I'm currently testing a pair of POC radios

<editor: can no one shut him up about the radios?>

No, go away.

So, where was I, oh yeah, POC radios. What are they, I hear you asking.

<editor: no one asked that>

Hush.

Push-to-talk Over Cellular. POC or PoC. In short, you get a SIM card with data. That goes in your radio. The radio transmits digitally to local cel towers. Ta-da. You have a "radio" that can reach around the world and back. These things are sold as "rely on POC radios for emergencies" and "stay in touch with the family during an emergency." Stuff like that.

Let's talk emergency for a moment. Last year, and this year (sadly, for both), there were two big weather events in the US: the hurricane and subsequent floods around the Carolinas, the floods in Texas. There have been a number of other, much smaller, more localized, events as well, just here in the US (lots overseas too, put in your own examples, power outage in Spain is just one). Want to know one of the things that failed at both these events (and Spain)? Cel service. But, why do so many people suddenly want radios? So they can communicate with loved ones in an emergency... when the cel service goes out... and you're using a cel serviced based radio as your emergency backup communication. Anyone else see the problem here?

Same with databases.

<editor: finally>

Hush I said.

If nothing has ever happened to the databases under your care, great! I'm jealous. Me, I've seen a ton of issues. I've researched even more, both to help better prepare myself, and to help teach preparedness. However, a whole lot of us have experienced data outages. From simple things like someone deleted a row they didn't mean to, to dropped databases, to actual outages caused by natural catastrophes (or unnatural ones, I still remember the data center that was taken out when a truck drove through the wall by accident).

What's more, I've seen, and read about, people's planning for the emergency go horribly awry. Myself, we had a database die, corruption. Went to restore it only to find that backups had been failing, without an error, for about four months. We didn't have monitoring in place to look at backup age. Oops. There's the case where people said, "Oh, we don't need a backup, we have disk redundancy," only to find that database corruption doesn't care about your redundant storage. There's the one where they only took log backups... for years... and never planned for, let alone tested, a restore. I can keep going (and going, and going, and...).

Planning for an emergency is good. However, your planning must be realistic. It must be tested. And it has to be resilient enough to actually survive the emergency that you're currently experiencing. Don't make the mistake of thinking that, hey, you don't need to mess with that weird radio technology when you can get a POC radio and talk around the world, no licensing or knowledge necessary. Oops. Maybe a bit of knowledge is useful. Same thing goes for your databases.

<editor: sigh... radios>

I said hush. I finished with databases.