
>
> I recently upgraded to the latest MySQL 5.0.95 package as part of the 
> CentOS 5.8 upgrade. I use MySQL query profiling on a regular basis for 
> performance testing. However, if I try to use any of the profiling 
> functionality I get this error:
>
> ERROR 1289 (HY000): The 'SHOW PROFILE' feature is disabled; you need 
> MySQL built with 'enable-profiling' to have it working
>
> Oddly enough, profiling is enabled in the MySQL package bundled with 
> CentOS 6.2, and was enabled in the package before on CentOS 5 (5.0.77 I 
> think)
>
> I'm assuming this is probably an upstream decision, but does anyone 
> know of any way around this other than compiling my own RPM/using a 
> different RPM from another repository? Query profiling is essential 
> functionality in my opinion, so I'm surprised it hasn't been enabled in 
> the package.
>

I will double check the spec file and see if they have turn it off on
purpose and post what I find out.



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL query profiling

[Andy Taylor]

Hi,

I recently upgraded to the latest MySQL 5.0.95 package as part of the 
CentOS 5.8 upgrade. I use MySQL query profiling on a regular basis for 
performance testing. However, if I try to use any of the profiling 
functionality I get this error:

ERROR 1289 (HY000): The 'SHOW PROFILE' feature is disabled; you need 
MySQL built with 'enable-profiling' to have it working

Oddly enough, profiling is enabled in the MySQL package bundled with 
CentOS 6.2, and was enabled in the package before on CentOS 5 (5.0.77 I 
think)

I'm assuming this is probably an upstream decision, but does anyone 
know of any way around this other than compiling my own RPM/using a 
different RPM from another repository? Query profiling is essential 
functionality in my opinion, so I'm surprised it hasn't been enabled in 
the package.

Thanks,

Andy
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Adam Tauno Williams]

On Mon, 2012-02-13 at 18:59 -0500, Boris Epstein wrote:
> Here we go:
> [root@gala ~]# ls -l /proc/18702/fd
> total 0
...
> lrwx-- 1 root root 64 Feb 13 18:58 7 -> /home/mysql/tmp/ibE0DGCO
> (deleted)
> lrwx-- 1 root root 64 Feb 13 18:58 8 ->
> /home/mysql/mysql_data/ib_logfile0
> lrwx-- 1 root root 64 Feb 13 18:58 9 ->
> /home/mysql/mysql_data/ib_logfile1
> [root@gala ~]#
> Now the files of interest are the ones marked deleted in /home/mysql/tmp .
> Now how do I tell what their size is?

I wouldn't bother pursuing this angle further;  I am not an MySQL admin,
but it must provide performance reporting, statistics, and explain
information [PostgreSQL does, as do most other database engines].  Check
the MySQL docs for how you monitor the utilization of temp spaces.  If
it can't provide this information choose another solution.

It is best not to dig around under engines.  Use the tools and let them
tell you,  otherwise you will build around implementation details.

-- 
System & Network Administrator [ LPI & NCLA ]

OpenGroupware Developer 
Adam Tauno Williams

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Boris Epstein]

On Mon, Feb 13, 2012 at 6:53 PM, Ljubomir Ljubojevic wrote:

> On 02/13/2012 11:31 PM, Boris Epstein wrote:
> >>
> > In fact, I did look in /proc/{pid#}/fd and found the file names, thanks!
> > Coud quite figure out the size of those invisible files - but no matter,
> > hopefully I've got enough room.
> >
>
> ls -l /proc/{pid#}/fd
>
> --
>
> Ljubomir Ljubojevic
> (Love is in the Air)
> PL Computers
> Serbia, Europe
>
> Google is the Mother, Google is the Father, and traceroute is your
> trusty Spiderman...
> StarOS, Mikrotik and CentOS/RHEL/Linux consultant
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>

Ljubomir,

Here we go:

[root@gala ~]# ls -l /proc/18702/fd
total 0
lr-x-- 1 root root 64 Feb 13 18:58 0 -> /dev/null
lrwx-- 1 root root 64 Feb 13 18:58 1 -> /var/log/mysqld.log
lrwx-- 1 root root 64 Feb 13 18:58 10 -> socket:[1245288]
lrwx-- 1 root root 64 Feb 13 18:58 11 -> /home/mysql/tmp/ibEvxyKc
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 12 -> socket:[1245289]
lrwx-- 1 root root 64 Feb 13 18:58 13 ->
/home/mysql/mysql_data/mysql/host.MYI
lrwx-- 1 root root 64 Feb 13 18:58 14 ->
/home/mysql/mysql_data/mysql/host.MYD
lrwx-- 1 root root 64 Feb 13 18:58 15 ->
/home/mysql/mysql_data/mysql/user.MYI
lrwx-- 1 root root 64 Feb 13 18:58 16 ->
/home/mysql/mysql_data/mysql/user.MYD
lrwx-- 1 root root 64 Feb 13 18:58 17 ->
/home/mysql/mysql_data/mysql/db.MYI
lrwx-- 1 root root 64 Feb 13 18:58 18 ->
/home/mysql/mysql_data/mysql/db.MYD
lrwx-- 1 root root 64 Feb 13 18:58 19 ->
/home/mysql/mysql_data/mysql/tables_priv.MYI
lrwx-- 1 root root 64 Feb 13 18:58 2 -> /var/log/mysqld.log
lrwx-- 1 root root 64 Feb 13 18:58 20 ->
/home/mysql/mysql_data/mysql/tables_priv.MYD
lrwx-- 1 root root 64 Feb 13 18:58 21 ->
/home/mysql/mysql_data/mysql/columns_priv.MYI
lrwx-- 1 root root 64 Feb 13 18:58 22 ->
/home/mysql/mysql_data/mysql/columns_priv.MYD
lrwx-- 1 root root 64 Feb 13 18:58 23 ->
/home/mysql/mysql_data/mysql/procs_priv.MYI
lrwx-- 1 root root 64 Feb 13 18:58 24 ->
/home/mysql/mysql_data/mysql/procs_priv.MYD
lrwx-- 1 root root 64 Feb 13 18:58 25 ->
/home/mysql/mysql_data/mysql/servers.MYI
lrwx-- 1 root root 64 Feb 13 18:58 26 ->
/home/mysql/mysql_data/mysql/servers.MYD
lrwx-- 1 root root 64 Feb 13 18:58 27 ->
/home/mysql/mysql_data/mysql/event.MYI
lrwx-- 1 root root 64 Feb 13 18:58 28 ->
/home/mysql/mysql_data/mysql/event.MYD
lrwx-- 1 root root 64 Feb 13 18:58 29 -> socket:[1245325]
lrwx-- 1 root root 64 Feb 13 18:58 3 -> /home/mysql/mysql_data/ibdata1
lrwx-- 1 root root 64 Feb 13 18:58 30 ->
/home/mysql/mysql_data/text_sort/local_list.MYI
lrwx-- 1 root root 64 Feb 13 18:58 31 ->
/home/mysql/mysql_data/text_sort/local_list.MYD
lrwx-- 1 root root 64 Feb 13 18:58 32 ->
/home/mysql/mysql_data/text_sort/local_list_sorted.MYI
lrwx-- 1 root root 64 Feb 13 18:58 33 ->
/home/mysql/mysql_data/text_sort/local_list_sorted.MYD
lrwx-- 1 root root 64 Feb 13 18:58 34 ->
/home/mysql/mysql_data/text_sort/tar_list.MYI
lrwx-- 1 root root 64 Feb 13 18:58 35 ->
/home/mysql/mysql_data/text_sort/tar_list.MYD
lrwx-- 1 root root 64 Feb 13 18:58 36 -> /home/mysql/tmp/MYbUgIFT
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 37 -> /home/mysql/tmp/MYizt5rL
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 38 -> socket:[1249477]
lrwx-- 1 root root 64 Feb 13 18:58 39 -> socket:[1340698]
lrwx-- 1 root root 64 Feb 13 18:58 4 -> /home/mysql/tmp/ibvjZa8E
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 40 ->
/home/mysql/mysql_data/text_sort/local_list.MYD
lrwx-- 1 root root 64 Feb 13 18:58 41 ->
/home/mysql/mysql_data/text_sort/local_list_sorted.MYD
lrwx-- 1 root root 64 Feb 13 18:58 5 -> /home/mysql/tmp/ib08cXW2
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 6 -> /home/mysql/tmp/ibtcHJLq
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 7 -> /home/mysql/tmp/ibE0DGCO
(deleted)
lrwx-- 1 root root 64 Feb 13 18:58 8 ->
/home/mysql/mysql_data/ib_logfile0
lrwx-- 1 root root 64 Feb 13 18:58 9 ->
/home/mysql/mysql_data/ib_logfile1
[root@gala ~]#

Now the files of interest are the ones marked deleted in /home/mysql/tmp .
Now how do I tell what their size is?

Thanks.

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Ljubomir Ljubojevic]

On 02/13/2012 11:31 PM, Boris Epstein wrote:
>>
> In fact, I did look in /proc/{pid#}/fd and found the file names, thanks!
> Coud quite figure out the size of those invisible files - but no matter,
> hopefully I've got enough room.
>

ls -l /proc/{pid#}/fd

-- 

Ljubomir Ljubojevic
(Love is in the Air)
PL Computers
Serbia, Europe

Google is the Mother, Google is the Father, and traceroute is your
trusty Spiderman...
StarOS, Mikrotik and CentOS/RHEL/Linux consultant
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Les Mikesell]

On Mon, Feb 13, 2012 at 4:06 PM, Boris Epstein  wrote:

> > >
> I haven't thought of it this way. In fact it matches with this description:
>
> http://dev.mysql.com/doc/refman/5.1/en/temporary-files.html
>
> I can kind of see the advantages; the disadvantages, of course, are that a
> major transaction can not be resumed in case the mysqld process is stopped.
>
>
I don't think it would make much sense to try to finish a query after a
restart since at that point the temp table might not reflect the real data
anyway.   Have you tried adding an index on the fields in your query so it
doesn't have to sort everything when you make the query?   Mysql isn't too
bright about optimizing 3-table joins, but with one or 2 and pre-indexed
fields it shouldn't need much time or extra space.

-- 
  Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Boris Epstein]

>
>
>
> If the process is stopped the transaction cannot be resumed for a myriad
> reasons;  loss of the temporary file is a trivial concern.
>
> --
> System & Network Administrator [ LPI & NCLA ]
> 
> OpenGroupware Developer 
> Adam Tauno Williams
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>

True - although if you somehow track your progress within the transaction
making it resumable could be possible.

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Adam Tauno Williams]

On Mon, 2012-02-13 at 17:06 -0500, Boris Epstein wrote:
> On Mon, Feb 13, 2012 at 4:58 PM, Adam Tauno Williams  > wrote:
> > On Mon, 2012-02-13 at 16:45 -0500, Boris Epstein wrote:
> > > Hello listmates,
> > > I got a rather strange situation that I can't quite make sense of. OK,
> > I've
> > > got a very large data file to sort (hundreds of millions of lines) and I
> > > decided to use MySQL for the purpose. I inserted the stuff into a table
> > > easily enough. Then I decided to sort it and got stuck as it turned out
> > > that MySQL, unless specifically configured to do otherwise, puts
> > temporary
> > > files in /tmp which simply was not sufficiently large. Then I changed
> > that
> > > directory to a partition that had more space (let's call the new temp
> > > driectory /home/big-temp) and now as I am running the query aimed at
> > > sorting the data it seems like space, according to the df, is no longer
> > > being used up under / (which was there /tmp was) but is now being used up
> > > in the right partition ( /home, the large one). Yet /home/big-temp is
> > still
> > > empty!
> > > So how is that possible?
> > Easy.  It is using temporary files the *correct* way.
> > 1. Open file
> > 2. Unlink file
> > 3. Use file
> > 4. Close file
> > This means (a) even if the process abends the resources allocated to the
> > file are released and (b) an external process can't see [or modify] the
> > temporary file.
> > When a file is unlinked it remains 'active' until all references to the
> > file are released - but the daemon is still holding a reference [because
> > it is using the file].
> > There is a file there, but nobody, not even root, can see it.
> > Actually you can; if you look in /proc/{pid#}/fd ...
> I haven't thought of it this way. In fact it matches with this description:
> http://dev.mysql.com/doc/refman/5.1/en/temporary-files.html
> I can kind of see the advantages; the disadvantages, of course, are that a
> major transaction can not be resumed in case the mysqld process is stopped.

If the process is stopped the transaction cannot be resumed for a myriad
reasons;  loss of the temporary file is a trivial concern.

-- 
System & Network Administrator [ LPI & NCLA ]

OpenGroupware Developer 
Adam Tauno Williams

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Boris Epstein]

>
>
>
> Easy.  It is using temporary files the *correct* way.
>
> 1. Open file
> 2. Unlink file
> 3. Use file
> 4. Close file
>
> This means (a) even if the process abends the resources allocated to the
> file are released and (b) an external process can't see [or modify] the
> temporary file.
>
> When a file is unlinked it remains 'active' until all references to the
> file are released - but the daemon is still holding a reference [because
> it is using the file].
>
> There is a file there, but nobody, not even root, can see it.
>
> Actually you can; if you look in /proc/{pid#}/fd ...
>
> --
> System & Network Administrator [ LPI & NCLA ]
> 
> OpenGroupware Developer 
> Adam Tauno Williams
>
>
In fact, I did look in /proc/{pid#}/fd and found the file names, thanks!
Coud quite figure out the size of those invisible files - but no matter,
hopefully I've got enough room.

Thanks.

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Boris Epstein]

On Mon, Feb 13, 2012 at 4:58 PM, Adam Tauno Williams  wrote:

> On Mon, 2012-02-13 at 16:45 -0500, Boris Epstein wrote:
> > Hello listmates,
> > I got a rather strange situation that I can't quite make sense of. OK,
> I've
> > got a very large data file to sort (hundreds of millions of lines) and I
> > decided to use MySQL for the purpose. I inserted the stuff into a table
> > easily enough. Then I decided to sort it and got stuck as it turned out
> > that MySQL, unless specifically configured to do otherwise, puts
> temporary
> > files in /tmp which simply was not sufficiently large. Then I changed
> that
> > directory to a partition that had more space (let's call the new temp
> > driectory /home/big-temp) and now as I am running the query aimed at
> > sorting the data it seems like space, according to the df, is no longer
> > being used up under / (which was there /tmp was) but is now being used up
> > in the right partition ( /home, the large one). Yet /home/big-temp is
> still
> > empty!
> > So how is that possible?
>
> Easy.  It is using temporary files the *correct* way.
>
> 1. Open file
> 2. Unlink file
> 3. Use file
> 4. Close file
>
> This means (a) even if the process abends the resources allocated to the
> file are released and (b) an external process can't see [or modify] the
> temporary file.
>
> When a file is unlinked it remains 'active' until all references to the
> file are released - but the daemon is still holding a reference [because
> it is using the file].
>
> There is a file there, but nobody, not even root, can see it.
>
> Actually you can; if you look in /proc/{pid#}/fd ...
>
> --
> System & Network Administrator [ LPI & NCLA ]
> 
> OpenGroupware Developer 
> Adam Tauno Williams
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos


Adam, thanks!

I haven't thought of it this way. In fact it matches with this description:

http://dev.mysql.com/doc/refman/5.1/en/temporary-files.html

I can kind of see the advantages; the disadvantages, of course, are that a
major transaction can not be resumed in case the mysqld process is stopped.

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] MySQL/file system question

[Adam Tauno Williams]

On Mon, 2012-02-13 at 16:45 -0500, Boris Epstein wrote:
> Hello listmates,
> I got a rather strange situation that I can't quite make sense of. OK, I've
> got a very large data file to sort (hundreds of millions of lines) and I
> decided to use MySQL for the purpose. I inserted the stuff into a table
> easily enough. Then I decided to sort it and got stuck as it turned out
> that MySQL, unless specifically configured to do otherwise, puts temporary
> files in /tmp which simply was not sufficiently large. Then I changed that
> directory to a partition that had more space (let's call the new temp
> driectory /home/big-temp) and now as I am running the query aimed at
> sorting the data it seems like space, according to the df, is no longer
> being used up under / (which was there /tmp was) but is now being used up
> in the right partition ( /home, the large one). Yet /home/big-temp is still
> empty!
> So how is that possible? 

Easy.  It is using temporary files the *correct* way.

1. Open file
2. Unlink file
3. Use file
4. Close file

This means (a) even if the process abends the resources allocated to the
file are released and (b) an external process can't see [or modify] the
temporary file.

When a file is unlinked it remains 'active' until all references to the
file are released - but the daemon is still holding a reference [because
it is using the file].

There is a file there, but nobody, not even root, can see it.

Actually you can; if you look in /proc/{pid#}/fd ...

-- 
System & Network Administrator [ LPI & NCLA ]

OpenGroupware Developer 
Adam Tauno Williams

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] MySQL/file system question

[Boris Epstein]

Hello listmates,

I got a rather strange situation that I can't quite make sense of. OK, I've
got a very large data file to sort (hundreds of millions of lines) and I
decided to use MySQL for the purpose. I inserted the stuff into a table
easily enough. Then I decided to sort it and got stuck as it turned out
that MySQL, unless specifically configured to do otherwise, puts temporary
files in /tmp which simply was not sufficiently large. Then I changed that
directory to a partition that had more space (let's call the new temp
driectory /home/big-temp) and now as I am running the query aimed at
sorting the data it seems like space, according to the df, is no longer
being used up under / (which was there /tmp was) but is now being used up
in the right partition ( /home, the large one). Yet /home/big-temp is still
empty!

So how is that possible? I know there are all kinds of file locking
possible depending on the circumstances, file system type, etc - but how
can it be that the files, if they exist, are not even visible to ls, even
"ls -a", even run by root?

Thanks in advance for any clarification.

Cheers,

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos