Just released new version, 0.6.0 (it should be visible on pgxn soon) of OmniPITR set of tools.
New version has one new feature – parallelism.
This works in omnipitr-archive and omnipitr-backup-* programs, and allows for parallel delivery to remote destinations (multiple -dr switches).
Also – if you're using compresses wal archive and omnipitr-backup-slave reading from it – all the wal files have to be decompressed before making backup – and this decompression can be parallelized too.
All parallelization is controled using -PJ option (–parallel-jobs), so you can add “-PJ 10" to get up to 10 decompressions at the same time or up to 10 deliveries at the same time.
Just released version 0.3.0 of our tool for handling WAL based replication in PostgreSQL – OmniPITR.
Version jump is related to addition of another tool – omnipitr-synch. This tool is used to copy PostgreSQL data dir (including all tablespaces of course) to remote location(s).
While this process is usually simple (call pg_start_backup(), transfer data, call pg_stop_backup()), thanks to the tool it can be wrapped as single call, with standardized logging, and tested logic. It also makes it trivial, and cheap, to setup more than one new slave at a time, without need to read data off master more than once.
Today, I released 2 new versions (new features in 0.5, and then quick bugfix in 0.51) of Pg::Explain Perl library.
This is the base of explain.depesz.com – the code that does heavy lifting of parsing plans, and providing them in uniform way (object).
Continue reading Big changes on explain.depesz.com
OmniPITR project that I wrote about some time ago is going on.
Just today I finished tests for omnipitr-backup-slave – part of OmniPITR which lets you make hot-backups of WAL-slave machine – without any additional load on master.
As previously – please download (svn co) and test. In case you have problems – please mail me or contact me on irc.freenode.net – I'm usually on #postgresql.
Thanks to the company I work for OmniTI I was working on pretty cool project. Name of the project is OmniPITR, and here is what it is, why, how, and where to get it.
Continue reading OmniPITR
I just updated explain.depesz.com with bugfix, which changes the way Bitmap Index Scan and Bitmap Heap Scan are displayed. Apparently index and table names were not shown previously.
Thanks go to Viktor Rosenfeld for spotting and reporting the bug.
I just updated explain.depesz.com with 2 new interface features:
- When you put mouse cursor over node, it will mark direct child nodes with
- When you click on node, all child nodes (even indirect) will be hidden, and the node you clicked on will be marked with
Both features are directed towards people who analyze longer plans – ability to hide parts or just visually see what the Nested Loop is calling in 2nd stage greatly helps me see the flow of query – you can test it, for example, here.
By the way – BIG THANK YOU for Metys for help on it (well, actually, for doing it).
I just modified the internals of explain.depesz.com. Now, it finally stores the plans in database (previously it stored the plans as files in dedicated directory).
Effect for enduser is just that history page should load faster.
But, having the data in database makes it possible to add more features.
One such feature is already added – ability to explain plan, but not list it on history page.
Right now, when you add new plan, you can specify if you want it to be listed with other previous explains – default value is “yes". But if you don't want just anybody to be able to click his way to your plan – there is option for it.
For obvious reasons all previous plans are marked as public now. If you want it to be changed, please contact me – we can probably do something about it.
And as last thing: I would like to express my big THANK YOU to all of you who use explain.depesz.com.
(side note: Catalyst is really cool)
I just updated explain.depesz.com with the newest explain-parsing library version (Pg::Explain v 0.09).
This version will hit CPAN mirrors in next couple of hours.
- Fix exclusive time calculations
- Make PE::Node understand Bitmap scans (heap and index)
- Add proper handling of nodes that were “never executed"
- Add ->is_analyzed method to PE::Node to make it easy to distinguish between EXPLAIN and EXPLAIN ANALYZE nodes
- Apply patch from Filip Rembiałkowski, which adds first header and last header display – for example to see time range the analysis was on.
SVN repo at: http://svn.depesz.com/svn/pgsql-tools/trunk