Comments on: mydumper

By: Tarto

Tarto — Fri, 08 May 2009 08:19:30 +0000

I removed PCRE_NEWLINE_ANY and it worked perfectly.
Really fast and easy to use!
I need to run more tests but it looks very promising.
Thank you for such a nice tool.

By: Domas Mituzas

Domas Mituzas — Thu, 07 May 2009 16:18:33 +0000

my fault – it has been fixed in trunk though, I was depending on too new PCRE version.

just removing PCRE_NEWLINE_ANY from the file helps ;-)

I’ll release 0.1.6 with that fix, I guess, was a bit slacking, and expecting more changes to accumulate.

By: ddddddddddddd

ddddddddddddd — Thu, 07 May 2009 13:09:45 +0000

[root@localhost mydumper-0.1.5]# yum install glib2-devel mysql-devel zlib-devel pcre-devel
Setting up Install Process
Parsing package install arguments
Package glib2-devel – 2.12.3-4.el5_3.1.i386 is already installed.
Package mysql-devel – 5.0.45-7.el5.i386 is already installed.
Package zlib-devel – 1.2.3-3.i386 is already installed.
Package pcre-devel – 6.6-2.el5_1.7.i386 is already installed.
Nothing to do
[root@localhost mydumper-0.1.5]# make
cc `mysql_config –cflags` `pkg-config –cflags glib-2.0 gthread-2.0` `pcre-config –cflags` -O3 -g -c -o mydumper.o mydumper.c
mydumper.c: In function ‘check_regex’:
mydumper.c:112: error: ‘PCRE_NEWLINE_ANY’ undeclared (first use in this function)
mydumper.c:112: error: (Each undeclared identifier is reported only once
mydumper.c:112: error: for each function it appears in.)
make: *** [mydumper.o] Error 1
[root@localhost mydumper-0.1.5]#

Thank you for “nice” software.

By: Domas Mituzas

Domas Mituzas — Sun, 19 Apr 2009 16:47:56 +0000

Baron,

I know about mk-parallel-dump options and about the –tab :). Actually in my tests gzip doesn’t matter (client has enough of CPU to facilitate that ;-) – and I don’t want to use tab formats (the whole idea was having the logical sql dump).

This project showcases that there are ways to do things better ;-)

By: Xaprb

Xaprb — Sun, 19 Apr 2009 16:36:19 +0000

What options did you use for mk-parallel-dump? If you give it –tab, it does NOT use mysqldump, it uses SELECT INTO OUTFILE. This might end up being faster than mydumper. Of course it only works if you run it on the same machine you’re dumping. Oh, and by default it’s gzipping the result of the dump, so if you disable that, it should be faster too.

Not that I disagree with you: Perl ain’t the solution here…

By: Domas Mituzas

Domas Mituzas — Tue, 24 Mar 2009 07:50:51 +0000

roy, mylvmbackup is neither logical, portable nor parallel ! ;-)

By: roy f

roy f — Tue, 24 Mar 2009 06:54:11 +0000

why not mylvmbackup ?

By: Log Buffer #134: A Carnival of the Vanities for DBAs

Log Buffer #134: A Carnival of the Vanities for DBAs — Fri, 06 Feb 2009 17:39:19 +0000

[...] might use Domas Mituzas’s just-released mydumper, of which Domas says: “I had that idea that there’s no good tool to do logical dump of [...]

By: mike

mike — Wed, 04 Feb 2009 03:48:40 +0000

This is an awesome tool for me, actually. I have been backing up MyISAM/InnoDB databases nightly for a client and once a week lately it seems the job gets stuck or takes so long it bottlenecks some cronjobs which never finish as well, and then run the next instance over and over until the box has to be forcibly rebooted.

This might help cut down on the time needed to do the dump!

I should probably also put a lockfile in, now that I think about it, so cronjobs exit without doing anything during the nightly backup. I don’t have extra resources for some sort of replication based backup, LVM backups would be proprietary and not plain SQL … I was looking for an alternative but to get a plain SQL dump it seems like this might wind up being the winner!

By: Domas Mituzas

Domas Mituzas — Tue, 03 Feb 2009 21:33:37 +0000

mysqlhotcopy would not work with InnoDB datastores (nor MyISAM, it locks stuff too much), innodb hot backup is proprietary, and neither perform logical backups.

By: roy f

roy f — Tue, 03 Feb 2009 16:59:06 +0000

Why not use mysqlhotcopy or innodb hot backup ?