About Me
Michael Zucchi
B.E. (Comp. Sys. Eng.)
also known as Zed
to his mates & enemies!
< notzed at gmail >
< fosstodon.org/@notzed >
Michael Zucchi
B.E. (Comp. Sys. Eng.)
also known as Zed
to his mates & enemies!
< notzed at gmail >
< fosstodon.org/@notzed >
Welcome to my own little corner of the web - one not paid for by advertising, and where neither you the reader nor I the author are the product.
For a little about me see the first blog entry on this site: A Hacker's Introduction originally written a decade ago on another service.
Use the menu above to navigate the information & entertainment(?) available here.
!Z
I think I've managed to convert everything over to zedzone.au - zedzone.space should still work for another week but then will go offline.
The bloody .space domain has suddenly got prohibitively expensive so i'm going to move to another one, probably zedzone.au. The certificate and some other shit might get messy for a while because I do this so rarely I've forgotten how to set it up but I guess we'll see.
Renewed VPS for another 2 years.
Renewed the VPS for another 13 months and took the opportunity to clean out and ugprade the server. It was using Ubuntu 16.04 that was well beyond end of life although I attempted to keep some important software up to date like apache.
I decided to change to debian, the ubuntu LTS end of life process rubbed me the wrong way a bit and debian is probably better for a server anyway. I think I've managed to translate all the old settings and features but i'll keep an eye out for any breakage.
httpd was getting a bit out of date and crashing out now and then so I've upgraded it from source. ubuntu's setup for the apache config is such a fucking mess I should really clean it up. I should also really update the distribution but i'm abit loathe to do that and have to set it all up again ... but I suppose if i do there will be some downtime.
Renewed the VPS for another 13 months.
This thing gets a lot of network traffic but very few hits - search engines to their pass is about all. Plenty of hacking attempts on the site. Occasionaly hacking attempts on ssh. Most of the traffic seems to just be IPV6 control requests though.
Added some new software projects and respositories, firefox extensions and vulkan experiments. I'm thinking of recovering some of the old google code stuff as well.
Fixed a couple of links in the software project pages and converted all the 'this page' links to use https.
And finally I changed the default background colour to a brighter grey and tweaked the blog layout for narrow displays (aka mobile phones).
I've added the ability to clone the git repositories on https://code.zedzone.au/.
These are of course read-only.
I probably would've done it earlier if i realised it was so simple.
I've been loosely monitoring the google search console. Submitting a sitemap has made a difference, it's already approximately doubled the number of pages indexed.
It's also shown a few broken places, some of which I knew about but never fixed such as the transfer of the software/ pages from my old site for the software I no longer bother to maintain. One point of interest is that it uses users.on.net as the canonical version of most of the pages in software/ so I just went throuhg the pages there and added meta redirects. I've retained the contents that they had but but i'm likely to remove it in the future.
I added a sitemap.xml to see if it makes any difference to search engines.
Since moving to my own site it's effectively vanished from the internet. Searches for jjmpeg for example only find an ancient fork someone made on github (as well as a different JJmpeg which was a port of ffmpeg to java - which was already a dead project 10 years ago).
It prompted me to got to the google (and bing) search 'console' and work out what was going on. In a word: not much. Mostly it's just ranked too low because it's not corporate backed I guess, but I also lost all the incoming links I had from blogspot - there weren't many but there were a couple.
I'll see if it makes any difference.
Changed to use a git browser for code.zedzone.au. I'm still setting it up so things might change for a while.
Cleaned up the software page a bit.
I also removed some of the Code links as that is out of date and unmaintained for the moment. Still thinking about a solution for that and I might disable it at some point.
A couple of days ago I tweaked the css and html to improve rendering on (some) phones. It's kinda shit but better sometimes.
I noticed the crawlers were referncing broken url's and found any internal-post links were broken, so i've fixed those. I think when I originally imported them I hadn't completely settled on the site layout rules like /post for posts, and so on.
Added Last-Modified headers to the blog generator. There's some timezone-related skew, I haven't implemented the time related If- headers, and it only goes by the post date itself, but maybe it's better than nothing.
Renewed the VPS for another 13 months. Whee!
I've been occupied with other concerns so there's not been any further development on the site yet.
Bit of a nasty one, I had a dumb bug in my fast-cgi parameter handling that broke with some requests. Just happened to affect mobile browsers more than the ones I use.
Should be fixed now - the blog should no longer give an internal server error for such browsers.
Another
discovery, Let's
Encrypt. So most all of the site is
now accessible via https. For the moment
code.zedzone.au is not.
Add a site icon. Workbench 2.0 palette.
Oh, I discovered the ETag header. As a quick-and-dirty I just update the etag whenever I recompile the blog driver and post it for every response. Since some of the information on every page is dynamic (like the tag counts) this is probably about as good as I can do anyway.
Made a small alteration to the tag url. Was "tag[/postid]?tag=tagname" is now "/tag/tagname[?post=postid]". Also tighter tolerances to bad input.
The blog driver now runs using FastCGI. It's a bespoke incomplete implementation but seems ok so far.
Manually fixed most of the old internode (users.on.net) links in various blog entries to point here. A couple of things I still need to move over though.
Implemented post tags and related indices. These don't match the blogger url's but they provide the same functionality.
Also started work on a system-wide stylesheet. This is now used by the blog system but only a few project pages use it so far.
There has been zero invalid login attempts since the hardening changes a couple of days ago and probe attempts are also down markedly.
Made some config hardening changes to stop invalid login attempts.
Converted the blogger blog over and wrote a basic (read-only) driver. See the Blog link above.
It's still work in progress and i'm working on the live site so things could break or change when i'm working on it. I do have the few comments that have been made and will probably get them added at some point.
Part of the conversion fixed (some of) the html, renamed and downloaded the image files and so on. I haven't done a whole lot of checking but any broken links are probably due to google code vanishing.
Initial setup, added software projects and cvstrac.
To access the cvstrac repositories see "Browse Source" at the top of any software pages where it is available.
Server went live, nothing of import going on