One of the things that has happened around the university lately hasbeen a review of our campus-wide student email offering. You canprobably guess the result; pretty much everyone involved (especiallythe students) was very firmly in favour of moving to GMail or somecompetitor, provided that various side issues could be adequatelydealt with.

(Yes, some of these side issues are not trivial .But they are apparently significantly smaller than they used to be, andaddressable.)

There's a part of me that

Here's something important if you're designing or considering a newanti-spam system (as we may be at some point). It may sound obvious,but I think it's not:

If you run an email system, part of your job is filtering spam foryour users .

It used to be that you could provide your users a collection ofanti-spam options and tools and settings and so on, and consider yourwork done. Those days are over and gone. Much as with

Here is something that we have been forced to think about lately: areUPSes really a good insurance policy against power problems, or are theyinstead an extra source of problems? In short, does using UPSes reallyincrease your net reliability?

The problem with UPSes used by themselves is that they are another pieceof machinery to fail (and they are a moderately complicated piece ofmachinery at that). And UPSes do fail; for example, we recently hadan incident where a UPS reset itself out of the blue,

We almost lost a ZFS pool today. More accurately, we did lose a ZFSpool, and then we were able to get it back because we were lucky enoughto have a Solaris 10 update 8 test machine handy. But except for thatbit of luck, we would be in the very uncomfortable position of tellingan important research group that for the first time ever we'd just lostnearly a terabyte of data, and it wasn't even because of a hardwarefailure, it was because of a

There are some local developments around the university that have hadme thinking about potential alternatives to our current system of spamfiltering; right now we rely on some commercial software that theuniversity has a site license for, but it's possible that the licensewon't be renewed at some point. In the process of this, I've realizedsomething.

One inobvious advantage of outsourced spam filtering is that it meansthat you are not in control of what gets filtered. That may sound likea disadvantage, but

In an ideal world, disks would either be working completely normally orobviously broken (either producing errors or being completely dead);if a drive wasn't actively reporting problems, you could assume it wasworking fine. I am here to tell you that sadly we do not live in thatworld, at least not with SATA drives.

What we've now seen several times is SATA drives that degraded quietly;they didn't particularly report errors, they just started performingterribly (by their usual standards)

Sysadmins have a habit of accumulating little personal scripts to makeour lives more convenient, often things that make perfect sense to usbut are too peculiar or specific to be worth installing generally.Since these are personal hacks, they're often bashed together in verycasual ways; if it mostly works for us, it's good enough.

From personal experience, I now have a suggestion for these scripts:make sure that they have clear or at least comprehensible errormessages, not little cryptic ones. If you do

A while back I wrote about how I monkey-patch Python modules fortesting , selectively replacing variousfunctions they provide with test ones that return the results I like.But it's not the only way to do this. If monkey-patching system modulesthis way makes you nervous (perhaps due to its potential side effects),there's another way to do it: you can entirely replace the system moduleitself in the module you're testing.

After all, import ed modules are (more or

One of the things that blogging gives me is the chance to be verywrong in public. Yesterday , I claimed thatC's stdargs didn't let you peel some arguments off the front of a va_list and then pass the shortened list to another function,such as vprintf() . Well, no, and now I'll tell you why I'm wrong.

I'm clearly wrong in practice on x86 Unix machines with gcc, as a simpletest program easily demonstrated once