Diagnosing an install problem: a case study in indirect failures
Today I tried to Kickstart-install two SATA-based machines we have in for evaluation, booting them from USB memory sticks. Unfortunately it didn't work; something aborted about when our customization stage took over. (I didn't get to see what, because it promptly rebooted.)
One of my longer-term irritations with Anaconda is that you only get a binary choice of 'always reboot afterwards' or 'never reboot afterwards'; there is no option for 'reboot if all went well,
On being nibbled to death by moths
Last Friday I used the phrase'being nibbled to death by moths' to describe part of my day. Isuppose I should elaborate on that, lest people think that Toronto hassome really peculiar insects.
This is my way of summing up all of the routine five and ten andtwenty minute bits of system administrator work that come upthroughout the day. Reading a stream of email, making sure the backupscompleted, checking the logs, monitoring incoming spam streams,checking for important tech news (like security updates)
Mutating Referer Spammers
Last week's Referer spammers have changedwhat they're shilling for and mutated their methods. Currently theyseem to be shilling for online poker, although instead of clickthroughpayments they seem to be angling for 'affiliate' payouts when they getpeople to sign up at places like pacificpoker.com, fairpoker.com,partypoker.com, and 888.com (which seem unrelated to each other).
All of the websites being Referer-spammed for are still at the IPaddress 64.4
Weekly spam summary on August 20th, 2005
The overall SMTP connection rate has dropped from last week , down to 140,000 SMTP connectionsfrom at least 36,000 different IP addresses. The SMTP frontend hit ahigh-water of 16 simultaneous connections, I believe relatively earlyin the week, so I suspect we saw the spillover from last week'straffic burst last Sunday and maybe Monday and then a normal rest ofthe week.
Kernel level IP rejections:
Host/Mask Packets Bytes207.2
Disk setup in our Fedora Core 4 Kickstart environment
The problem with working on automatic installers is that testing themis such a slow and tedious process. These days, it takes about 30minutes to reinstall Fedora Core 4, with patches and local changes, onmy test machine. (Then when something looks like it works, I have todo three tests instead of just one: upgrade from Fedora Core 2 (withan extra reinstall to get back to FC2), reinstall Fedora Core 4, andinstall from a USB memory stick.)
Because of this
The March of the Cheap
Once upon a time, a collection of companies ruled the workstationmarket, selling what were then called '1M^3' machines, machines withone million MIPs, one megabyte of RAM, and one million (black andwhite) pixels. With computing so clearly heading their way, companieslike Apollo, DEC, IBM, HP, SGI, and Sun had futures so bright they hadto wear shades.
Computers have long since surpassed the '1M^3' level of performance,yet of the companies
The pains of modern disk storage
I spent today building some of the infrastructure that will hopefullylet us safely Kickstart-install Fedora Core 4 on the new generationof SATA-based PC workstations. (Safely, because destroying a serverby accidentally starting a Kickstart install on it is one of my quietnightmares.)
What makes this difficult is a a simple decision, by itself sane andrational: the Linux kernel people decided that rather than making up anew set of block devices for things like USB memory sticks or SATAdisks, they would just reuse
Parallelizing DNS queries with split
So there I was the other day ,with 35,000 IP addresses to look up inthe SBL to see if they were there.Looking up 35,000 IP addresses one after the other takes a longtime. Too long a time.
The obvious approach was to write a SBL lookup program that internallyworked in parallel, perhaps using threads. I was using Python and ithas decent thread support, but when I started going down this route itrapidly started looking like too much
Annoying RSS Feed Tricks
The RSS feed tricks that are really annoying me right now areall the different ways people have invented to serve partial entrycontent. Almost all of them are bad, plus the basic idea is bad too.
Serving partial entries implies that the blog authors don't expecttheir readers to be interested in most of their words (otherwise, whymake them go through extra effort to read them). The only good reasonsfor this that I can think of offhand are that very long entries orentries on a huge variety
Remember to think about the scale of things
One of the famous computer programming quotes is 'prematureoptimization is the root of all evil' (C.A.R. Hoare quoted by DonaldKnuth; attribution dammit (tm)).
A related issue is 'think about the scale of what you're planning'.A recent LiveJournal story provides a lovely example of this. To quote from it:
An increasing number of companies (large and small) are reallyinsistent that we ping them with all blog updates, for reasons Iwon't rant