The Life of a Sysadmin

Happy 2010 everyone! Now that it seems to be well and truly under way, I feel I can say that safely.

It's been busy so far. All the stuff I didn't do in 2009 is still on my plate...which is obvious, right? but it still caught me by surprise after the 3 days doing Xmas maintenance on my own. It was easy to forget that there are, you know, people waiting to show up and do work.

Like the new students we've got for one of the faculty members. I'd upgraded OpenSuSE on their new workstations over the holidays, then when they came in yesterday the carefully-tweaked dual monitor displays weren't working. Arghh.

Or the guy who's let me know that he wants to get moving on the MySQL/PHP website he's building...which reminds me that I've still got to move the website to a virtualized machine. I'm tempted to do that RIGHT NOW and put his site in there, but I don't think that'll be the best way to do it.

Or the new project my boss is part of, which involves researchers from across Canada. For me, it's a new website, hardware recommendation and purchases, maybe a new LDAP server. I could add a new root suffix to the existing LDAP server, but

a. we don't need it yet a. that seems like it'll make it more difficult to move later a. while I can create one in the existing LDAP server (Fedora/389/CentOS DS), the cn=config tree seems suspiciously empty of any entries related to the new root...so I'm leery of trusting it.

I still haven't sat down yet and tried to plan my year. Partly I've been busy, partly my planning tools are a bit of a mess (daytimer + orgmode + RT). But at some point I need to get my priorities straight and oh, how I long to have them straight. I feel a bit like I'm spinning my wheels right now.

Ah well. In other news, Xmas was good; my kids got two guitars (one acoustic with an Elmo sticker, one fake double-neck electric) which makes four guitars they have now. Since they no longer have that to fight over, they've taken to fighting over a microphone (cardboard tube stuck in a toy that acts like a stand). But damnit, they're still cute.

Finally: Just for fun right now I did a word count of all my blog entries. I've been blogging since 2004, and I've got something like 158,000 words. Amazing. And there are still some entries I've got to grab from my old Slashdot journal.

From (I think) a fellow Canuck:

A few weeks ago, I was sent a power point presentation on the "[Dynamic
Planning for COIN in Afghanistan][3]". I looked at it briefly, but thought
that it was some kind of joke; so, I flushed it immediately. However,
I received it from another source. So, it appears the joke is on me.

A quick look at this bird’s nest of a concept, would seem to suggest
that Dilbert or some escapee from the Project Management Institute has
taken over planning for COIN operations in Afghanistan. What I see is
yet another attempt to take a complex human activity and turn it into
an MBA project management flowchart. I can see the thinking, “Now that
we have the power point correct, we are sure to win the war in
Afghanistan!” In fact, I’m sure that, if we showed this power point to
the insurgents, they would throw in the towel, convinced that our
superior power point skills indicate that we cannot be
defeated. Really, I don’t know how we fought wars before power point.

Original post here. Found via WarHistorian.org (well recommended).

• This Nagios check looks for extra Wordpress admins.$ARG1$ should look include "-w min:max -c min:max", giving the acceptable ranges; in my case, I know I should have 3 and only 3 admins, so I have "-w 3:3 -c 3:3".

  define command{
  command_name check_wp_admins
  command_line $USER1$/check_mysql_query -q 'SELECT COUNT(wp_users.user_login) AS "Admins"
                         FROM wp_users, wp_usermeta
                         WHERE wp_usermeta.meta_value LIKE "%administrator%" AND
                         wp_usermeta.user_id=wp_users.ID' -H $HOSTADDRESS$ $ARG1$
  }
  

• This one looks for nasty Wordpress posts. Note the dependency on MySQL's regex command. In my case, I know that I do not have any posts with these words, so in $ARG1$ I have "-w 0:0 -c 0:0".

  define command{
  command_name check_wp_nasty_posts
  command_line $USER1$/check_mysql_query -q 'SELECT COUNT(*)
                         FROM wp_posts
                         WHERE post_content REGEXP "iframe|noscript|display"' -H $HOSTADDRESS$ $ARG1$
  }
  

• And a Python script that picks out a random client, job and file from Bacula's database and tries to retrieve it. It's not ideal -- checking for sanity is left as a DARPA Grand Challenge -- but at least it's one way of exercising backups. I anticipate running this often. If anyone's interested, let me know.

The more I work with Python, the more I don't just like it but admire it.

Ugh...not much more right now. I've got a blocked eustachian tube that I'm self-medicating with a Python script^W^Wcold medicine, and the acetominiphen in it is making me hazy.

I came across these pictures taken by the Cassini probe, while looking for pictures of Saturn to show my oldest son. They're beautiful, and I'm heartsick that I'll never get to see these views first-hand.

Here's how to piss me off:

• Refuse to update your software so that it uses a Makefile. Yes, I know you're not only 'way smarter than I am but you've got a source tree going back to 1977. I don't care; editing six different shell scripts is not the way to do things.

• Sprinkle those six scripts with assumptions about which software is present and where the source code is being compiled. Document most of them.

• Run tests but carefully delete all the results; don't include an option to save them. That way I have to edit your scripts to figure out what the hell went wrong.

• Assume the presence of csh for everything, rather than POSIX-standard sh.

• Put configurable options inside shell scripts, rather than in a configuration file or allowing them to be set by arguments to those scripts.

• Include directions like "Perhaps change foo, bar and baz", without explaining why or what they're set to. When tests later fail because you didn't properly set foo, bar and baz, don't explain where these are set or how they affect the tests.

• Set a hard-coded location for temporary output. Die silently when those locations aren't present, rather than explaining why or offering to create them or using /tmp. Refuse to overwrite already-present files, but don't explain this anywhere; instead, say that they might be useful next time.

• Have important variables, like the hard-coded location for temporary output, set in two or more different places. Suggest editing some of them.

• Have test failure indicated by "!!FAILED", ensuring a moment's confusion about whether that means "FAILED!", "NOT FAILED!" or "NOT NOT FAILED".

_Update, April 20th 2010:

• Be so glad that you're done installing this software that you never come back to document how you did it in the first place, leaving no clues but a short rant on your blog. Sigh.

• Dive into Mark writes about the iPad:

  Once upon a time, Apple made the machines that made me who I am. I became who I am by tinkering. Now it seems they’re doing everything in their power to stop my kids from finding that sense of wonder.

• Paul Vixie on the proposed EDNS0 extension to:

  allow Authoritative Nameservers to return varying replies based upon the network address of the client that initiated the query rather than of the client's Recursive Resolver.

  His response?

  if we're going to add client identity to the query, can we do so in a more general way? i'd like to know lat-long, country, isp, language, and adult/child. and the ip address should be multiprotocol, covering ipv6.

• More from Paul Vixie on What DNS Is Not:

  The Internet economy rewards unlimited creativity in the monetization of human action, and fairly often this takes the form of some kind of intermediation. For DNS, monetized intermediation means lying.

• Bacula released version 5.0.0, including file deduplication using base jobs (project #6).

Off to go craft a budget for next year.