A pleasant surprise with a Thunderbolt 3 10G-T Ethernet adapter

Recently, I tweeted :

I probably shouldn't be surprised that a Thunderbolt 10G-T Ethernetadapter can do real bidirectional 10G on my Fedora laptop (a Dell XPS13), but I'm still pleased.

(I am still sort of living in the USB 2 'if it plugs in, it'sguaranteed to be slow' era.)

There are two parts to my pleasant surprise here. The first partis simply that a Thunderbolt 3 device really did work fast


An unpleasant surprise with part of Apache's AllowOverride directive

Suppose, not entirely hypothetically, that you have a generaldirectory hierarchy for your web server's document root, and youallow users to own and maintain subdirectories in it. In order tobe friendly to users, you configure this hierarchy like the following:

Options SymLinksIfOwnerMatchAllowOverride FileInfo AuthConfig Limit Options Indexes

This allows people to use .htaccess files in their subdirectoriesto do things like disable symlinks or enable automatic directoryindexes (which you have turned off here by default in order to avoidunpleasant accidents, but which is inconvenient


Using grep with /dev/null, an old Unix trick

Every so often I will find myself writing a grep invocation like this:

find .... -exec grep  /dev/null '{}' '+'  

The peculiar presence of /dev/null here is an old Unix trick thatis designed to force grep to always print out file names, evenif your find only matches one file, by always insuring that grephas at least two files as arguments. You can wind up wanting todo the same thing with a direct use of


Thinking about the merits of 'universal' URL structures

I am reasonably fond of my URLs here on Wandering Thoughts (although I've made a mistake or two in their design), but I havepotentially made life more difficult for a future me in how I'vedesigned them. The two difficulties I've given to a future self arethat my URLs are bare pages, without any extension on the end oftheir name, and that displaying some important pages requires aquery parameter.

The former is actually quite common out there on the Internet, asmany


Open protocols can evolve fast if they're willing to break other people

A while back I read an entry from Pete Zaitcev , where he said, amongother things:

I guess what really drives me mad about this is how Eugen [theauthor of Mastodon] uses his mindshare advanage to drive protocolextensions. All of Fediverse implementations generaly communicatefreely with one another, but as Pleroma and Mastodon develop, theygradually leave Gnusocial behind in features. In particular, Eugenfound a loophole in the protocol, which allows to attach pictureswithout using up the space in the message for the URL.


'Scanned' versus 'issued' numbers for ZFS scrubs (and resilvers)

Sufficiently recent versions of ZFS have new ' zpool status ' outputduring scrubs and resilvers. The traditional old output looks like:

scan: scrub in progress since Sat Feb  9 18:30:40 2019      125G scanned out of 1.74T at 1.34G/s, 0h20m to go      0B repaired, 7.02% done

(As you can probably tell from the IO rate, this is a SSD-based


Making more use of keyboard control over window position and size

For a long time, I've done things like move and resize windows in my highly customized X environment through themouse. Using the mouse for this is the historical default in Xwindow managers and X window manager setups, and I have a wholeset of fvwm mouse bindings to make this fastand convenient. I may have used keyboard bindings for things like raising and lowering windows and lockingthe screen , but not beyond that.

(I was familiar with the idea, since keyboard-driven tiling windowsare very


A touchpad is not a mouse, or at least not a good one

One of the things about having a pretty nice work laptop with a screen that's large enoughto have more than one real window at once is that I actually useit, and I use it with multiple windows, and that means that I needto use the mouse. I like computer mice in general so I don't objectto this, but like most modern laptops my Dell XPS 13 doesn't havea mouse, it has a trackpad (or touchpad, take your pick). You


Using a single git repo to compare things between two upstreams

The other day I wrote about hand-building an updated upstreamkernel module . One of the thingsthat I wanted to do in that is to compare the code of the nct6775 module I wanted to build between the 4.20.x branch in the stabletree and the hwmon-next branch in Guenter Roeck's tree .In my entry, I did this by cloning each Git repo separately andthen running diff by hand, but this is a little awkward and Isaid that there


A problem with strict memory overcommit in practice

We've used Linux's strict overcommit mode onour compute servers for years to limit memory usage to the (large)amount of RAM on the machines, on the grounds that compute jobsthat allocate tens or hundreds of GB of RAM generally intend to useit. Recently we had a series of incidents where compute machineswere run out of memory, and unfortunately these incidents illustrateda real problem with relying on strict overcommit handling. You see, in practice, strict overcommit kills random processes . Oh, notliterally