On Tuesday 20 April 2010 13:08:49 Dejan Muhamedagic wrote:
> Hi,
>
> On Tue, Apr 20, 2010 at 09:32:03AM +0200, Florian Haas wrote:
> > Hello,
> >
> > a little while back, Marian Marinov made an excellent contribution in
> > the form of proof-of-concept code that
,
> Florian
>
> On 2010-03-19 14:57, Marian Marinov wrote:
> > On Friday 19 March 2010 09:24:52 Florian Haas wrote:
> >> On 03/17/2010 03:47 PM, Marian Marinov wrote:
> >>> Hello,
> >>> I have fixed all portability issues with these patches.
>
On Friday 19 March 2010 09:24:52 Florian Haas wrote:
> On 03/17/2010 03:47 PM, Marian Marinov wrote:
> > Hello,
> > I have fixed all portability issues with these patches.
> >
> > I have added the check before seting readonly in mysql_start.
> > I have migrated ev
On Friday 19 March 2010 09:26:15 Florian Haas wrote:
> On 03/18/2010 09:43 AM, Marian Marinov wrote:
> > Hello,
> > My last list of patches is not working. I has a few mistakes in the code.
> > I have fixed them all, but can I submit one big patch fixing them or I
> >
On Thursday 18 March 2010 10:43:26 Marian Marinov wrote:
> Hello,
> My last list of patches is not working. I has a few mistakes in the code.
> I have fixed them all, but can I submit one big patch fixing them or I have
> to split it on parts?
>
> With these fixes it seams th
Hello,
My last list of patches is not working. I has a few mistakes in the code.
I have fixed them all, but can I submit one big patch fixing them or I have to
split it on parts?
With these fixes it seams that the RA is ready and working at least on my test
cluster.
--
Best regards,
Marian
# HG changeset patch
# User Marian Marinov
# Date 1268837185 -7200
# Branch mysql-ms
# Node ID 13e001a78d4ae612797024bf8abebe997bb6eed8
# Parent 94532073f46a421cf6ee753ee3d7013b9860ab61
Medium: RA: mysql: fixed bug with missing arguments from mysql show
master/slave status
The bug was that
# HG changeset patch
# User Marian Marinov
# Date 1268824675 -7200
# Branch mysql-ms
# Node ID 94532073f46a421cf6ee753ee3d7013b9860ab61
# Parent 207cd11a968a40381509d0ab14bb242f7e051c09
Low: RA: mysql: added note for .my.cnf in the metadata
Added note for the root password in the metadata
diff
# HG changeset patch
# User Marian Marinov
# Date 1268268329 -7200
# Branch mysql-ms
# Node ID 1c9dbac45db3a417be9d7a29bb15f96f35875277
# Parent c5304bc37c11e348f113b5701f2cb680d452deaa
Low: RA: mysql: updated preference calculation, better log checks
Swapped master_file to with master_log for
# HG changeset patch
# User Marian Marinov
# Date 1268822787 -7200
# Branch mysql-ms
# Node ID 207cd11a968a40381509d0ab14bb242f7e051c09
# Parent 1c9dbac45db3a417be9d7a29bb15f96f35875277
Low: RA: mysql: mysql_start, set readonly, only if the resource is MS
Set readonly, only if the resource is
# HG changeset patch
# User Marian Marinov
# Date 1268267087 -7200
# Branch mysql-ms
# Node ID cacb3e119fd9da6a7ea3c9f1d2d4f19d8e5b3bf6
# Parent 31ea5ae9c8e773edb89648cf482907ff3028d456
Low: RA: mysql: Set readonly
When starting the server, always make sure that the server is running in read
# HG changeset patch
# User Marian Marinov
# Date 1268267209 -7200
# Branch mysql-ms
# Node ID c5304bc37c11e348f113b5701f2cb680d452deaa
# Parent cacb3e119fd9da6a7ea3c9f1d2d4f19d8e5b3bf6
Low: RA: mysql: fixed invalid var slave_user in is_slave
I have changed the variable name but not at all
# HG changeset patch
# User Marian Marinov
# Date 1268266846 -7200
# Branch mysql-ms
# Node ID 31ea5ae9c8e773edb89648cf482907ff3028d456
# Parent 7d5becb0cbd1bf5b3558dd85dede99d7f3ef2a3c
Medium: RA: mysql: fixed the 'while read k v' problem
Fixed all places where I used while re
Hello,
I have fixed all portability issues with these patches.
I have added the check before seting readonly in mysql_start.
I have migrated eveything to use set.
Fixed the grep -P to use -E.
Added aditional documentation.
I did pull before emailing but I don't know why it sends 7 patches instead
epo. Please respond to Lars' comments.
>
> Cheers,
> Florian
>
> On 03/11/2010 01:45 AM, Marian Marinov wrote:
> > # HG changeset patch
> > # User Marian Marinov
> > # Date 1268266846 -7200
> > # Branch mysql-ms
> > # Node ID 31ea5ae9c8e773edb8
On Monday 15 March 2010 12:50:41 Florian Haas wrote:
> On second thought...
>
> On 03/11/2010 01:45 AM, Marian Marinov wrote:
> > diff -r 31ea5ae9c8e7 -r cacb3e119fd9 heartbeat/mysql
> > --- a/heartbeat/mysql Thu Mar 11 02:20:46 2010 +0200
> > +++ b/heartbeat/mysq
# HG changeset patch
# User Marian Marinov
# Date 1268268329 -7200
# Branch mysql-ms
# Node ID 1c9dbac45db3a417be9d7a29bb15f96f35875277
# Parent c5304bc37c11e348f113b5701f2cb680d452deaa
Low: RA: mysql: updated preference calculation, better log checks
Swapped master_file to with master_log for
# HG changeset patch
# User Marian Marinov
# Date 1268267209 -7200
# Branch mysql-ms
# Node ID c5304bc37c11e348f113b5701f2cb680d452deaa
# Parent cacb3e119fd9da6a7ea3c9f1d2d4f19d8e5b3bf6
Low: RA: mysql: fixed invalid var slave_user in is_slave
I have changed the variable name but not at all
# HG changeset patch
# User Marian Marinov
# Date 1268267087 -7200
# Branch mysql-ms
# Node ID cacb3e119fd9da6a7ea3c9f1d2d4f19d8e5b3bf6
# Parent 31ea5ae9c8e773edb89648cf482907ff3028d456
Low: RA: mysql: Set readonly
When starting the server, always make sure that the server is running in read
# HG changeset patch
# User Marian Marinov
# Date 1268266846 -7200
# Branch mysql-ms
# Node ID 31ea5ae9c8e773edb89648cf482907ff3028d456
# Parent 7d5becb0cbd1bf5b3558dd85dede99d7f3ef2a3c
Medium: RA: mysql: fixed the 'while read k v' problem
Fixed all places where I used while re
I tried to fix everything that you pointed guys.
Also I have updated the calculate_preference to have better handle for the log
files.
However I don't agree that we need to check the show binary logs and so I
didn't added such checks.
Best regards,
Marian
___
Wednesday 10 March 2010 23:00:44 Lars Ellenberg wrote:
> On Wed, Mar 10, 2010 at 10:53:54AM +0200, Marian Marinov wrote:
> > + mysql --password=$OCF_RESKEY_replication_passwd \
> > + --user=$OCF_RESKEY_replication_user \
> > + -h $master_host \
n you please pull from mysql-ms in my
> repo, and rebase your patch on that?
>
> As a general rule, please pull before you prepare and submit a patch; it
> makes merging a whole lot easier. :) Thanks!
>
> Cheers,
> Florian
>
> On 03/10/2010 09:53 AM, Marian Marinov wrote
# HG changeset patch
# User Marian Marinov
# Date 1268210842 -7200
# Branch mysql-ms
# Node ID 283dfb08a1ee6e378725e006f1a98467fd134cfb
# Parent 7d5becb0cbd1bf5b3558dd85dede99d7f3ef2a3c
Medium: RA: mysql: readonly and updated preference handling
Make all servers readonly during start and switch
lock? Can we help?
>
> Cheers,
> Florian
>
> On 03/02/2010 02:00 AM, Marian Marinov wrote:
> > I have introduced preference points calculation based on the log
> > position. I haven't tested the points, they will need some ajustments and
> > I have to test some situati
ly depending on whether they are
> invoked as "is_slave | mysql_demote 1" vs. just "is_slave | mysql_demote"?
>
> Thanks.
>
> Cheers,
> Florian
>
> On 03/02/2010 02:00 AM, Marian Marinov wrote:
> > I have introduced preference points calculation based on
# HG changeset patch
# User Marian Marinov
# Date 1267491017 -7200
# Branch mysql-ms
# Node ID b45e84f55882e98cbd57a71ab3e483f192945f04
# Parent 0c1c598c4f5ab4e758f7140b29aab86b2d8cd1a5
Medium: RA: mysql: fixed wrong parameters in is_slave
Fixed wrong comparation parameter $a(which is undefined
# HG changeset patch
# User Marian Marinov
# Date 1267490738 -7200
# Branch mysql-ms
# Node ID 0c1c598c4f5ab4e758f7140b29aab86b2d8cd1a5
# Parent 7c35f7fc2d3765060579a7c2328d8dc7dc684835
Medium: RA: mysql: added preference points calculation
Added preference points calculation during start, stop
I have introduced preference points calculation based on the log position.
I haven't tested the points, they will need some ajustments and I have to test
some situatiations which will be problematic for points calclulation.
I have fixed a problem that I introduced into is_slave function when I was
send the preference
and readonly patches.
Best regards,
Marian Marinov
___
Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
# HG changeset patch
# User Marian Marinov
# Date 1267453055 -7200
# Branch mysql-ms
# Node ID be4026374c679d0b873c39c30dcdc9eca89cd920
# Parent dc0dad7b935739987478e876d431e068f7ca1373
Low: RA: mysql: cleaned more bashisms
Changed the shell back to /bin/sh
Changed all == comparations to
# HG changeset patch
# User Marian Marinov
# Date 1267193043 -7200
# Branch mysql-ms
# Node ID 2da5a4ff4b7c17899376fc69a68dfc6b1d28c140
# Parent 0d95421fed862e5053e65f369f67f87ae172d5a6
Medium: RA: mysql: added replication_port variable
Added replication_port variable
Added initialization of
# HG changeset patch
# User Marian Marinov
# Date 1267196151 -7200
# Branch mysql-ms
# Node ID dc0dad7b935739987478e876d431e068f7ca1373
# Parent 2da5a4ff4b7c17899376fc69a68dfc6b1d28c140
Medium: RA: mysql: removed all bashimisms
All regular expressions are now used with grep
All arrays are
# HG changeset patch
# User Marian Marinov
# Date 1267191536 -7200
# Branch mysql-ms
# Node ID c6bc95f4dfff0d97ec7aaf98c811168c38153ace
# Parent 435f3775df694592d1233eeddfb55ea6ae76199e
Low: RA: mysql-ms: Added more information for the replication_user var
Updated the long desription of
# HG changeset patch
# User Marian Marinov
# Date 1267192583 -7200
# Branch mysql-ms
# Node ID 0d95421fed862e5053e65f369f67f87ae172d5a6
# Parent c6bc95f4dfff0d97ec7aaf98c811168c38153ace
Low: RA: mysql-ms: Moved all common variables out of the BSD check
diff -r c6bc95f4dfff -r 0d95421fed86
; [ "$h" = "$HOSTNAME" ] && return 0
> > done
> > return 1
> > }
>
> The RA also should make sure that the list only contains one value. But
> yes, this should work.
>
Is it actually possible to have more t
Thank you Florian.
I'll try to have the master preference functionality today.
--
Best regards,
Marian Marinov
signature.asc
Description: This is a digitally signed message part.
___
Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org
On Friday 26 February 2010 09:55:40 Florian Haas wrote:
> On 2010-02-25 07:48, Marian Marinov wrote:
> > On Thursday 25 February 2010 04:38:16 Marian Marinov wrote:
> >> Hello,
> >> again thanks for the comments. In this version of the patch I tried to
>
d lists, so the trailing whitespace shouldn't
> actually hurt.
Lars, sorry but it seams my shell scripting skills are not good enough to
solve this one...
How would I match two hostnames:
host_a='shrek'
host_b='shrek '
I don't know how to do that without RE in sh
On Thursday 25 February 2010 04:38:16 Marian Marinov wrote:
> Hello,
> again thanks for the comments. In this version of the patch I tried to fix
> all mentioned problems.
>
> However there are a few things:
>
> 1. the whitespaces after the HOSTNAME should remain since t
ling this?
>
> how about
> local slave_info
> slave_info=$(mysql blafoo)
> # No, not local slave_info=$(), because then $? is the exit code
> # of declaring that variable local, which is 0 (unless your
> # shell crashes half way through...)
>
Hello,
since in my resource agent it is important to know if the resource is
master/slave or other type I added one simple check in the .ocf-shellfuncs
ocf_is_ms. This is the patch.
--
Best regards,
Marian Marinov
diff -r 1a692bb1e5b0 heartbeat/.ocf-shellfuncs.in
--- a/heartbeat/.ocf
On Friday 21 November 2008 11:50:50 Michael Schwartzkopff wrote:
> Hi,
>
> this week there was the announcement about a GPL version of DRBD+, which
> means 3 partitions in one DRBD. How is this integrated into pacemaker?
> Multistate resource running 3 times in the cluster?
>
> Any experiences?
>
>
43 matches
Mail list logo
