Snowing for about 24 hours now… Okay, so not as bad as up north… but I’m in Dallas, TX, we shouldn’t have snow…

Statement as of 7:10 PM CST on February 11, 2010

… Greatest all-time calendar day snow on record set at Dallas Fort Worth…

… Record 24-hour snow for February set at Dallas Fort Worth…

… Record daily maximum snowfall for February 11th set at Dallas Fort Worth…

Through 7 PM CDT… Dallas Fort Worth Airport has recorded 7.9 inches of snow. This breaks the greatest calendar day snow on record. The old record is 7.8 inches set on January 15… 1964 and January 14… 1917.

This also breaks the record 24-hour snowfall for February which was 7.5 inches… set on February 17… 1978 and February 25… 1924.

Obviously… .this shatters the record daily maximum snowfall for February 11th… of 1.4 inches which was previously set in 1988.

With the snow continuing… the snow total will increase. An updated record event report will be sent later this evening.

Back in November, a post on SysAdmin’s Journey had a follow-up to a Yahoo document titled Best Practices for Speeding Up Your Website, specifically the fourth section. The author stumbled across an issue with AOL users not being able to see images at all.

The Yahoo document is a great source of little hints and tips, and the SAJ article is a good gotcha on one of those tips. Handy to keep around if you’re planning on tweaking your servers for a little bit of a performance boost for your users. Worth reading the document for some ideas…

I’m obviously playing catch up with a number of posts I’ve been meaning to do. This was something from something I read back in November by Tom Limoncelli¹, but was something I had planned on writing about anyway. The post, titled “Run, run, run, dead”, brings a nice analogy of things breaking in the analog/digital eras, and points out that as system administrators, we should be using the analog method of monitoring.

“Major Hayden” has a handy tip for those running Irssi on a server in a different timezone…

/load perl
/script exec $ENV{'TZ'}='CST6CDT';

Also links back to the Irssi documentation with other handy tips.

For a while, I’d been using an offline reader¹, then Google did some upgrades, so I swapped. I’ve been using Google Reader since late 2006, diligently serving my news to me, in its simple, and easy to use interface.

Recently I decided to tinker with a new offline reader, RSSOwl. Simple interface, easy to use, customizable searches, article flagging, and sharing features. Well worth a peek…

PacketLife as an excellent collection of cheat sheets for networking professionals. Well worth a quick look and bookmarking for later.

Sysadmin1138 has an interesting followup to an Ask Slashdot question about IT admin abuse… It’s a pretty complete answer, and good insight into what we system administrators have access to, and what we can do.

It’s easy to see from the response that abuse is easy for us, we have the access, the temptation is there, but we’re not all bad. I’ve known people to cross the fine line between doing their job, and looking a little deeper at somebody’s email account whilst “fixing”. It’s not hard for us to peek at chat conversations whilst we sniff out connectivity issues for servers.

Whilst we may possess the power to do so, most of us follow ethics, if not officially documented ethics, our own personal ones. And whilst there are always a few bad apples, it applies to all industries, the large majority of us are good, don’t let the few bad ones make you think we’re all out to spy on you.

Courtesy of LOPSA, I stumbled across Matt Simmons blog¹. A recent post discussed us all being a large source of information, and how we all suck it in from various sources. He lists off his recent change in using mailing lists, and several other sources of information (including an extensive feed export).

Courtesy of Matt, my feed collection is huge, instead of just large, and was the reason I just flipped over to RSSOwl (see my previous post). I did some reorganization, and my new collection can be found here. I did some tidying, some reorganization, removing of duplicate and broken feeds (something RSSOwl flags clearly). He has some interesting sources, and contacts, worth checking out his sources of information.

Funnily, I realized that Matt is also using the Cleaker theme, which I decided to take a look at over 2 years ago now. Interesting how sysadmins brain’s work ;)

Due to some unforeseen issues, we had to move our RAS server into a new DHCP scope, and VLAN today. This is usually relatively easy…

• Change Server IP
• Change switch port VLANs
• Change firewall to point to new server

This went smoothly, all being completed in about 2 minutes. But on checking the DHCP scope, the RAS server wasn’t requesting any IP addresses. I recycled the RAS service, but that didn’t seem to help, only hint of an issue for RAS was the generic message about DHCP server not being reachable…

Unable to contact a DHCP server. The Automatic Private IP Address 169.254.71.62 will be assigned to dial-in clients. Clients may be unable to access resources on the network.

Apparently my brain has been on overdrive, and I forgot that moving the server to a different VLAN also made the DHCP requests not reach it. This is easily resolved with a few commands on our switches…

# config t
(config)# int vlan77
(config-if)# ip helper-address 10.13.1.3
(config-if)# exit
(config)# exit
# write

A quick restart of the RAS service, and the new VLAN is hosting its assigned IP scope.

Unfortunately from 1716 CST through to 1845 CST, the Netdork server was unreachable due to a runaway process. Unable to get onto the server I was forced to initiate a remote reboot of the machine to regain control.

Now the server is back up and stable, I’ll be implementing controls to stop these processes from running away with themselves (I’m sure they went to Vegas).