Activity for Kamil Maciorowski‭

Type	On...	Excerpt	Status	Date
Edit	Post #291290	Initial revision	—	about 1 month ago
Answer	—	A: What are the concrete security risks of forcibly terminating a process? > What security risks can be introduced this way, and how? Consider the following scenario: 1. A legitimate server listens on some network port. Usually¹ no other process can listen on the port. 2. The server exits. This makes the port available to other processes for listening. 3. Anothe... (more)	—	about 1 month ago
Comment	Post #290163	@#64656 "accessible in the first place?" – In Bash there is `${PIPESTATUS[@]}`, it is not affected by the `!` the answer is about. You can think of `$?` as derived from `${PIPESTATUS[@]}` according to `pipefail` (`set -o | grep pipefail`) and the presence (or not) of `!`. (more)	—	about 1 month ago
Comment	Post #291149	Why? Because `getent group` with common `nsswitch.conf` uses `/etc/group` as the database. For each user the information about the primary group is in `/etc/passwd` though, not in `/etc/group`. (more)	—	about 1 month ago
Edit	Post #289899	Post edited: removed pointless code	—	7 months ago
Comment	Post #289899	Because I forgot to remove this line. The code you see is not the first version, the answer evolved before being published. In one of my tries the script was supposed to print the content of `$lines` from within some trap. I set the variable to empty before defining this exact trap; the purpose was t... (more)	—	7 months ago
Comment	Post #289899	I wanted the script to start ignoring SIGINT as early as possible. OTOH `trap leave …` before `leave() …` would create a time window when the trap, if triggered, tries to run `leave` anyway and either fails or "succeeds" by running some other `leave`. IMO it's better here to define the trap a moment ... (more)	—	7 months ago
Comment	Post #289899	@Quasímodo The `$(…)` in your code would strip the trailing newlines I want to save. We need to append `x` *before* `$(…)` acts, only then `x` makes the trailing newline characters in fact non-trailing. Appending `x` after `$(…)` acts is too late. (more)	—	7 months ago
Comment	Post #289899	`setterm` and `tput` output escape sequences the terminal is supposed to interpret (see e.g. `tput smcup | xxd`). The sequences are garbage when not interpreted by a terminal. When not interpreted by the terminal, they cannot configure the terminal. By printing to `/dev/tty` (also from `tee`) I make ... (more)	—	7 months ago
Comment	Post #289899	`$(…)` removes *all* trailing newline characters; adding exactly one newline character when doing `printf` may or may not compensate this. By appending `x` and one newline character (this is what `echo x` does), allowing `$(…)` to remove this newline character and removing `x` "by hand" later, I capt... (more)	—	7 months ago
Edit	Post #289899	Initial revision	—	7 months ago
Answer	—	A: How to overwrite each line of STDOUT with the next one? This other answer tries to truncate each line to make it fit the width of the terminal. It's hard to do this reliably because e.g. a tab character counts as one, but it looks like several spaces; on the other hand unprintable characters and escape sequences do not show at all. My answer is a compl... (more)	—	7 months ago
Comment	Post #289875	Another thing: each tab character in the input will count as one, but in general the terminal will move the cursor more and the truncated line will look longer than `$COLUMNS` and still wrap (see `man 1 expand`, this may help). On the other hand there may be unprintable characters that may cause the ... (more)	—	7 months ago
Comment	Post #289875	The answer does not state `$COLUMNS` is not portable. It works in "big" shells (Bash, Zsh) but `sh` is not required to support it. (more)	—	7 months ago
Comment	Post #289794	Yes, [`find-bash` is arbitrary](https://superuser.com/q/1526229/432690). (more)	—	8 months ago
Edit	Post #289794	Post edited:	—	8 months ago
Edit	Post #289794	Post edited:	—	8 months ago
Edit	Post #289794	Initial revision	—	8 months ago
Answer	—	A: Higher-order functions in Bash? 1. Do not embed `{}` in the shell code. If your `{}` gets expanded (by `xargs`) to `./rogue name $(reboot).txt` then `bash -c 'my-func "{}"' ` will become `bash -c 'my-func "./rogue name $(reboot).txt"' ` and `reboot` will be executed. Pathnames containing `"` will also be problematic. The right way ... (more)	—	8 months ago
Edit	Post #289608	Initial revision	—	8 months ago
Answer	—	A: Documentation for double asterisk glob expansion While `` belongs to the POSIX standard, `` is an extension. You are right it's not in `man 7 glob`. Tools that treat `` specially (i.e. not as `` simply followed by ``) tend to agree what it means, but AFAIK it's only because there is a consensus, not because they all use some common library (like `g... (more)	—	8 months ago
Comment	Post #286257	What exactly is wrong with `dd … || the_other_method`? Does it waste time and effort because `dd` fails with I/O error not early enough? Or does `dd` return a different error after actually doing its job with a "clear data" DVD? Or what? (more)	—	9 months ago
Comment	Post #288789	I think it depends on architecture if SIGUSR1 is 10. This is one of the reasons I wrote "I won't elaborate". (more)	—	11 months ago
Edit	Post #288789	Post edited:	—	11 months ago
Edit	Post #288789	Initial revision	—	11 months ago
Answer	—	A: How can I use SIGUSR1 or SIGUSR2 without risk of terminating the process? If you are about to start the tool Start the tool from a process that already blocks or ignores SIGUSR1 and/or SIGUSR2 and let the tool inherit these settings. A blocked signal, when generated, will not be delivered to the process until the process unblocks it. An ignored signal, when delivered... (more)	—	11 months ago
Edit	Post #288788	Initial revision	—	11 months ago
Question	—	How can I use SIGUSR1 or SIGUSR2 without risk of terminating the process? SIGUSR1 and SIGUSR2 are user-defined signals. Imagine there is a tool designed to do something useful upon receiving one or the other. The problem is the default action for these signals is `Term` (see `man 7 signal`), so it's safe to send SIGUSR1 or SIGUSR2 to the tool only after it started handl... (more)	—	11 months ago
Edit	Post #288739	Post edited:	—	11 months ago
Edit	Post #288739	Post edited:	—	11 months ago
Edit	Post #288739	Initial revision	—	11 months ago
Answer	—	A: Reverse shell with named pipe and netcat I'm not sure which exact fragment, functionality or aspect is problematic to you. Here I will make points (paragraphs) about what I used to struggle to understand, or about what I suspect may be not-quite-easy to understand. Some of the below paragraphs are important for understanding later paragr... (more)	—	11 months ago
Comment	Post #288405	What tool do you use to display these names in a way that makes use of the newlines? Or how exactly do you "compare files titles"? Many modern tools sanitize such names when displaying to a human (i.e. in GUI or when printing to a terminal) and it takes some effort to make them stop. Do you use like ... (more)	—	11 months ago
Comment	Post #288401	There is this article [*Fixing Unix/Linux/POSIX Filenames: Control Characters (such as Newline), Leading Dashes, and Other Problems*](https://dwheeler.com/essays/fixing-unix-linux-filenames.html) by David A. Wheeler. Users who find the question interesting will probably find the article interesting ... (more)	—	11 months ago
Edit	Post #288384	Initial revision	—	11 months ago
Answer	—	A: When a command takes filenames as argument, how can I avoid creating temporary files? This other answer uses process substitutions. Not every shell supports this feature. If your OS provides pathnames for file descriptors (`/dev/fd/N` or `/proc/self/fd/N`) then you can use them to achieve the desired result without process substitution and without temporary files: ls /home/alic... (more)	—	11 months ago