Server outage updates…

On July 31st, the GeoIngenuity server crashed. Hard.

That’s exactly 1 day after the release of GeoJournal. Just in case you thought otherwise, I can tell you first hand that Mr. Murphy is still quite alive and well, and keeps the most impeccable timing…

Luckily, I use VMWare to virtualize the server, which makes it handy if the underlying server hardware goes kaput. You just upload the server image to a new VMWare server, and bring it up just like nothing happened…

Well, I did have a backup, but what I didn’t account for was the pathetically slow upload speed of my Comcast cable modem. They tout 384k upload speeds, but in reality it was more like 80k. Wonderful. It took about 3 days, on and off, to upload the image to the new temporary server. Hopefully, I will have new server hardware and I can transfer the image back to the new hardware. I wasn’t expecting the new hardware expense, but Murphy doesn’t usually let you know when he’s coming for a visit, so it’s just one of those things…

Why is this important to you, the GeoJournal user? Well, simply put, the Google Map feature of GeoJournal must communicate with this server initially before it accesses the map data from Google. This is a requirement based on Google’s terms of service and program API. When my server goes offline, the maps will not work. The good news is with DNS changes I can redirect the map queries to another server, and that’s what I did when the server died. I had map calls redirected within a couple hours, while the hardware was down for days (and is still down, actually).

I’ll post a new announcement when the new hardware is going in. I promise it wont take days to upload this time…