I’ve been meaning to upgrade the SOTA cluster’s infrastructure for a while now and finally got round to doing this last Saturday. The cluster is written in the Python programming language and was written for use with Python2. Whilst Python2 is a fine language, it has reached the end of development and only now gets important security fixes. The future is Python3 which is essentially the same but complete different!
The upgrade took about 30mins to do and I introduced one not so subtle bug that was spotted by Pat KI4SVM which was fixed also on Saturday. It’s been running since then and my inbox has remained unusually quiet so I’m taking that as a good sign.
Users should see no changes at all apart from the version number at sign on having changed. But if you do find a bug left me know.
You can find the cluster at elgur.dtdns.net 7300 and it should work with most cluster aware software apart from HRD. Or you can just log in directly using your callsign as username. You don’t need a password as it’s only a source of SOTA spots derived from SOTAwatch.
Nice. I’m a long-time Python fan. My first production Python code was written in version 1.3 almost 20 years ago.
I think I first used Python when it was at the 1.6 or 2.1 stage back in 2003, it was used to provide a complicated build system where I was working. I moved on and didn’t see it used anywhere till an article about it on a website piqued my interest again. I was already running an SMS spotter that was written in C++ running on a PowerPC 604 machine. I downloaded and built Python2.7 for that and then learned ‘enough’ Python by converting the C++ program to Python in about 5 evenings. I swapped the spotter over and nobody noticed so I decided at that point I was an expert!
I should have gone straight to Python3 then but played about using Python2.7 for everything, often because something else was only Python2. I took the plunge and updated the SMS spotter to Python3 when I moved it to the VM it runs on in Iceland. There were a few bits to comprehend, the fact that strings are not byte sequences as in C. But once you get your mind around the fact you need to encode a string to UTF-8 (or ASCII) on output and decode it back to a string on input then it all becomes straightforward.