Depending on iconv -t ascii//translit on Red Hat was a mistake
I recently re-learned that moving code from one Linux distribution to another doesn’t always pan out as expected. Especially when that second distribution is Red Hat Enterprise Linux.
I recently re-learned that moving code from one Linux distribution to another doesn’t always pan out as expected. Especially when that second distribution is Red Hat Enterprise Linux.
I decided I wanted to host my WordPress installation as a hidden service on Tor instead of backporting all my existing content to Hugo. I previously ran Hugo on my onion site and even though I still want to make that move eventually, for now, I’m sticking with what I already know. Besides, putting arguably the worst content management system ever invented on the dark web seemed like a fun venture.
Lately I’ve noticed a steady increase in the amount of referrer spam I’m getting, so I decided to see if there was a simple way to trap and ban these bots. The typical approach is usually to maintain a blacklist of domain names and deny them using mod_rewrite rules. The downside to this approach is the amount of time and effort that goes into maintaining your blacklist.
I recently bought a new Raspberry Pi 3 and installed Slackware ARM current (hard float) on it. My goal was to compare the performance of the hard float port against Slackware ARM 14.2 (soft float), which is currently powering this RPi3 hosted website.
That tag “stuff” is not working on our corporate website, please fix asap the costumer complained. Sure, will do immediately I replied confidently, believing this to be a simple matter of purging some old cache or refreshing permalinks. Sadly that was not to be the case so I ended up having to get my hands dirty. To my absolute horror, the site was running one of those godawful themeforest themes.
So the annual “Slackware needs PAM and Kerberos” thread is going strong over at linuxquestions.org at the moment. This particular topic always seem to awaken a collective inferiority complex within the Slackware community, where users are aggressively refuting any claim that Slackware is not a viable choice for business use (you can do anything with some lines of bash right?…). At the opposite side you have users arguing that Slackware has become a niche hobbyist distribution due to its reluctance to implement mainstream technologies.
My plan was to have a Zimbra e-mail server running in VirtualBox using the internal networking model with a local domain. To make this work, I have a virtual pfSense box running on the same internal network to provide “outside” access. Additionally, pfSense is also running a DNS resolver so I figured it would be a trivial matter to add the local records I needed.
Looking through my server logs I noticed how Baidu’s web spider was causing an unexpected redirect loop while trying to index an image attachment page. Since I deliberately redirect all attachment page requests to the actual post owning the attachment, I decided to take a closer look. The following request triggered the loop:
Microsoft bingbot is a repeat offender on my RPi and as a consequence it’s subjected to extended periods of jail time. However, being a good netizen I like to provide an even playing field for all web crawlers. With that in mind, what would be the easiest way of purging all IP addresses from fail2ban?
Let’s Encrypt suffered from a major service disruption today leaving users unable to access various services. The cause of the problem seems to have been an update to Boulder (ACME CA) which has since been reversed.
In the last few days I’ve noticed a few unusual GET requests for supposedly exposed SSH private keys. All requests are following the same pattern:
It’s live, prepare to self-destruct in 3..2..1..
This website recently celebrated its second year of Raspberry Pi based hosting. It’s currently running on a RPi3 with Slackware ARM 14.2 (32-bit soft float). Somewhat to my surprise, the second year went by without a single glitch.
Perl upgrades on Gentoo Linux have been laborious for me in the past, but with Perl 5.24 that was no longer the case. I’ve previously managed Perl upgrades by using the oneshot option and manually resolving any remaining conflicts afterwards. With this upgrade though, it looked to be close enough so I decided to try with the backtrack option as suggested from the following output.
My plan to install the latest Windows 10 Creators Update fell short during the weekend due to an error identified as 0xc1900200. I was using the “Windows 10 Update Assistant” to perform the upgrade, and the assistant did initially confirm that my system was ready for the upgrade.
When you’ve been running GNU/Linux distributions for an adequate number of years, I do believe you’ll eventually find yourself walking the path to Mount Gentoo in hope of joining the ancient Greybeards. Many have met their demise on the road ahead, but armed with the Gentoo handbook we’re confident that it’s within our reach.
My office workstation recently went trough a Slackware release upgrade by following the excellent systemupgrade article from the Slackware Documentation Project. Personally I experienced a few snags along the way so I’ll add a few notes for future reference.
Time for another unfortunate run in with the OTRS 5 Daemon. Again I got a call from a customer informing me that OTRS had been idle for a day without creating any new tickets.
Just because Greg KH said that all users of the 4.4 kernel series must upgrade ;-) I’m happy to report that everything seems to be working as expected and have yet to notice any regressions.
Dlackware is not your average “GNOME for Slackware” project but instead aims to take the slack out of Slackware. What you get in return is the latest in “enterprise” technology. Dlackware delivers a fully functional GNOME 3.22 desktop with PAM, Wayland and systemd.