OK - the new server is up and running, and the wiki is back in its proper place; editing is re-enabled and it should be a damned sight faster now.
It may take an hour or two before your ISPs DNS catches up (the time-to-live on the record is only 1 hour, but some ISPs seem to have DNS servers that don't seem to honour the TTL and take days to update, ho hum...)
It would be useful to have the Variable and ParserFunctions extenstions installed for wiki use; with those in place, I think that the OXPs list could be made quite a lot more manageable (with a template or two).
Incidentally, security support for etch will cease in 11 months and 4 days…
It seems like the hardware is STILL flaky, and it's gone down again. I'm travelling at the moment so it will have to just stay down until I get a chance to transfer it back to the old UltraSPARC while the proper box is sorted out *again*.
I vaguely remember an urban legend where a hospital cleaner would disconnect a bed's artificial lung/kidney/heart from a specific bed when doing her rounds. For months the doctors couldn't figure out why all the patients given that particular bed would keep dying so regularly. I wonder if there's something like that happening inside the data centre....
It's back on the old version (at least the update I did to my page and the OXP pages has rolled back and gone), and judging by the speed on the back-up hardware again.
Either that or something has gone very weird and my work network here is having cache fun...
edit:are still there are indeed gone. I had my page changes cached.
Anyone wanting to make sure, you could try adding '?dummy=1234' to the base address you want to look at. That's usually enough to force any cache to fetch the actual page... using the url above as an example,
usually having a ? is enough to convince any cache that the page you're trying to look at is dynamic, and shouldn't be cached anyway. As they say, YMMV!