This week I found myself pondering on a question that has come up a few times in the past, what exactly does a sysadmin do when they are not firefighting problems?
Luckily, not too much of this pondering took place. Storm Eunice had barrelled through the country, stating on Friday and had caused significant disruption all round. While we escaped fairly lightly here in London, we still had experienced some disruption to power and phones etc.
When I arrived on Monday there was a peculiar problem with the internet. We had an internet connection, however the were a number of things not working properly. Most notably, the internet. We could only really access sites hosted by Google. So I did what any good IT person would do and turned everything off and on again. Specifically, the router and firewall. This restored full connectivity and enable me to go and make a coffee.
Further investigation showed that we must have lost power to the most at some point and consequently our primary domain controller had failed to reboot. Reattaching the hard disk (I believe I have ranted extensively about this in previous posts) allowed for a successful reboot but also required a manual reinstall of Azure AD connect in order to fully restore the sync service. That took at least half a dozen attempts to get working as I ran into a number of different errors including having to manually reimport the AD power shell command-lets. A further knock on was that a number of students were unable to access their accounts due to password issues, luckily a password change and sync resolved all of those issue. I am sure there were a few other issues, but they have fallen from my fallible memory.
Thursday brought another first for me. I was visited by the lovely Mark and Jamie from Misco in order to discuss the work that we are putting out to tender. Primarily, re-cabling the entire network. It was a great experience and was really helpful to talk through the issues with someone far more knowledgeable than me in regard to networking.
After that, the fun didn’t stop. Over the course of Thursday, our printers all began giving error messages and intermittently refusing to print. Some investigation revealed that the issue may well have been caused by an update failing to install. A simple fix I though foolishly. Alas it turned out the reason that the update had not installed on the print server was because automatic updates had been disabled on the server sometime in early 2019. In total it took the better part of 8 hours to bring the server up to date. Full functionality restored.
So that at least answers one of the questions, when you aren’t fighting fires make sure updates are being applied properly. I also discovered the excellent resources on reddit.com/r/sysadmin . They have a wealth of knowledge in the wiki and I have purchased a book recommendation to help me understand better what I should be doing.
