Urgh...I've been fortunate enough to avoid things like this myself
until recently, when some of my PHP/MySQL/Apache stuff is getting
tossed out in favor of Cold Fusion/MS-SQL Server/MS-IIS. Anyway, my
At 8:35 AM -0500 8/16/01, Boget, Chris wrote:
>Recently, we presented MySQL as a database option for a website that
>we might be working on. We've used it as our database in the past and
>we plan on using it in the future as possible.
>With that said, I confess I don't have as intimate a knowledge of mySQL
>to address some of the things in the email that was sent to me. I'd like
>to hear what some of you have to say/think about this. I know some
>of the things said below aren't entirely correct, but I'm not 100% sure
>about some of the others.
>MySQL - as I said at our meeting, we would not be comfortable with this
>as an enterprise strength solution. MySQL is unsupported freeware and
>lacks enterprise management functionality.
Do these people even know what they mean by 'enterprise management
functionality', or did they just crib from an Oracle brochure? DO
they have any specifics about what they require?
>It has a small limited feature
>set compared to ORACLE, DB/2 and is lacking the functionality to support
>data replication and has little capability for generating management info.
No surprise that these folks haven't been following MySQL development
for quite a while, and probably don't know about its replication
features. I haven't used 'em myself, though, so I can't vouch for
As far as the feature set & manageability, it's true - there's a
lotta things MySQL made a conscious decision to leave out (unions,
views, triggers, stored procedures, subselects [i know, coming soon],
foreign key support, etc.) in favor of speed/small memory footprint.
And you have to go to third-parties for
If your application requires such, then maybe MySQL _isn't_ the right
solution; however - depending on your app - Oracle/DB2/whatever might
be sheer overkill. Administrative overhead for systems like those
might far outweigh any advantages they have for you.
>There are question marks around the scalability of the product, I'm not
>sure of the locking algorithms used (whether row level or record level) -
It depends on table type; AFAIK, it can be table (ISAM/MyISAM),
page-level (BDB), or row-level (InnoDB). See:
You've got a choice! This used to be considered a good thing...
>fact that it is not generally used in multi-user solutions is a good enough
>indication that this is not accepted database technology for
>The fact that it is unsupported freeware would mean that an end user would
>potentially be "held to ransom" by a DBA with specific knowledge.
This kinda of statement is beginning to REALLY rile me when I hear
it. Even if you discount the fact that this mailing list provides
better support than the majority of PAID support programs, if you
want to, the MySQL folks would be more than happy to take a large
amount of your $$$ to provide excellent support:
- this can include customizing MySQL for you! There are also
individual consultants & firms that will support you as well. How
anyone could actually back up a claim of MySQL being 'unsupported' is
>security model is also not sufficiently developed for any system that
I dunno, with some combination of encrypted fields, database server
behind a firewall, SSH-tunnelled communication and good DB/system
administration, you'd have a plenty secure system. After all, I don't
think any of the recent and not-so-recent credit-card number thefts
have been on MySQL systems.
OK, back to work for me. But first, some Mountain Dew...
+------------------------ Open source questions? ------------------------+
| Steve Edberg University of California, Davis |
| sbedberg@stripped Computer Consultant |
| http://aesric.ucdavis.edu/ http://pgfsun.ucdavis.edu/ |
+----------- http://pgfsun.ucdavis.edu/open-source-tools.html -----------+