Come ottenere risultati

Quando hai un server il cui produttore ha mollato il business sette anni fa, il cui controller RAID ha causato otto kernel panic lo scorso weekend, e su cui gira ancora una copia di Apache con una cinquantina di virtual host di cui nessuno si vuole assumere la responsabilità, neanche per dire quali servono ancora e quali no, cosa puoi fare? Mandare un messaggio alla mailing list techie della compagnia che recita

"XXX has gone down eight times last weekend. This is clearly intolerable.
This is the list of virtual hosts on the box

...

I want to know which ones must go on another machine and which ones must go, period.
I will kill one virtual host every half hour until my demands are met. There will be no negotiations."

Venticinque minuti dopo avevo la lista completa. Certe volte adoro questo lavoro.

---

When you have a server whose manufacturer went out of business seven years ago, whose RAID controller caused eight kernel panics last weekend, that still runs Apache with some fifty virtual hosts for which nobody is willing to claim responsibility, not even to say whether they are still needed or can be shut down, what can you do? You send a message to the company's techie mailing list, one formulated more or less like this:

"XXX has gone down eight times last weekend. This is clearly intolerable.
This is the list of virtual hosts on the box

...

I want to know which ones must go on another machine and which ones must go, period.
I will kill one virtual host every half hour until my demands are met. There will be no negotiations."

Twenty-five minutes later I had the complete list. Sometimes I love my job