00:12:38 is there a man page for the /var/log/messages file? 00:12:59 kerneldove_: what are you hoping to find in said man page? 00:13:24 not sure 00:15:14 in /var/log/messages config, "*.notice;authpriv.none;kern.debug;lpr.info;mail.crit;news.err" will that include /var/log/security content? (security.*) 00:15:19 kerneldove_: well, without any additional info from you, i can only suggest looking at the syslog(3) manpage 00:15:23 i'm guessing not but wanna confirm 00:15:41 and syslog.conf(5) 00:16:30 for the above, it would only include security.notice. nothing else from security.* 00:19:46 security.notice and more severe, right? 00:20:22 nope 00:20:32 *.notice is what would be matched 00:20:34 oh 00:20:44 what are you trying to do? 00:20:50 gimmie somethin' to work with here 00:20:50 :) 00:20:53 zi: wait, really? 00:21:08 was curious if tailing /var/log/messages was enough to stay abreast of security issues reported to the system 00:21:16 "kern.notice" would be kern.notice and anything higher, "*.notice" only matches notice exactly? 00:21:27 since /var/log/messages is kinda the main funnel of all important system logging 00:23:45 kerneldove_: might start out by enabling all.log in /etc/syslogd.conf and then figuring out what you like and go from there 00:24:38 ivy: nah, i misspoke. =notice would be required to match exact 14:48:01 I'm trying to decide if it's worth blowing away my experimentation BSD machine and putting 15-BETA on it as I turn it into a NAS 14:48:24 * CrtxReavr has no time for OS betas. 14:48:39 it's not too far off normal 14-RELEASE really, just a couple tweaks so it knows how to power off properly and my shell is brightly coloured 14:49:09 zip: what did you fix with poweroff? is that an OS bug we should fix? 14:49:21 hang on let me check 14:49:40 I have to WOL that sucker and wait for it to boot 14:51:06 hw.efi.poweroff=0 14:51:23 hey is there a way to replace jpeg-turbo with mozjpeg without removing and reinstalling 278 packages? 14:51:28 according to the comment, this is because it'll reboot instead of powering off 14:54:56 ivy: did you see my PM? 14:56:08 black2: yes, but unless this issue is private somehow, it's probably better to ask in the channel. otherwise just ask the question, don't ask to ask :-) 14:56:17 zip: ah okay, i guess that's some hw/fw-specific issue 14:57:04 ivy: okay I will deal with it very soon 15:03:33 ivy: okay but I strongly recommend doing that in PM as I have told plenty of people about my idea but they don't seem to understand what I am saying. 15:03:40 are you sure you want to do it here? 15:06:48 black2: i... have no idea? tbh if no one understands your idea, you might try describing it differently, but if you really want to PM that's fine 15:07:28 okay then let's do PM. I hope you understand. 15:22:22 Just wondering, is it possible to do CUDA programming on FreeBSD natively, or does that require Linuxulator? 15:23:12 <[tj]> it requires linuxulator 15:23:44 "Nvidia, fuck you!" ~ Linus Torvalds 16:39:51 is there a syntax checker for the kernel config file? i cannot get around syntax error for a config change. i do not know where is the issue. it has no obvious syntax errors 16:40:19 config: /home/oleh/kernel/CHAKLUNCHIK:416: syntax error 16:41:16 https://termbin.com/dhu5 16:41:29 i see no errors on line 416 16:42:24 angry_vincent: which line is 416? 16:43:01 device hwpmc 16:43:13 the last line of config 16:44:40 angry_vincent: you're missing a newline at the end of the file. config(8) should probably produce a better error there 16:48:52 thx. this fixed. though it looks really weird to me why would it ever need a new line 16:49:43 it is standard for all lines in a text file to end with a newline. most editors do this by default, although some do not 16:50:47 well, then emacs does not do it by default 16:53:39 there is apparently an option for this: https://www.gnu.org/software/emacs/manual/html_node/emacs/Customize-Save.html 17:01:09 i added this. thx 17:10:10 May times software will consider a file to be corrupt if it doesn't end with a newline. 17:10:15 Many times 17:20:18 such is life with lex/yacc 17:22:59 now i know. but, indeed, message could be more helpful instead of syntax error 17:23:24 i put that issue on my todo list, unless anyone else gets around to it first 17:23:54 OS/2 v2.5 (I think.) has this MS-CardFile alike program. . . 17:23:57 can it be solved 17:24:20 ivy: thx 17:24:30 It had bugs where it would sometimes neglect that trailing newline, and getermine the file was corrrupt the next time you tried to open it. 17:24:35 Was a bloody pain. 17:30:29 On the pedantic level a text file is defined as a file of lines terminated by newlines. If a file contains a line NOT terminated by a newline such as at the end of file then it is not a text file but a binary file. 17:31:29 That's why sometimes we see config files with a "# Do not remove this comment." line at the end of the file. Because then if /that/ line is missing a newline it was in a comment and probably ignored without any errors being introduced. Those annoy me though because I know how to ensure that the file is a text file and ends with a newline. 17:31:57 In the original days the editors were ed/vi and those are line editors and impossible to save a file without a newline at the end of every line. 17:33:05 Emacs may be a well known Unix-land offender but Microsoft editors are now used by a lot of people working on MS-Windows and then either remotely editing or copying the file and that's where I mostly run into files not terminated by newlines now. 18:54:55 what. 18:57:58 that's... one definition of a text file I suppose 18:58:04 I've never really seen a formal one 19:05:42 posix has a formal definition that, expanded out and reading a little between the lines, is exactly that 19:06:09 it's a series of zero or more lines, nad a line is a sequence of zero or more non- characters plus a terminating character 19:10:24 https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap03.html#tag_03_403 a text file contains characters organized into zero or more lines. 19:10:48 https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap03.html#tag_03_206 a line is a sequence of zero or more non-newline characters plus a terminating newline character. 19:22:08 Not sur I agree with that. 19:22:14 printf "foo\n\nbar\n" 19:22:30 Is the "blank" row not a line? 19:23:06 CrtxReavr: the blank line is "zero or more non-newline characters plus a terminating newline character", so this does not conflict with the definition 19:23:45 Yeah. . . I guess I read the "zero or more" part a bit too fast. 19:23:58 ivy: bro I've been waiting... you wanna do it now? 19:24:24 black2: waiting for what? you said you would send a PM then you never did 19:33:38 posix notably has definitions for both a blank line and an empty line 19:34:35 a partial line at the end does not a binary file make, it just an incomplete line makes, practically speaking 19:35:15 yeah, it's just not a valid text file 19:36:24 so i told yasu@ on pkgbase that freebsd-update does what they want, which is true, but i think we're still updating packages that don't need to be updated for some reason 19:36:32 er, wrong channel 20:09:42 ahaa :) 20:47:12 cperciva: i tested .MAKEFLAGS: -W and it seems to fix this issue: https://www.le-fay.org/tmp/7d/ZkSMsCRwGh 20:47:20 wrong channel again, ffs 21:21:35 where's the right one 21:27:35 It left. 21:29:31 ketas: a secret channel for cool people 21:35:33 like freemasons 21:38:21 freebsdmasons* 21:41:18 yes 21:41:32 freebsdsons 21:56:04 epic5-coded 22:09:58 LOL 22:10:22 do you have decoder rings? 22:10:30 someone say freemasons? https://freemasonsfordummies.blogspot.com/ 22:11:43 * Macer checks the significance of the US$1 bill