Wednesday, November 12, 2008
LISA 2008 San Diego
It's been a while since I last posted to this dev blog..not a lot of activity in the last year. As LISA 08 has approached, I've had a renewed inspiration in MachDB and am excited to be working on it again. I'm at the LISA conference Wed, Thurs & Friday if anyone would like to meet up and discuss the future of the MachDB, features you'd like to see or gripes/complaints. Hit me up on the blog and we'll find a way to meet up!
Tuesday, April 22, 2008
Web 2.0 Expo in SF
I'll be at the Web 2.0 Expo in San Fran this week, Wed-Friday. Hit me up if you're there and want to talk MachDB or just get a beer!
I've been busy philosophizing with others about the future direction of the MachDB. I have some code to write and some cool new features coming soon. I'll post more after the conference is over.
I've been busy philosophizing with others about the future direction of the MachDB. I have some code to write and some cool new features coming soon. I'll post more after the conference is over.
Wednesday, December 26, 2007
Freshmeat and Sourceforge
We now have some more project features, courtesy of the nice people at freshmeat.net and SourceForge.
http://freshmeat.net/projects/machdb/
and
https://sourceforge.net/projects/machdb/
I'm not using a lot of the features at SourceForge, but the bug and feature tracking systems will come in handy...if I can get people to use it. As for Freshmeat, I'll use that to get info out to the community.
http://freshmeat.net/projects/machdb/
and
https://sourceforge.net/projects/machdb/
I'm not using a lot of the features at SourceForge, but the bug and feature tracking systems will come in handy...if I can get people to use it. As for Freshmeat, I'll use that to get info out to the community.
Saturday, December 22, 2007
Documentation: Data Flow Diagram
In the interest of education, I have been working on docs. I started off with a data flow diagram, available in PDF and JPG form.


Wednesday, December 19, 2007
machdb-0.5.1 released
I put out a new version tonight of the code and a slightly updated XML spec. The following bugs and issues have been addressed:
I also added some contributed gathering script files. We now have them for PPC Linux, Sparc Solaris and Debian i386. Thanks to Chris and Kyrre for those contributions!
- Filesystem and disks have had the limit raised from 2TB to 16 zettabytes.
- OS Name limit now 100 characters
- Added Base Architecuture to host node
- Added Unified Cache to CPU node
- Split up 'version' and 'release' into 2 different fields for packages. 'release' is optional input
- Various UI bug fixes
- Renamed files, added contrib directory, in an attempt to standardize
mysql -u mysqluser -p mysqldatabse < update_0.5.0-0.5.1.sqlTo update the data in your tables and your schema at the same time:
php update_0.5.0-0.5.1.phpThis will update your packages table to include release information. If you don't want this, comment it out. If you do, you'll want to check your gathering scripts to make sure they also sent the release information properly.
I also added some contributed gathering script files. We now have them for PPC Linux, Sparc Solaris and Debian i386. Thanks to Chris and Kyrre for those contributions!
Friday, December 14, 2007
Bugs, Specs and new Sites
Moved to a new website, machdb.org Please update your bookmarks. I hope the URL change will make it into the Feb issue of LOGIN. Either way, I'll set up redirects at the old site to get people here. I also moved the blog, to be more central. When I get some time I'll make a new wiki on this domain as well.
Bugs!
Fixed an XML parsing bug in collect.php, thanks to the quick eyes of Josh, Steve and Stephen.
Found a big bug in my spec thanks to Chris. From my response:
"So I found out why your disk data showing a max of 2TB. See...this is where seeing it run in other environments teaches me the right way to do things. ;) Apparently I had some bad math when I calculated what MySQL's int(11) datatype could do. So I guess I've either got to modify the MySQL tables to be BigInt or change the spec to start with MB. With BigInt and KB, the limit is 9,223,372,036,854,775,807 KB, or 8,192 exabytes. Yep, I said exabytes. But that makes the DB bigger. With int(11) and MB, the limit is 2,048 terabytes. Thoughts? I'll noodle on this tomorrow and run it by a DBA buddy and get his feedback as well. Until then, sit tight. I will have this resolved in the next day or three."
I updated the XML spec just now to describe the limitation, so as not to confuse others.
Chris has also sent me a patch to include a unified cache in the CPU information, which will go into the next release next week. I'd like to get this out before Xmas, since I won't be around for a bit afterwards.
These bug reports are great everybody, keep em coming. :)
Bugs!
Fixed an XML parsing bug in collect.php, thanks to the quick eyes of Josh, Steve and Stephen.
Found a big bug in my spec thanks to Chris. From my response:
"So I found out why your disk data showing a max of 2TB. See...this is where seeing it run in other environments teaches me the right way to do things. ;) Apparently I had some bad math when I calculated what MySQL's int(11) datatype could do. So I guess I've either got to modify the MySQL tables to be BigInt or change the spec to start with MB. With BigInt and KB, the limit is 9,223,372,036,854,775,807 KB, or 8,192 exabytes. Yep, I said exabytes. But that makes the DB bigger. With int(11) and MB, the limit is 2,048 terabytes. Thoughts? I'll noodle on this tomorrow and run it by a DBA buddy and get his feedback as well. Until then, sit tight. I will have this resolved in the next day or three."
I updated the XML spec just now to describe the limitation, so as not to confuse others.
Chris has also sent me a patch to include a unified cache in the CPU information, which will go into the next release next week. I'd like to get this out before Xmas, since I won't be around for a bit afterwards.
These bug reports are great everybody, keep em coming. :)
Monday, December 10, 2007
XML Spec Published & Wiki
After the release last night, I sent the code to a fellow sysadmin from LISA who had e-mailed me with interest in the MachDB. He's already got it running, and with not too much effort. My first user, cool! I'm very excited to see it working somewhere else other than my 2 environments. To that end, I worked diligently tonight to document the XML spec that describes what info you can send to the MachDB. You can find the doc linked off the main project website, or you can click here.
Comments on this are welcome, so let me know what you think.
Also, I set up a temporary wiki page at http://www.mybrainhurts.com/wiki/MachDB. You can go there for information from other people that have installed it. I will be keeping a FAQ and install instructions there. Once the permanent wiki is up, all docs will go there.
And if anybody knows of an easy way to syntax highlight code on a web page, you must tell me. ;)
Comments on this are welcome, so let me know what you think.
Also, I set up a temporary wiki page at http://www.mybrainhurts.com/wiki/MachDB. You can go there for information from other people that have installed it. I will be keeping a FAQ and install instructions there. Once the permanent wiki is up, all docs will go there.
And if anybody knows of an easy way to syntax highlight code on a web page, you must tell me. ;)
Labels: machdb
Subscribe to Posts [Atom]