00:07:17 <polyex27> cant believe im actually fixing stuff and not just begging others for help
00:07:22 <polyex27> it's like a whole new world
00:09:17 <koobs> asking for help is a sign of strength
00:10:29 <mason> It's being fully human, as we are communal beings.
00:15:11 <polyex27> ya but i gotta be able to admin a box myself and till now i needed mostly begging to do it
00:24:28 <polyex27> so i got caddy running again and it has http/3 quic now. starts up fine, runs fine. but i see in caddy log message that failed to increase receive buffer size
00:24:41 <polyex27> so https://github.com/lucas-clemente/quic-go/wiki/UDP-Receive-Buffer-Size#bsd says to increase it. that advice safe?
00:24:42 <VimDiesel> Title: UDP Receive Buffer Size · lucas-clemente/quic-go Wiki · GitHub
00:26:43 <polyex27> i dont trust non-bsd ppl
00:34:22 <V_PauAmma_V> polyex27, I've been here (since I returned) 3.5 years, and I happily do both - sometimes alternating, sometimes together, as in "help! I started fixing this document and I don't understand what it should say!"
00:36:55 <polyex27> well im sure ill still ask for help, like i am now with the maxsockbuf advice. it's just amazing to actually be fixing shit on my own box myself. never thought i'd see the day
01:06:42 <polyex27> i just doubled the default maxsockbuf
01:23:30 <gh00p> Can I get a hand making fusefs go? Or if someone else can tell me a good way to read an exfat filesystem....
01:24:15 <gh00p> I haven't used fuse, and I'm not sure about .. the daemon, where it is or how to start it.. or how to tell fuse to use exfat, even?
01:32:21 <gh00p> So, I bought this 16TB m.2 SSD which I do not trust. `file -s /dev/da1p2` tells me it is. a `DOS/MBR boot sector".  I can `od -c /dev/da1p2` and I see the word "EXFAT"  in the output. Am I on the right track?
01:56:32 <ebonheart> gh00p: How much did you pay for the ssd? Too good to be true? I would expect a disk like that to show up as an NVMe...
01:57:51 <ebonheart> gh00p: and if you bought it, aren't you going to wipe it and put your own FS on there?
06:22:04 <debdrup> koobs: I don't disagree at all, but it's also a proper _good_ feeling when you do figure something out on your own
06:22:51 <debdrup> gh00p: I can almost guarentee you got scammed.
06:24:16 <debdrup> There's no way 16TB fits on any M.2 form factor; the 16TB SSDs that you can buy take up either 2.5" via U.2 or the "Ruler" form factor that Intel is pushing.
06:24:41 <debdrup> Actually, I think the 16TB SSD I saw was 3.5"
06:33:24 <grahamperrin> <debdrup> "Graham Perrin: the project isn't..." <- The conversation began in a FreeBSD area where GitHub is primary.
06:43:07 <meena> polyex27: why is caddy opening a UDP port??
06:44:32 <meena> oh, probably because QUIC…
06:52:55 <nacelle> debdrup: they make legit 2.5" 16TB ssd
06:53:17 <nacelle> but tehy're also thousands of bucks new
06:53:46 <nacelle> heh, what do you know
06:54:57 <nacelle> you can get a 16TB m.2 nvme
06:55:01 <nacelle> https://semiconductor.samsung.com/resources/data-sheet/Product_Brief_Samsung_PM983_NF1_NVMe_SSD_1806.pdf
06:55:23 <nacelle> but its not exactly 2280...
06:55:53 <nacelle> oh pardon, thats ngsff being compared to m2, <-- learning
06:56:14 <nacelle> Measuring only 30.5mm x 110mm x 4.38mm, the NGSFF
06:56:14 <nacelle> form factor offers three times the capacity of a 2.5 inch while maintaining
06:56:15 <nacelle> serviceability in an M.2 size.
06:56:44 <nacelle> (I do imagine you're right about it being a scam fwiw)
07:25:18 <Plasmoduck> how can I change my pkg server?
07:41:45 <lavaball> plasma, it's in /etc/installurl.
07:42:18 <lavaball> oh, i meant the other guy.
07:42:19 <lavaball> my bad.
07:55:51 <plasma> nvm ;)
08:06:35 <grahamperrin> V_PauAmma_V: now I realise, the example screenshot in GitHub-provided documentation is misleading (as a result of dimming parts of the shot).
08:06:36 <lavaball> oh, also i'm in the wrong channel. so i'm guessing you also have /etc/installurl?
08:08:51 * grahamperrin uploaded an image: (79KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/GMybxoPkiCQrUeTiIftnpWyk/image.png >
08:08:52 <grahamperrin> Now I see, edits from maintainers allowed by default, and not only with drafts:
09:38:37 <meena> anyone here using doas to give out privileges to accounts, rather than sudo?
09:47:58 <V_PauAmma_V> No. I didn't even know that was a thing.
09:51:13 <meka> Well, it almost isn't a thing. I don't know what, but something from OpenBSD is missing in FreeBSD which makes doas work nice. I think it's about remembering you for a while like sudo does, so you don't have to type password every single time
09:51:37 <meka> But take it with a huge grain of salt, as I didn't do the research
10:21:44 <yuripv> i used it previously until they decided to make it "secure" and it stopped doing something sudo can (i don't remember what exactly, didn't look back)
10:30:09 <steew> As far as I know, it's just that the OpenDOAS port has the persistence flag disabled by default on compilation, but it can be built with it and added to doas.conf as `permit persist :wheel`
10:30:45 <steew> They mention it follows the same method sudo does of relying on timestamps, and supposedly OpenBSD has a kernel API for this kind of auth
10:31:08 <steew> but yeah, it can be made to persist though not (supposedly) as securely as OpenBSD
11:27:52 <f451> is there a known problem with pkg right now on arm64 -current ?
13:28:46 <meena> f451: much easier to tell us what you're experiencing
13:29:19 * grahamperrin toys with alternatives to a2x(1) …
13:29:25 <grahamperrin> w3m -dump -cols 2147483647 https://www.freebsd.org/status/report-2022-01-2022-03/ > /tmp/test.txt && less /tmp/test.txt
13:29:26 <VimDiesel> Title: FreeBSD Quarterly Status Report First Quarter 2022 | The FreeBSD Project
13:30:57 <f451> meena: lol, point taken ;)
13:31:30 <meena> steew: FreeBSD doesn't package OpenDoas: it packages https://github.com/slicer69/doas/
13:31:32 <VimDiesel> Title: GitHub - slicer69/doas: A port of OpenBSD's doas which runs on FreeBSD, Linux, NetBSD, and illumos
13:31:33 <f451> in fairness, there was a ton of output
13:32:06 * meena can't read a ton of output either… 
13:32:21 <f451> ok in summary
13:32:25 <f451> The package management tool is not yet installed on your system.
13:32:48 <f451> Bootstrapping pkg from pkg+http://pkg.FreeBSD.org/FreeBSD:14:aarch64/latest, please wait...
13:33:06 <f451> Verifying signature with trusted certificate pkg.freebsd.org.2013102301... done
13:33:23 <f451> Child process pid=97234 terminated abnormally: Segmentation fault
13:34:01 <f451> it's "not installed" rn because i tried to fix the issue by deinstalling and reinstalling it
13:35:06 <f451> unfortunately gdb isn't installed either
13:36:56 <meena> f451: so, summary-summary: "pkg on CURRENT / aarch64 is segfaulting"
13:37:05 <f451> yes
13:40:26 <f451> maybe it's cos ive got retpoline enabled
13:41:12 <f451> make: "/usr/src/share/mk/bsd.prog.mk" line 68: warning: Retpoline requested but not supported by compiler or linker
13:42:24 <f451> i thought retpoline was all good?
13:42:28 <f451> maybe not
13:44:36 <f451> nah, same error
17:18:21 <SpaceBass_laptop> I've got a port I just upgraded... did a make reinstall, says it is installing v 1.59 but whenever I check the version of the bin it shows an older version. I did a find to see if there were two bins and there are not
17:18:28 <SpaceBass_laptop> Any ideas how to resolve that?
17:20:44 <SpaceBass_laptop> I guess I can copy from /u/s/p/net/rclone/work/bin/rclone to /usr/bin
17:22:29 <gh00p> Hi. In rsync, is there a way to preserve permissions *except* group? I am trying to back up a zfs volume to exFAT. I don't care about keeping the group, but I would like to be free of the error messages.
17:24:20 <meena> gh00p:         -a, --archive               archive mode; equals -rlptgoD (no -H,-A,-X)
17:24:35 <meena> gh00p: so if you expand that yourself, and drop the -g, you'll get what you want
17:27:41 <gh00p> meena, ah, tysm. I guess I have more options to remove to make this happy with exFAT though. :/
17:28:27 <gh00p> exFAT sucks. Is it likely that I can get away with formatting a 16TB SSD with ufs instead?
17:29:22 <rwp> gh00p, Since FAT stores only to the 2-second level you will also need --modify-window=1 too.
17:30:10 <gh00p> rwp, oh, that makes sense. Thank you, I would not have figured that error out.
17:30:38 <grahamperrin> SpaceBass_laptop: what's the port?
17:30:46 <grahamperrin> rclone?
17:32:23 <rwp> gh00p, It doesn't actually produce an error message. It just can never make the target "up to date". Because the timestamp on odd second files can't be equal.
17:32:47 <rwp> And therefore rsync will always try to update the file thinking that the timestamp does not match.
17:33:00 <gh00p> Yes, errors without error messages are some of the worst. :)
17:34:04 <rwp> My digital audio player uses FAT and I run into that issue rsync'ing files there.
17:37:38 <gh00p> But FAT is not exFAT. Does exFAT have the same time limitation?
18:06:48 <rwp> gh00p, I guess not! https://en.wikipedia.org/wiki/ExFAT "Timestamp granularity of 10 ms for creation and modified times (down from 2 s of FAT ..."
18:06:49 <VimDiesel> Title: exFAT - Wikipedia
19:31:23 <last1> how can I sort files found by find ?
19:31:32 <last1> I tried find ./ -type f -print0 | sort -z
19:31:45 <last1> but it prints them all in one line, whereas I need them one per line
19:34:53 <dodo> last1: remove -print0
19:36:29 <last1> genius - thank you! so find already outputs the null char ?
19:37:07 <dodo> if you tell it to
21:28:22 <rwp> last1, the sort -z reads zero terminated strings on input and *writes zero terminated strings on output*.
21:28:41 <rwp> That's why they appeared to you as being all on one line.  They were zero terminated not newline terminated.
21:33:34 <grahamperrin> last1: I guess, you're also aware of option -s in find(1)?
21:33:46 <grahamperrin> find -s ./ -type f -print
21:34:02 <grahamperrin> Results may differ, as mentioned in the manual page
21:35:14 <grahamperrin> Alternatively, with sysutils/bfs:
21:35:17 <grahamperrin> bfs -s ./ -type f -print
21:45:23 <rwp> If it is me and I know there are no whitespace problems with files then I just use traditional newline terminated strings.  Like: find . -type f -print | sort
21:46:01 <rwp> Because it seems silly to use zstrings and then convert at the end. Ex: find . -type f -print0 | sort -z | xargs -0rL1
22:04:34 <nacelle> do you seriously need to use -print still?
22:04:43 <nacelle> even openbsd's find dropped that years ago
22:09:42 <meena> nacelle: that's for print0
22:11:32 <meena> (and some people never get used to not using certain parameters…)
22:12:02 <nacelle> I see a raw -print on the line above it :-)
22:12:44 <nacelle> (the only reason I got better at unix commands is after peers poked fun at me for using arcane syntax, maybe this helps, maybe its doesnt)