I’m looking to setup a site with 49 edge switches connected by fiber to a central switch. 3 VLANs will be setup to handle data, telephony, and streaming media. each edge switch should have provision for 2 SFP modules for failover, and the core switch needs to have the provision to handle this failover. i’m getting lost on the Cisco site with their specs and recommendations. if anyone could suggest a suitable model number for the core switch and the edge switch, it would be really appreciated.” ...

This is a bit long-winded, and wordy. If you’ve come here looking for tips on improving commvault backup performance and/or throughput, then you should click here for the good stuff. It’s been a long time since I blogged. It’s been a really long time since I blogged about anything we’ve been doing at $Dayjob. I’ve spent the better part of the year working on sorting out the backup solution here. ...

Lessons learnt from the Unlock London Hackathon. I had an email on May 16th, asking for some assistance in setting up the wifi network for another hackathon. After my impromptu assistance at LondonRealtime went down so well, and “saved the network”; apparently I was a natural choice for the next one. At least I knew (sorta) the network layout at White Bear Yard. The main difference between this one and the last, was a bit more warning on the side of “We’re gonna need wifi”, but there was still a clause that it’s gotta be “flawless” - Their words, not mine. ...

As time goes on, and I find myself dealing with more and more large companies, it’s pretty obvious that there’s a recurring pattern. Many large companies are awful to work with. Dell are pretty bad. On one recent occasion, we had the need at work to call Dell Support for a mysterious problem with our shiny new blade servers. The biggest problem when dealing with Dell in particular is their scripted support technicians on 1st line. Most other companies seem to hire people with the capability of abstract thought. Dell hires chimpanzees to read from scripts. Woe befalls you if you attempt at any point to deviate from the script, and heaven forbid you ask a question that they don’t have the answer to. If that happens, there’s a 1 in 3 chance of any of the following outcomes. ...

Smokeping is one of my favourite diagnostic tools for tracking down sporadic network issues. You install it, configure it with a list of hosts, and it pings them regularly, and keeps track of the round-trip times, latency, packetloss, and so on. The web frontend is a Perl CGI script, and as a result, it’s a bit of a bugger to make it work on Nginx. I wasn’t gonna install Apache just for this one thing… ...

Does this look suspicious to you? 85.25.184.184 - - [13/Mar/2012:00:41:09 +0000] "POST /3293b4da67abffca2460244619d9a8bca3f4c401.php HTTP/1.1" 200 1096 "-" "Mozilla/5.0 (Windows; U; MSIE 9.0; WIndows NT 9.0; en-US))" 85.25.184.184 - - [13/Mar/2012:00:42:57 +0000] "POST /3293b4da67abffca2460244619d9a8bca3f4c401.php HTTP/1.1" 200 359 "-" "Mozilla/5.0 (Windows; U; MSIE 9.0; WIndows NT 9.0; en-US))" 85.25.184.184 - - [13/Mar/2012:00:42:57 +0000] "POST /3293b4da67abffca2460244619d9a8bca3f4c401.php HTTP/1.1" 200 450 "-" "Mozilla/5.0 (Windows; U; MSIE 9.0; WIndows NT 9.0; en-US))" 85.25.184.184 - - [13/Mar/2012:01:14:12 +0000] "POST /3293b4da67abffca2460244619d9a8bca3f4c401.php HTTP/1.1" 200 354 "-" "Mozilla/5.0 (Windows; U; MSIE 9.0; WIndows NT 9.0; en-US))" Sure as hell looks dubious to me. Looks like something doing something that it oughtn’t. ...

This is a continuation to my earlier blogpost about hacking initrd.gz. After it’s installed with the modified installer kernel, and finished building, the installed kernel doesn’t have the e1000e driver. This is because the netboot installer pulls in a kernel and it’s wherewithall from apt. It also gets a new initrd. As a result, I’ve decided to build an e1000e-dkms module, and we’ll specify the preseed installer to install that, along with linux-headers-generic, linux-headers-2.6.32 and build-essential. ...

This morning, I did something unquestionably naughty, and totally got away with it. A little background. We just had some brand new workstations delivered.. They turned up yesterday afternoon. They’re high performance 3d workstations with an Intel DX79TO mainboard. This mainboard has the intel 87529 Gigabit Ethernet controller. I wouldn’t normally pay so much attention to the controllers and so on that are actually on a board, but in future, I will. ...

Well, that’s certainly another strange problem. We have a tendency to build our own DMG images for certain bits of software we roll out here. Sometimes we’ll incorporate our own patches, other times it’s just to make the application structure more FHS compliant, and stop it “shitting all over the filesystem”, as we so charmingly term it. In the past, we’ve used InstallEase to build a DMG and PKG installer for OSX. It’s been pretty good up until last week, when one of our engineers built a PKG that had a nasty side effect of destroying the System Preferences once installed. We initially pegged this as “another weird thing about Lion”, and rebuilt the image, and so on. ...

Puppet really is great. Don’t ever get me wrong there. It’s saved me masses of time and money over the last few years, and allowed me to do my job quickly and efficiently. That said, it really does have issues with scalability. After about 20-30 clients using WEBBrick, everything kinda falls over a bit. We had this problem at Baseblack. We’ve now got ~60 workstations and rendernodes all using Puppet for configuration management and software deployment. It’s great. It far simplifies the process of rolling out updates and upgrades to new build machines. ...