A while back I had a postiion as a system administrator for a farm services company. They had previously been with an MSP who treated them poorly, and were looking to hire in so here I came. I won’t bore you with the details, but at some point I needed to decomission an OOOOLLLDDDD netapp disk shelf thing and replace it, so I decided to try my hand at created an iSCSI lun on a physical linux host. I don’t recall the exact circumstances surrounding it, but needless to say it shit the bed one morning. One of the luns, anyway. I don’t remember if I panicked and had to just figure something else out or what, but I ended up rolling Windows Server and deploying the StarWind iSCSI software on that server. When I left it was still running solid, but the real question is what the hell could have possibly happened with the targetcli setup?

After going through targetcli setup again on a new system, it is actually really intuitive. I admit I know significantly more about Linux administration now than I did, but I can’t really think of a scenario where this thing would poop out like that. I mean, maybe the disk the LUN was on filled up or something? I think it staticly allocated the data so that’s probably not the case. In any manner, at the end of the day I don’t really understand this software, just basically how to set it up.

This can be a scary prospect. It’s not like VMware, which I’ve been using for 10 years at this point and am completely comfortable with. Or even to a lesser extent Linux as a server OS, which I’ve also been using for a long time and am fully comfortable with. Getting elevated at work to an increased resposnibility position has helped with this immensely. Before Chris left, he told me that “you’re going to have to start googling more shit and trying to figure it out yourself”, and he was absolutely right.