Graylog – store, search, and analyze log data from any source.
xkcd is spot on once again.
Fedora 23 has been released yesterday, and as a big fan and a long time user I had to upgrade my laptop (from Fedora 22) immediately. Or at least try.
Usually, the process is quite simple and doesn’t take much figuring out. This time it was somewhat different though. At first, the recommended upgrade command changed slightly from the nice and simple fedup to dnf system-upgrade.
The first attempt downloaded all packages, ended with a cryptic error along the lines of “No updates for kernel packages were found“. Hmm, weird. I thought maybe this was caused by me forgetting to run “dnf update” before the upgrade process. So I did. Some updates were installed, but there were no kernel packages among them. I rebooted the laptop just in case.
The second attempt for some reason failed to find any of the previously downloaded packages, so I had to wait until the almost 2 GB get pulled down again. Result – same error about missing kernel updates. Hmm, again.
After poking around for a bit, I realized that I had previously configured yum to exclude some packages from updates (kernel, kmod-wl, kmod-VirtualBox, and VirtualBox). These settings were picked up by dnf. Editing both /etc/yum.conf and /etc/dnf/dnf.conf to disable the exclude fixed the issue. “dnf update” now pulled some updates to the kernel. Another reboot.
The third attempt once again lost all the downloaded packages and I had to wait some more. This was annoying, especially at 1am now. The process of downloading finished OK and this time there was no errors. So “dnf system-upgrade reboot” should do the trick now, right? Wrong!
Surprisingly, there was no boot menu for system upgrade upon reboot. So I went with Fedora 22 boot option. Which resulted in a brief screen saying “Preparing for upgrade“. Which then disappeared and the system booted back into Fedora 22. Now that was interesting.
It took me a while to find the issue. The problem was that my Fedora 22 laptop wasn’t a clean install, but an upgrade from Fedora 21. That shouldn’t be a problem, but it is, due to this bug. My /etc/os-release file didn’t have the VARIANT and VARIANT_ID variables (thanks to this blog). So after adding these lines:
VARIANT="Workstation Edition" VARIANT_ID=workstation
I’ve rebooted once again, and started the fourth attempt. Downloaded packages were gone once again, so I had to wait a bit more. This time the faster office Internet connection helped to save some time. Download finished OK. Another reboot. Once again, there is no option to upgrade Fedora, so I’m going into Fedora 22 boot. Finally, now there is the upgrade screen! After some preparation time, the packages were installed, the machine rebooted, and now I’m Fedora 23 user.
Checking my /etc/os-release file I see that both variant variables are gone now. This feels weird, but hopefully won’t cause issues in the future. If it will, I’ll probably Google for three hours before finding this very blog post.
Fedora Magazine covers “6 great monospaced fonts for code and terminal in Fedora“. Their choices are:
- Source Code Pro
- Fira Mono
- Droid Sans Mono
- DejaVu Sans Mono
It’s been a while since I considered a change to the monospaced fonts that I’m using. The top three fonts in my list from a while back are Fixedsys Excelsior, Monaco, and Microsoft Consolas. I used Fixedsys Excelsior almost exclusively in all my terminal windows.
Linux Weekly News reports that Red Hat acquires Ansible. There are quite a few configuration management tools around, and it was only the matter of time until Red Hat, with all its corporate client base, would buy one. Or pledge allegiance. My personal preference would be in Puppet, but Puppet comes from the Ruby world, where’s Red Hat is more of a Python shop.
Ansible’s simple and agentless approach, unlike competing solutions, does not require any special coding skills, removing some of the most significant barriers to automation across IT. From deployment and configuration to rolling upgrades, by adding Ansible to its hybrid management portfolio, Red Hat will help customers to:
- Deploy and manage applications across private and public clouds.
- Speed service delivery through DevOps initiatives.
- Streamline OpenStack installations and upgrades.
- Accelerate container adoption by simplifying orchestration and configuration.
The upstream Ansible project is one of the most popular open source automation projects on GitHub with an active and highly engaged community, encompassing nearly 1,200 contributors. Ansible automation is being used by a growing number of Fortune 100 companies, powering large and complex private cloud environments, and the company has received several notable accolades, including a 2015 InfoWorld Bossie Award, recognizing the best open source datacenter and cloud software.
Regardless, though, of my personal preferences, these are good news for configuration management and automation.
One of the very few things we still rely on from Microsoft at work is Office 365. Not because it is so great, but because I simply didn’t have the time to move away yet (quiet Christmas season is coming soon). Most people don’t get exposed much to it anyway, using Evolution or Thunderbird email clients, or forwarding everything to Gmail altogether. But as an administrator of the service I get constantly annoyed by the “there are problems with your domain” notification. After ignoring it for about a year, I decided to finally fix it. All that was needed is a couple of records in the DNS zone.
Unfortunately, the instructions Microsoft provides don’t quite well apply to Amazon Route 53 user interface. It took me a few tries and some searching to find the right ones. Here they are, thanks to a comment on this page:
1. Log into Route 53.
2. in the hosted zones window, check the box next to YOURDOMAIN.com
3. click [ > Go to Record Sets]
4. Click [Create Record Set]
5. Enter _sipfederationtls._tcp , in the Name field
6. Switch type to SRV – Service locator
7. In the Value window Enter: 100 1 5061 sipfed.online.lync.com
8. Click [Create]
9. Click [Create Record Set]
10. Enter: _sip._tls , in the name field
11. Switch type to SRV – Service locator
12. In the Value window Enter: 100 1 443 sipdir.online.lync.com
13. Click [Create]
“Five Linux-Ready, Cost-Effective Server Control Panels” reviews 5 some alternatives to cPanel, which, they say, is rather expensive. My beef with cPanel is not the price, but the technical merit. Even though I love the fact that it is written in Perl, I don’t agree with its “let me handle everything” approach.
cPanel installs all the software that it helps to manage. This might be a “so what” issue for most people, but not for me. I like my servers clean. And I want to utilize the tools that already come with my server – RPM, yum/dnf, etc. Control panels can help with routine, but when something breaks, I should be able to go to the config files and deal with the problem using the distribution’s recommended ways. cPanel, unfortunately, breaks that. It downloads sources, applies patches, locally compiles things, and has its own layout for configuration files. That’s too much mess for me.
I haven’t used any of the other control panels reviewed in the article (I usually prefer the command line way), but I hope they aren’t as intrusive and abusive as cPanel. Sometimes control panels are useful for providing a bit of help to non-technical users (create mailbox, change email password, backup the website, etc), but if they are as needy as cPanel, thanks, but no thanks.
HTTP Status Dogs – Hypertext Transfer Protocol response status codes. And dogs. If you are even a tiny bit familiar with HTTP or dogs, this will put a smile on your face. I’m thinking to use these as default error pages from now on.
OverAPI.com – Collecting All Cheat Sheets