I’ve hacked around a bit to get PBXT to compile in tree, and pulled in the Maria engine. Both are latest source.
So, want to try out Maria?
Want to try out PBXT?
Just want to do ./configure and go with it, just like building a normal MySQL Server?
Grab the -stew tree. Source tarball here:
(it’s based on something close to 5.1.24… and I’m going to switch some of my systems over to it rather soon… already done some good benchmarks on one of my apps).
feedback much appreciated.
UPDATE: Got x86-64 Linux. Try my binary tarball (built from above src tarball). Built on Ubuntu Gutsy (my laptop). So it may (or may not) work. If it kills your squirrel, not my fault.
Stewart,
Having a small issue building and running the mysql-test-run.pl comes back with:
mysql-test-run: *** ERROR: Could not find any of /tmp/stew/mysql-5.1.24-stew1/storage/maria/release/maria_chk
During the make process, stdout / stderr never reports anything to do with maria (make | grep maria – basically shows nothing).
Basically:
./configure
make
cd mysql-test
./mysql-test-run.pl [tried a number of options such as –skip-test=maria]
Am I missing something? I wouldn’t be surprised if it was user error, however, can’t seem to get it right.
are you building on windows? the ‘release’ is something that Visual Studio creates…. I do have to admit i only tested on Linux… and only on my box.
but i’d love to have it solved…. does the maria tree work for you?
try
./BUILD/compile-mad64-max
and then:
make bin-dist
and then try running in that tarball
(if you’re linux)
./BUILD/compile-mad64-max didn’t exist (and running x86)
but
./BUILD/compile-pentium-max worked a treat !
For some feedback: http://paste-bin.com/13247
Lost query to MySQL Server – I’m not much of a mysql qa tester, but I know that error message when our cluster decides it’s time that time of the month. Thoughts ?
Also, the mysql-5.1.23a-maria-alpha compiled fine for me using ./configure && make, so not sure what was going on, since then I had install the complete Development Tools yum group – so could have simply been that difference.
Got some results from the mysql built sql-bench/test-transactions script:
http://spreadsheets.google.com/pub?key=p-hhLx9MuqgBnJQ99ySwsBw
For total average operations all pretty on par, which, personally I find a little surprising, as InnoDB has had much more time to perfect performance – as I’d expect Maria and PBXT to have concentrated closer on stability and new features.
Also, take the results with a large grain of salt, as these are far from perfect benchmarks.
hrrm… I saw something like this too…
I’ll investigate… good to have something to fix in the next release… -stew2 (which will be made easier when the maria tree is updated to a slightly newer base (currently it’s somewhere between 5.1.23 and 5.1.24… and the way i brought it into my tree made mine 5.1.24-ish… (well… somewhere between 23 and 24)).
Of course.. it could be a Maria bug… I’ll investigate.
my guess is it’s only a problem when selecting from INFORMATION_SCHEMA.TRIGGERS after creating tables with triggers with russian names.
(in future, i’ll actually test better before publishing too)
Yeah, in regards to the INFORMATION_SCHEMA.TRIGGERS thoughts I’d say your right cause a future test (main.ddl_i18n_koi8r) also fails when selecting from the information_schema.triggers table. Not sure if it’s a Maria bug though, as my steps to reproduce didn’t specify the engine (my default was MyISAM), but that’s just my thoughts – don’t work much with the internals as I’d like.