Squid and IPv6

As it turns out if you want to have IPv6 support in the popular web proxy squid then you need to be using the cutting edge code available via CVS. This of course means that you are trying to get working code that others are still working on. Not the best situation but I'm hoping that I can get around anything that is still broken.

So far I have found a dependance on uudecode (for the icons). Of course uudecode is no longer distributed with Fedora. You only get a program called uudeview which I can only assume has the same calling semantics as uudecode. Here's hoping because currently the icons extraction doesn't work and therefore the make install doesn't working. So I am now manually trawling through the code doing manual makes where ever I think they are needed to get around this.

Getting closer.

IPv6 Proxy

Finally squid has been updated to work with IPv6. I am now in the process of testing this out. This is the ultimate solution to getting IPv6 working within your network. The proxy can work out the correct type of connection required and thus provide transparent access to either IPv4 or IPv6 hosts that it services.

Currently, anytime I wish to access an IPv6 site I have to disable my proxy because it doesn't understand. This is a pain and deflates my experience of IPv6. Now that squid supports IPv6 I will be able to leave my proxy on and leave the choice of protocol to the proxy depending on what type of address records it gets back from DNS.

So here's hoping I can get the beta of squid 3.x working with IPv6 support.

Grrr, ie Sucks!

Can someone who actually develops and uses IE fix it!!

When I do a shortcut to load a URL I would really just like to type it straight into the location bar. Why the need for a separate box? And when I inevitably make a mistake in the typing of said URL please don't make me reach for the mouse to click to correct it. Why can't I just hit my shortcut again and fix the few characters that I mistyped?

This is something where firefox has got it right. Keeping it simple. No wonder it is stealing market share from IE.

Anyway that's my gripe for the day.

Grrr, Debuggin SOAP and PHP

I don't know about you but something shouldn't be this hard. I'm currently trying to get my final SOAP code working and for the life of me I can't figure it out. The PHP module reports useless errors like "Whoops looks like we got not XML" or words to that effect.

Which means diddily squat to me. Do you know what that means? If you do can you let me know? All my other SOAP interactions work fine but the server side is running on a linux machines with the lastest updates for the PHP and SOAP code.

This last one is running on a hand compiled apache install running on solaris. I'm sure there are many areas of danger there.

So I'm stuck with this. I guess I should download and compile all the latest patches and see where that ends up.

The frustrating thing about all this is the SoapClient can understand my WSDL file present the list of functions available but cannot for reasons unknown call any of them!

I'm sure I will eventually find a solution but seriously, get some useful error messages guys!