[Maria-developers] 2PC between two engines + binlog
Hi Hakan, all With PBXT and the latest merge, this should now technically be enabled (again). That is, a transaction should be consistent between InnoDB+PBXT +binlog, that is either all or none should commit. But as I understand, when this was last tried with Falcon it went kaboom in MySQL, so likely there are some bugs. I think it would be good to at least create/enable some tests for this, then we can see what the exact code status is, and hopefully get it all working as well. Right now I don't think anybody really expects something like this to work, but in theory it should work so presumptions *may* be made (I know this sentence appears kinda contradictory, hope it makes sense anyhow ;-). Thus it's important for us to know what happens, so that we can tell people "don't do this now, please track bug #", or "please try this". Thanks Cheers, Arjen. -- Arjen Lentz, Exec.Director @ Open Query (http://openquery.com) Exceptional Services for MySQL at a fixed budget. Follow our blog at http://openquery.com/blog/ OurDelta: packages for MySQL and MariaDB @ http://ourdelta.org
Hi Hakan, Unfortunately there is a bug in 1.0.09f which prevents PBXT XA from working correctly with the binlog. I have pushed the patch to: lp:~paul-mccullagh/maria/maria-pbxt-1009g We have already tested the patch with MySQL 5.1.41. I will propose the patch for merge as soon as I have finished testing MariaDB. After that, all should work (theoretically), as Arjen says... :) On Dec 21, 2009, at 1:29 PM, Arjen Lentz wrote:
Hi Hakan, all
With PBXT and the latest merge, this should now technically be enabled (again). That is, a transaction should be consistent between InnoDB+PBXT +binlog, that is either all or none should commit.
But as I understand, when this was last tried with Falcon it went kaboom in MySQL, so likely there are some bugs. I think it would be good to at least create/enable some tests for this, then we can see what the exact code status is, and hopefully get it all working as well.
Right now I don't think anybody really expects something like this to work, but in theory it should work so presumptions *may* be made (I know this sentence appears kinda contradictory, hope it makes sense anyhow ;-). Thus it's important for us to know what happens, so that we can tell people "don't do this now, please track bug #", or "please try this".
Thanks
Cheers, Arjen. -- Arjen Lentz, Exec.Director @ Open Query (http://openquery.com) Exceptional Services for MySQL at a fixed budget.
Follow our blog at http://openquery.com/blog/ OurDelta: packages for MySQL and MariaDB @ http://ourdelta.org
_______________________________________________ Mailing list: https://launchpad.net/~maria-developers Post to : maria-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-developers More help : https://help.launchpad.net/ListHelp
-- Paul McCullagh PrimeBase Technologies www.primebase.org www.blobstreaming.org pbxt.blogspot.com
Hi Paul, I will wait for your update. Cheers, Hakan On 21.12.2009, at 14:21, Paul McCullagh wrote:
Hi Hakan,
Unfortunately there is a bug in 1.0.09f which prevents PBXT XA from working correctly with the binlog.
I have pushed the patch to: lp:~paul-mccullagh/maria/maria-pbxt-1009g
We have already tested the patch with MySQL 5.1.41. I will propose the patch for merge as soon as I have finished testing MariaDB.
After that, all should work (theoretically), as Arjen says... :)
On Dec 21, 2009, at 1:29 PM, Arjen Lentz wrote:
Hi Hakan, all
With PBXT and the latest merge, this should now technically be enabled (again). That is, a transaction should be consistent between InnoDB+PBXT +binlog, that is either all or none should commit.
But as I understand, when this was last tried with Falcon it went kaboom in MySQL, so likely there are some bugs. I think it would be good to at least create/enable some tests for this, then we can see what the exact code status is, and hopefully get it all working as well.
Right now I don't think anybody really expects something like this to work, but in theory it should work so presumptions *may* be made (I know this sentence appears kinda contradictory, hope it makes sense anyhow ;-). Thus it's important for us to know what happens, so that we can tell people "don't do this now, please track bug #", or "please try this".
Thanks
Cheers, Arjen. -- Arjen Lentz, Exec.Director @ Open Query (http://openquery.com) Exceptional Services for MySQL at a fixed budget.
Follow our blog at http://openquery.com/blog/ OurDelta: packages for MySQL and MariaDB @ http://ourdelta.org
_______________________________________________ Mailing list: https://launchpad.net/~maria-developers Post to : maria-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-developers More help : https://help.launchpad.net/ListHelp
-- Paul McCullagh PrimeBase Technologies www.primebase.org www.blobstreaming.org pbxt.blogspot.com
-- Hakan Küçükyılmaz, QA/Benchmark Engineer, Stuttgart/Germany Monty Program Ab, http://askmonty.org/ Skype: hakank_ Phone: +49 171 1919839
Hi Hakan, All tests look good, so I have proposed the bug fix mentioned below for merging: lp:~paul-mccullagh/maria/maria-pbxt-1009g On Dec 21, 2009, at 3:09 PM, Hakan Kuecuekyilmaz wrote:
Hi Paul,
I will wait for your update.
Cheers,
Hakan
On 21.12.2009, at 14:21, Paul McCullagh wrote:
Hi Hakan,
Unfortunately there is a bug in 1.0.09f which prevents PBXT XA from working correctly with the binlog.
I have pushed the patch to: lp:~paul-mccullagh/maria/maria-pbxt-1009g
We have already tested the patch with MySQL 5.1.41. I will propose the patch for merge as soon as I have finished testing MariaDB.
After that, all should work (theoretically), as Arjen says... :)
On Dec 21, 2009, at 1:29 PM, Arjen Lentz wrote:
Hi Hakan, all
With PBXT and the latest merge, this should now technically be enabled (again). That is, a transaction should be consistent between InnoDB+PBXT +binlog, that is either all or none should commit.
But as I understand, when this was last tried with Falcon it went kaboom in MySQL, so likely there are some bugs. I think it would be good to at least create/enable some tests for this, then we can see what the exact code status is, and hopefully get it all working as well.
Right now I don't think anybody really expects something like this to work, but in theory it should work so presumptions *may* be made (I know this sentence appears kinda contradictory, hope it makes sense anyhow ;-). Thus it's important for us to know what happens, so that we can tell people "don't do this now, please track bug #", or "please try this".
Thanks
Cheers, Arjen. -- Arjen Lentz, Exec.Director @ Open Query (http://openquery.com) Exceptional Services for MySQL at a fixed budget.
Follow our blog at http://openquery.com/blog/ OurDelta: packages for MySQL and MariaDB @ http://ourdelta.org
_______________________________________________ Mailing list: https://launchpad.net/~maria-developers Post to : maria-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~maria-developers More help : https://help.launchpad.net/ListHelp
-- Paul McCullagh PrimeBase Technologies www.primebase.org www.blobstreaming.org pbxt.blogspot.com
-- Hakan Küçükyılmaz, QA/Benchmark Engineer, Stuttgart/Germany Monty Program Ab, http://askmonty.org/ Skype: hakank_ Phone: +49 171 1919839
-- Paul McCullagh PrimeBase Technologies www.primebase.org www.blobstreaming.org pbxt.blogspot.com
Paul McCullagh <paul.mccullagh@primebase.org> writes:
All tests look good, so I have proposed the bug fix mentioned below for merging:
lp:~paul-mccullagh/maria/maria-pbxt-1009g
Thanks! I have merged this into the MariaDB 5.1.41 release clone (will later also be merged into main lp:maria). - Kristian.
participants (4)
-
Arjen Lentz -
Hakan Kuecuekyilmaz -
Kristian Nielsen -
Paul McCullagh