How old our servers are (as of 2026)
Back in 2022, I wrote about how old our servers were at the time , partly because they're older than you mightexpect, and today I want to update that with our current situation. My group handles the generaldepartmental infrastructure for the research side of the department(the teaching side is a different group), and we've tended to keepservers for quite a while. Research groups are a different matter;they often have much more modern servers and turn them over muchfaster.
New old systems in the age of hardware shortages
Recently I asked something on the Fediverse :
Lazyweb, if you were going to put together new DDR4-based desktop(because you already have the RAM and disks), what CPU would youuse? Integrated graphics would probably be ideal because my needs aremodest and that saves wrangling a GPU.
(Also I'm interested in your motherboard opinions, but the motherboardneeds 2x M.2 and 2x to 4x SATA, which makes life harder. And maybe4K@60Hz DisplayPort
Canonical's Netplan is hard to deal with in automation
Suppose, not entirely hypothetically, that you've traditionallyused /etc/resolv.conf on your Ubuntu servers but you're consideringswitching to systemd-resolved, partly for fast failover if yournormal primary DNS server is unavailable and partly because it feels increasingly dangerous not to, sinceresolved is the normal configuration and what software is likelyto expect. One of the ways that resolv.conf is nice is that you canset the configuration by simply copying a single file that isn'tused for anything else
Considering mmap() verus plain reads for my recent code
The other day I wrote about a brute force approach to mappingIPv4 /24 subnets to Autonomous System Numbers (ASNs) , where I built a big, somewhatsparse file of four-byte records, with the record for each /24 ata fixed byte position determined by its first three octets (so0.0.0.0/24's ASN, if any, is at byte 0, 0.0.1.0/24 is at byte 4,and
Early notes on switching some libvirt-based virtual machines to UEFI
I keep around a small collection of virtual machines so I don'thave to drag out one of our spare physical servers to test things on. These virtual machineshave traditionally used traditional MBR-based booting ('BIOS' inlibvirt instead of 'UEFI'), partly because for a long time libvirtdidn't support snapshots of UEFI based virtual machines and snapshots are very important for myuse of these scratch virtual machines . However,I recently discovered that libvirtnow can do snapshots of UEFI based virtual machines
Going from an IPv4 address to an ASN in Python 2 with Unix brute force
For reasons , I've reachedthe point where I would like to be able to map IPv4 addresses intothe organizations responsible for them, which is to say their Autonomous System Number (ASN) ,for use in DWiki , the blog engine of Wandering Thoughts .So today on the Fediverse I mused :
Current status: wondering if I can design an on-disk (read only)data structure of some sort that would allow a Python 2 program toefficiently map an IP address to an ASN. There are
Fedora's virt-manager started using external snapshots for me as of Fedora 41
Today I made an unpleasant discovery about virt-manager on my (still)Fedora 42 machines that I shared on the Fediverse :
This is my face that Fedora virt-manager appears to have beendefaulting to external snapshots for some time and SURPRISE, externalsnapshots can't be reverted by virsh. This is my face, especially asit seems to have completely screwed up even deleting snapshots on somevirtual machines.
(I only discovered this today because today is the first time Itried to touch such a snapshot
Mass production's effects on the cheapest way to get some things
We have a bunch of networksin a number of buildings , and aspart of looking after them, we want to monitor whether or not they'reactually working. For reasons beyond the scope of this entry wedon't do things like collect information from our switches throughSNMP, so our best approach is 'ping something on the network in therelevant location'. This requires something to ping. We want thatthing to be stable and always on the network, which typically rulesout machines and devices run
A traditional path to getting lingering duplicate systems
In yesterday's entry I described a lingering duplicate system andhow it had taken us a long time to get rid of it , but I got too distracted by the storyto write down the general thoughts I had on how this sort of thinghappens and keeps happening (also, the story turned out to be longerthan I expected). We've had other long running duplicate systems,and often they have more or less the same story as yesterday'sdisk space usage tracking system .
The first system
Lingering duplicate systems and the expense of weeding them out (an illustration)
We have been operating afileserver environment for a longtime now, back before we used ZFS . When youoperate fileservers in a traditional general Unix environment , one of the things you need isdisk usage information. So a very long time ago, before I evenarrived, people built a very Unix-y system to do this. Every night,raw usage information was generated for each filesystem (for a whilewith ' du '), written to a special system directory in the filesystem,and then used to