Another aphorism of system administration

Noticing when something shows up is easy; detecting when it goes awayis hard.

Like all aphorisms, this has exceptions. And if you want to see it thatway, it's a corollary of an earlier aphorism .

(An aphorism brought to mind as I contemplate our DHCP configurationfiles and wonder just how many of those Ethernet addresses are currentlymouldering in a dump somewhere.)


What I currently know about Fibrechannel versus iSCSI versus AoE

We're planning a significant capacity increase for our local SANstorage pool, which means that we've been trying to figure out whichSAN technology we want to go with. We don't have high performance IOneeds, so we're going for bulk storage: large SATA disks in RAID 5 insome sort of SAN RAID controller. We plan to use Solaris 10 on x86for our NFS servers that will use the SAN, with DiskSuite forfailover. DiskSuite has to own full disks


Weekly spam summary on February 17th, 2007

This week, we:

  • got 15,925 messages from 244 different IP addresses.
  • handled 23,465 sessions from 1,341 different IP addresses.
  • received 244,268 connections from at least 75,016 different IPaddresses.
  • hit a highwater of 16 connections being checked at once.

This is about the same as last week .The per day figures show some significant fluctuations:

DayConnectionsdifferent IPs
Sunday36,6

Why we do NFS fileserving with a SAN

Our storage infrastructure here has a number of NFS servers sittingin front of a pool of SAN RAID storage boxes using commodity SATAdisks. This is a somewhat unusual setup for a comparatively smallenvironment like ours; a far more common setup is to have the disksdirectly attached to the fileservers.

We have a SAN setup for a simple reason: failover between the NFSserver machines . We consider the server machines to be the things mostlikely to suffer failures, either hardware or software, or just to needdowntime. With


Programming fun

Programming fun is spending a couple of hours writing, revising, andtuning a DWiki feature that I'm not sure I'm actually going to like wellenough to keep.

I'm serious, not being sarcastic; I like programming, but not all ideasfor improving a program pan out. An idea that seems great in my mind canbe less attractive once I've made it concrete and explored all of thebits that I could gloss over when it was just thought-stuff. So I can


QOTD: There are three types of authentication

There are three types of authentication :

They are:

  1. Something you've lost,
  2. Something you've forgotten, and
  3. Something you used to be.

The full entry includes an illustrative story and bonus comments (and,unfortunately, a certain amount of comment spam, at least right now).

(From Richard Johnson of river.com.)


I hate hardware (Dell 2950 edition)

The Dell 2950 is a decent 2U server that has recently started beingpopular around here; like most recent servers, it has dispensed withPS/2 connectors and only has USB for the keyboard and mouse. It has fourUSB connectors, two on the front and two on the back.

If you plug your keyboard into the back USB connectors, the Ubuntu6.06.1 x86_64 server kernel hangs with ' BUG: soft lockup detectedon CPU#0!


Something all full-service backup systems should have

Having spent much of today wrestling with this very question, I have asmall suggestion for people designing full-service backup systems (bywhich I mean ones that have individual file indexes and an environmentfor restoring single files):

Please provide a command that summarizes all of the versions ofa file that you know about.

Most full-service backup systems can go back in time, so you can ask forthings like 'the version of the file on January 28th'. But what I reallywant is


RPM tricks for dealing with multiarch machines

Why do we need some new RPM tricks for dealing with multiarch machines?Simple:

; rpm -q glibcglibc-2.5-10.fc6glibc-2.5-10.fc6

That's not an error, that's just my machine having both the 32-bit andthe 64-bit glibc RPMs installed. On a multiarch machine it's routineto have RPMs with identical names, and rpm itself hasn't been fullyadapted for this situation


On Python's grammar

Python's grammar looks imposingly complex from the outside, but themore I've thought about it the more I've realized that it's simple butclever. In particular, the way the work is split between the tokenizerand the actual grammar simplifies both.

(Technically you could claim that Python's grammar is straightforwardbecause the tokenizer takes care of all the hard bits , but I consider this ducking the issue.)

The actual grammar istypical for an Algol-style language. You have the