- developers - lists.mariadb.org

Re: [Maria-developers] [Commits] Rev 4376: MDEV-6676: Speculative parallel replication in http://bazaar.launchpad.net/~maria-captains/maria/10.0
by Jonas Oreland 22 Sep '14

22 Sep '14

> > Intermediate commit. Patch is far from complete, but this small patch was > nevertheless sufficient to be able to sysbench-0.4 OLTP with full > parallelisation. > "full parallelisation" does that mean that X threads on master make slave achieve k*X higher throughput ? /Jonas

4 16

[Maria-developers] Need suggestions for benchmarking MDEV-6676 parallel replication
by Kristian Nielsen 22 Sep '14

22 Sep '14

Hi Axel, I'd like to ask your advice on some benchmarking of MDEV-6676. MDEV-6676 is about extending in-order parallel replication to be significantly more aggressive, and potentially get a _lot_ more parallelism on the slave. The basic idea is to just try to run _everything_ in parallel. If there is a conflict, the code will roll back the erroneous transaction and retry it, this time not in parallel. I have run a basic sysbench 0.5 OLTP on my laptop, and results were promising. However, I need something better than that. My main problem is that often when I run write-load benchmarks, the bottleneck is InnoDB log write or purge, or something like that. So the load does not really scale with multiple threads, even on the master, and parallel replication is not really that useful even if perfectly implemented. So do you have a suggestion for a write-load benchmark that scales well for medium concurrency, like 8-32 threads or something like that? And for a machine on which to run it? The best benchmark I think would be one that is I/O bound on a working set that is not in memory, on a machine with a disk subsystem that has good throughput at high concurrency. Because this should resemble the real-world use cases where parallel replication is most needed. Do you have any experience on running such benchmarks, and do we have any machines that could be used? Often the benchmarks I see in blogs are in-memory... Or do you have another suggestion? Thanks, - Kristian.

1 0

[Maria-developers] MDEV-4682 revision
by Roberto Spadim 20 Sep '14

20 Sep '14

hi guys, i sent some patch about qc_info plugin ( https://mariadb.atlassian.net/browse/MDEV-4682) via pull request at github (https://github.com/MariaDB/server/pull/6) could anyone check if it's ok? i included test file changes there's only one thing todo, that's include/not include statistics inside query cache, since it increase memory use, i think could be nice at runtime change this variable and include statistics on/off (more info inside mdev) -- Roberto Spadim

1 0

[Maria-developers] doubt about tests files
by Roberto Spadim 20 Sep '14

20 Sep '14

hi guys! i git cloned the last verison of mariadb from github, but i didn't found test files is it outside mariadb github source? -- Roberto Spadim

1 1

Re: [Maria-developers] [Commits] c8496e4: %B
by Roberto Spadim 19 Sep '14

19 Sep '14

working =P 2014-09-19 18:04 GMT-03:00 Igor Babaev <igor(a)askmonty.org>: > revision-id: c8496e48c4b7c35e9702f98daa19cdb4ef190097 > parent(s): 0486b3b796ad7ec597bff775da248096731c58d5 > committer: Igor Babaev > branch nick: 10.1-test-commit > timestamp: 2014-09-19 13:47:43 -0700 > message: > > %B > --- > sql/opt_range.cc | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/sql/opt_range.cc b/sql/opt_range.cc > index 27e20c0..c35bde6 100644 > --- a/sql/opt_range.cc > +++ b/sql/opt_range.cc > @@ -14,6 +14,8 @@ > along with this program; if not, write to the Free Software > Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 > USA */ > // test commit in git > +test 2 > + > /* > TODO: > Fix that MAYBE_KEY are stored in the tree so that we can detect use > _______________________________________________ > commits mailing list > commits(a)mariadb.org > https://lists.askmonty.org/cgi-bin/mailman/listinfo/commits > -- Roberto Spadim SPAEmpresarial Eng. Automação e Controle

1 0

[Maria-developers] Free Software Foundation address
by Matej Mužila 17 Sep '14

17 Sep '14

Hi, the address of the Free Software Foundation has changed and I think itshould be changed in sources too. Nowdays the correct one is: Free Software Foundation 51 Franklin Street, Fifth Floor Boston, MA 02110-1301 USA http://www.fsf.org/about/contact/ Regards Matej Mužila

1 0

[Maria-developers] master.info and relay-log.info location
by Tim Callaghan 15 Sep '14

15 Sep '14

Has any work been done in MariaDB 10.0 or 10.1 to move them from their file-based location to within XtraDB (and thus potentially TokuDB)? The knowledge base article at https://mariadb.com/kb/en/mariadb/development/general-development-informati… lists the following as a roadmap feature: - Transactional storage of slave state, rather than file-based master.info and relay-log.info . So the slave can recover consistently after a crash.

2 1

[Maria-developers] contribution procedure
by Jonas Oreland 12 Sep '14

12 Sep '14

Hi Serg, hope all is well. I created https://mariadb.atlassian.net/browse/MDEV-6712 for a plugin feature I wrote, and now I humbly ponder on how I get this review/triaged/something. What would be your best advice? /Jonas

2 1

[Maria-developers] How to get all fields in a given index?
by Rongrong 10 Sep '14

10 Sep '14

Hi, I'm trying to implement some feature that requires to get the list of fields for a given index. I'm trying to follow the optimizer code to find some clue but am not successful so far. Can anyone give me some pointers? Thanks! Rongrong

2 2

[Maria-developers] Total breakdown of development procedures :-(
by Kristian Nielsen 10 Sep '14

10 Sep '14

Today, I learned that a large change (the galera merge) has been pushed into 10.1 on August 27. At that point, there was not a _single_ builder in buildbot that was green! As a result, we now have an extremely serious regression in the server caused by this push, easily seen in the testsuite. This is a disaster! Someone (or more likely everyone) have forgotten the basic rules, rules that have been in place since the beginning of MariaDB, and which are essential to ensuring the necessary quality of a product used in production-critical environments every day: - No pushes into a tree with known test failures in Buildbot. - No pushes of larger changes without first seeing them green in Buildbot in a feature tree. - No pushes without review. - No pushes of new features that are not of beta quality. Must be feature complete, must have passed tests and review, must have documentation ready. I have asked Jan to revert the Galera merge until it can be tested green in Buildbot and passed RQG replication tests by Elena. But the problem here is clearly larger than any single patch. Things have degenerated far beyond what is even remotely acceptable. We need to step back and start following the rules to the letter to recover from this. - Kristian.

1 0