ANSI colours aren't consistent across X terminal programs
There is a long standing set of ' ANSI colour codes ' in terminalemulators, including terminal programs for X. Here is a table ofthem ,and fidian/ansi will provide youwith a convenient Bash script that will show you what these colorslook like in your terminal program. The latter is potentiallyrelevant because, shockingly, no two X terminal programs I've triedrender these ANSI colours exactly the same (between xterm, urxvt,Gnome Terminal, and konsole; xfce4-terminal may render the same
Special tcpdump filtering options for OpenBSD's pflog interface
One of the convenient things that OpenBSD's pf packet filteringsystem can do is log packets ofinterest, as covered in the Packet Filtering section of the pf.confmanual page .These packets are logged to a special pflog network interface (where adaemon will generally write themto disk). Since this is a network interface, you can monitor trafficon it with OpenBSD's version of tcpdump (or use tcpdump to read thelog file).
As part of this, the OpenBSD tcpdump has some special
An unexciting idea: Code changes have context
I recently read Mark Dominus's I wish people wouldstop insisting that Git branches are nothing butrefs ( via ).One of my thoughts afterward is that this feels like an instance ofa broader thing, which is that (code) changes have context; here,one part of that context is where they happen (ie, what branch theyhappen on). Of course we already know that in a sense, because Git(and pretty much every other version control system) considers itimportant to record both
How secure is merely discarding (TRIMing) all of a SSD's blocks?
Suppose hypothetically that you have some SSDs to securely disposeof, and that for one reason or another you can't use built-in SSDsecure erase on them, for example (apparently) because your BIOS automaticallylocks out that option when it boots. You might wonder how wellprotected you are if you simply tell the SSD to discard all ofits data . Unsurprisingly, the answer isthat it depends.
First off, any SSD you want to use today will support what's called'Deterministic Read After
When securely erasing disks, who are you trying to stop?
Various people, us included,periodically have the need to securely dispose of disk drives thatwe no longer need or want, where by 'securely' we mean that peopleshouldn't be able to get our data from the drives after we've gottenrid of them. Often there are questions of what you need or wantto be doing in order to achieve this security. In my view, part ofthe answer to this is depends on who you want to stop from gettingyour data (and how many
Modern email addresses can be in UTF-8
Over on the Fediverse, I noted :
It has been '0' days since someone's email client helpfully let themuse a Unicode '‐' instead of an ASCII '-' in dash-separated emailaddresses. Or perhaps the client automatically used the Unicodecharacter instead of the ASCII dash.
You may not be surprised to hear that email systems, ours included,don't consider the two to be the same. I'm not sure how it even works,although some sending MTAs appear to
A gotcha with Systemd's DynamicUser, supplementary groups, and NFS (v3)
Today I used bind mounts in an odd way :
Today's crazy Linux bind mount usage:
# cp [-a] /a/nfs/mount/special-file /root/special-file
# mount --bind /root/special-file /a/nfs/mount/special-fileThis was the easiest way to make a systemd service withDynamicUser=yes and a supplementary group get access to special-file,which is only accessible by said group. (The normal version of
Ubuntu is a Canonical product
A while back I wrote that from an outside perspective, Ubuntu isCanonical's thing , in that Canonical runs theshow despite having outside contributors. But in the wake of wrestling with Canonical's advertisements in a stock 22.04 LTSmachine and losing ,I want to amend that observation with an important additional one.Ubuntu is not merely Canonical's, Ubuntu is a Canonical product.Which is to say, Ubuntu exists to make money for Canonical .Further, the current evidence suggests that Canonical
Future Internet PKI schemes need to be bootstrapped through web PKI
On the Fediverse, I said
Thesis: any realistic, viable Internet PKI scheme inthe moderate future will have to bootstrap from web PKI, because webPKI is where the usage is to drive people to address ('solve' in anon-mathematical sense) the hard problems. You don't literally have touse HTTPS, but you need public TLS.
Counterpoint: end to end encrypted messaging. But I think that broadlythat has an 'introduction' (identity) problem.
(Brought on by thinking
Universities are often environments with distributed accounts and identities
Over on the Fediverse, I said something about university accountsystems :
Welcome to the university. We have a central account andidentification system, of course. Does everyone use the centralaccount system? Of course not. Does everyone know what central accountIDs their local accounts have? I see you're new here. Is there alwaysa one to one mapping between local and central accounts? Not a chance.
Universities are fun places.
There are all sorts of reasons for this distributed and anarchicenvironment. An obvious