00:25.59 | *** join/#devuan qin (~qin@unaffiliated/soalokin) |
00:27.52 | *** join/#devuan Xenguy (~Xenguy@devuan/community/Xenguy) |
00:38.19 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
00:41.34 | Xenguy | fsmithred: That 'pstree' utility is a nice, easy-to-parse visualization of PC processes. Surprised I didn't run across that sooner. |
00:42.17 | Xenguy | Coincidentally I re-discovered 'tree' recently too |
00:46.22 | *** join/#devuan qin (~qin@unaffiliated/soalokin) |
01:04.28 | fsmithred | Xenguy, you know pgrep? |
01:06.32 | *** join/#devuan arnoldoree (~arnoldore@113.210.63.210) |
01:08.45 | *** join/#devuan unixman_home (~unixman2@unaffiliated/eracc) |
01:09.02 | *** join/#devuan qin (~qin@unaffiliated/soalokin) |
01:11.31 | Xenguy | fsmithred: yes, pgrep and pkill come together I think |
01:12.43 | Xenguy | Very useful, and frequently used. Used to use 'ps aux |grep foo, or killall was popular too, before I came across pkill |
01:13.23 | Xenguy | Katolaz made me aware of pidof too |
01:13.37 | Xenguy | (through his conference presentation |
01:13.41 | Xenguy | ) |
01:13.43 | fsmithred | was trying to think of that one |
01:14.45 | Xenguy | minimalism was the theme of his talk, and I really enjoyed watching that (I liked the chocolates he was throwing too) |
01:15.01 | *** join/#devuan MasterdonX (~masterdon@45.9.249.244) |
01:16.01 | *** join/#devuan Volk (~Volk@a95-94-47-32.cpe.netcabo.pt) |
01:18.23 | *** join/#devuan unixman_home (~unixman2@223-179-152-216.mtaonline.net) |
01:18.23 | *** join/#devuan unixman_home (~unixman2@unaffiliated/eracc) |
01:30.50 | *** join/#devuan jotaxpe (jotaxpe@gateway/vpn/protonvpn/jotaxpe) |
01:41.39 | *** join/#devuan scalar (~scalar@140.82.138.217.360broadband.net) |
01:45.57 | *** join/#devuan ac_laptop (~ac_laptop@186.2.247.129) |
01:53.27 | *** join/#devuan krkini (~kini@unaffiliated/kini) |
01:59.23 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
02:18.37 | *** join/#devuan jotaxpe_ (jotaxpe@gateway/vpn/protonvpn/jotaxpe) |
02:42.24 | *** join/#devuan jotaxpe_ (~jotaxpe@201.219.239.192) |
02:43.52 | *** join/#devuan CAPTCHA_REQUIRED (44f2e88408@nat4.nuegia.net) |
02:44.19 | CAPTCHA_REQUIRED | Hi |
02:44.52 | CAPTCHA_REQUIRED | I'm getting quark: pthread_create: Resource temporarily unavailable when my system has more than enough resources available |
02:44.57 | CAPTCHA_REQUIRED | Why else could this happen? |
02:45.26 | CAPTCHA_REQUIRED | Here is the relevant syscalls |
02:45.27 | CAPTCHA_REQUIRED | clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f379e668a10) = 18629 |
02:45.27 | CAPTCHA_REQUIRED | wait4(-1, quark: pthread_create: Resource temporarily unavailable |
02:45.27 | CAPTCHA_REQUIRED | [{WIFEXITED(s) && WEXITSTATUS(s) == 1}], 0, NULL) = 18629 |
02:49.22 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
02:54.40 | gnarface | CAPTCHA_REQUIRED: probably some ulimit thing. check documentation on ulimit and /etc/security/limits.conf |
02:56.33 | CAPTCHA_REQUIRED | Im using stock limits |
02:56.58 | CAPTCHA_REQUIRED | However, I am using a kernel control group limit of 2G |
02:57.59 | CAPTCHA_REQUIRED | I've still got 1.7Gi available within this namespace only only 19 pids running inside of it, So I don't think it could be a memory thing |
02:58.07 | CAPTCHA_REQUIRED | Also, I'm only trying to create 4 threads |
02:58.16 | gnarface | how many files are you trying to open at once? |
02:58.30 | CAPTCHA_REQUIRED | Hmm |
02:58.39 | CAPTCHA_REQUIRED | Is there a way to tell that gnarface? |
02:58.50 | gnarface | this isn't something you wrote yourself? |
02:59.15 | CAPTCHA_REQUIRED | No |
02:59.16 | CAPTCHA_REQUIRED | Well |
02:59.19 | CAPTCHA_REQUIRED | Not mostly |
02:59.45 | CAPTCHA_REQUIRED | It's a kernel cgroup namespace of prosody, tor daemon (relay+directory), and quark |
03:00.44 | CAPTCHA_REQUIRED | Hmm |
03:00.51 | gnarface | what is the stock # of available file handles now? 4096 still? because that's not even enough to make wine or steam happy. |
03:01.17 | CAPTCHA_REQUIRED | file locks (-x) unlimited |
03:01.38 | gnarface | you can use lsof to check the open files count |
03:02.03 | gnarface | lsof | wc -l |
03:02.05 | gnarface | something like that |
03:02.12 | gnarface | or: lsof -p [pid] |
03:02.15 | CAPTCHA_REQUIRED | Yeah |
03:02.16 | gnarface | i think |
03:02.30 | CAPTCHA_REQUIRED | Oh looks like it's taking a while to populate |
03:02.34 | CAPTCHA_REQUIRED | Hey |
03:02.43 | CAPTCHA_REQUIRED | I do have a LOT of TCP sessions open |
03:02.59 | CAPTCHA_REQUIRED | For all the XMPP s2s connections and tor relays |
03:03.00 | gnarface | there are limits on tcp connection counts and amount of ram used by them |
03:03.13 | gnarface | check also /etc/sysctl.conf for that stuff |
03:03.29 | CAPTCHA_REQUIRED | Give me a sec it's trying to do rdns |
03:03.49 | CAPTCHA_REQUIRED | 1495 |
03:04.13 | CAPTCHA_REQUIRED | Oh wait |
03:04.17 | CAPTCHA_REQUIRED | Outside of the namespace |
03:04.44 | gnarface | hmm, i dunno but the kernel has some defaults to prevent processes from going off the rails and taking down the system, but sometimes if you have very high-end hardware they can strangle your throughput unless you adjust them a bit |
03:04.46 | CAPTCHA_REQUIRED | The lsof command failed printing lsof: no pwd entry for UID 1000 |
03:04.46 | CAPTCHA_REQUIRED | and lsof: no pwd entry for UID 999 |
03:04.46 | CAPTCHA_REQUIRED | over and over again |
03:04.56 | CAPTCHA_REQUIRED | Then returned 31219 |
03:05.08 | gnarface | and if you're using a virtualization container of some sort then that can cause extra complications between the guest and host's limits |
03:05.34 | CAPTCHA_REQUIRED | I'm using lxc |
03:05.45 | gnarface | i could almost have guessed that |
03:06.15 | gnarface | note that there are both hard and soft limits with different defaults |
03:07.43 | CAPTCHA_REQUIRED | Just to be clear gnarface you don't think it's a thread limit, but somehow my system has run out of file descriptors |
03:08.04 | CAPTCHA_REQUIRED | And I need to find out what is taking all my descriptors, or increase the limit |
03:08.29 | gnarface | CAPTCHA_REQUIRED: well there's a bunch of limits and i'm just guessing it might be one of them, not necessarily the file descriptor one, but that one is the one i ran into troubles with the most, so it was my first guess |
03:09.07 | gnarface | note that if you're using lxc, there might be situations where you have to adjust the limits for both the host and the guest |
03:09.36 | gnarface | i'm not very familiar with lxc, but it's unlikely the host for a proper virtualization solution would allow it's guest's limits to exceed it's own |
03:10.33 | *** join/#devuan debdog (~debdog@194.140.112.222) |
03:10.40 | CAPTCHA_REQUIRED | If i was exceeding a limit would it show up in a log somewhere? |
03:11.08 | gnarface | probably, but the error you pasted was sufficient for google to corroborate my hypothesis |
03:11.34 | gnarface | you can also manually check each limit with the ulimit tool |
03:12.03 | gnarface | (try it and note that you'll get different responses depending on whether you're in our out of the guest, whether you're root or not, and whether you're ssh'd or on a physically local terminal or not) |
03:12.22 | gnarface | (and don't forget to check both hard and soft limits) |
03:13.46 | CAPTCHA_REQUIRED | » ~# lsof 2>/dev/null | wc -l |
03:13.46 | CAPTCHA_REQUIRED | » 31188 |
03:14.14 | CAPTCHA_REQUIRED | This is from the host |
03:14.31 | CAPTCHA_REQUIRED | I had to redirect stderr because the uids were subuids |
03:14.36 | gnarface | you gotta constrain that by individual process |
03:14.42 | CAPTCHA_REQUIRED | Hmm |
03:15.25 | CAPTCHA_REQUIRED | fs.file-max = 401614 |
03:15.28 | gnarface | try: lsof -p [pid of guest] |
03:15.58 | CAPTCHA_REQUIRED | I don't think it works like that gnarf |
03:16.07 | fluffywolf | trying new versions of things after the beowulf upgrade... gimp's tool icons seem to be broken. they're light grey shapes that are essentially invisible, and are the same no matter which of the four themes I pick. |
03:16.11 | CAPTCHA_REQUIRED | The entire ptree is exposed to the hos |
03:16.12 | CAPTCHA_REQUIRED | T |
03:16.19 | CAPTCHA_REQUIRED | Unless you mean pid of guest's init |
03:16.46 | fluffywolf | nevermind. changing icon themes fixed it. apparantly the default is "invisible". |
03:16.47 | fluffywolf | ... |
03:18.23 | CAPTCHA_REQUIRED | There's only one thing that concerns me lxc.cgroup.memory.limit_in_bytes = 2G |
03:19.01 | CAPTCHA_REQUIRED | Memory use: 820.73 MiB |
03:19.06 | CAPTCHA_REQUIRED | KMem use: 28.78 MiB |
03:19.11 | gnarface | CAPTCHA_REQUIRED: sorry i don't know enough about lxc |
03:19.26 | CAPTCHA_REQUIRED | No worries |
03:19.49 | gnarface | i'm relatively certain i've put you on the right path, but closing the loop is gonna require reading some documentation about the matter |
03:21.11 | CAPTCHA_REQUIRED | Yeah |
03:21.13 | CAPTCHA_REQUIRED | Well |
03:21.20 | CAPTCHA_REQUIRED | I get the error on the host too |
03:21.20 | CAPTCHA_REQUIRED | p# /var/tmp/quark -p 80 -h :: -u www-data -g www-data -d /usr/share/vim |
03:21.20 | CAPTCHA_REQUIRED | /var/tmp/quark: pthread_create: Resource temporarily unavailable |
03:21.29 | gnarface | fluffywolf: maybe related, maybe not but a long time before beowulf i ran into a problem with the default icon themes not having icons which manifested in that basic way. i installed tango-icon-theme and put gtk-icon-theme-name="Tango" in my ~/.gtkrc-2.0.mine |
03:21.36 | CAPTCHA_REQUIRED | Maybe it's a problem with the code |
03:22.40 | gnarface | i would look for it opening stuff in a loop but not closing it after |
03:22.52 | gnarface | but if this is online, this can also be caused by incoming network connections |
03:23.18 | CAPTCHA_REQUIRED | Yeah it was a problem with the code |
03:23.24 | CAPTCHA_REQUIRED | I went back to an earlier revision |
03:23.35 | CAPTCHA_REQUIRED | Before threading was implemented |
03:24.18 | gnarface | huh, is it perl? |
03:24.29 | CAPTCHA_REQUIRED | » <gnarface> but if this is online, this can also be caused by incoming network connections |
03:24.30 | CAPTCHA_REQUIRED | can you elaborate? It's been a while since I had to tune a linux kernel for lots of long-sustained TCP sessions |
03:24.35 | CAPTCHA_REQUIRED | gnarface: no it's just C |
03:24.56 | gnarface | well, like i said, there's other limits, not just on number of open files |
03:25.02 | CAPTCHA_REQUIRED | gnarface: here is the source https://git.suckless.org/quark/ |
03:25.04 | gnarface | some of them are in the kernel sysctl.conf |
03:25.38 | gnarface | now it is a known type of attack for public facing servers to be DOS'd by simply exhausting the available network connections |
03:25.53 | gnarface | clients will just find some service that responds by opening a connection and waits politely to close, then they just don't close it |
03:26.00 | gnarface | then they open more |
03:26.01 | gnarface | in a loop |
03:26.48 | gnarface | if they don't care about a valid response then it doesn't matter that the connection hangs |
03:27.01 | gnarface | all they care about is using up the available connections faster than the timeout clears them |
03:27.28 | gnarface | it's usually easier to deal with these vandals at the firewall level though |
03:36.28 | CAPTCHA_REQUIRED | Yeah i've dealt with those in the past |
03:37.07 | CAPTCHA_REQUIRED | Usually they aren't very smart about it and do the synflood from a singal ip range |
03:37.23 | CAPTCHA_REQUIRED | Which is easier enough to filter out with a bucket |
03:48.01 | *** join/#devuan kini (~kini@unaffiliated/kini) |
04:11.46 | *** join/#devuan captainfixerpc14 (~captain@32.211.181.220) |
04:40.12 | *** join/#devuan ar3itrary (~hacker@2a03:4000:6:8177:2::1) |
05:03.12 | *** join/#devuan KREYREN (~kreyren@gateway/tor-sasl/kreyren) |
05:23.35 | *** join/#devuan bpmedley (~bpm@69-174-155-24.bltnilaa.metronetinc.net) |
05:25.23 | *** join/#devuan unixman (~aunixman@209.165.134.7) |
05:25.24 | *** join/#devuan unixman (~aunixman@unaffiliated/eracc) |
05:54.33 | *** join/#devuan DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
06:09.22 | *** join/#devuan kini (~kini@unaffiliated/kini) |
06:11.10 | *** join/#devuan KnoP (~KnoP@87.190.255.4) |
06:11.41 | *** join/#devuan KnoP (~KnoP@business-176-095-149-105.static.arcor-ip.net) |
06:11.55 | *** join/#devuan chomwitt (~quassel@2a02:587:dc2f:b800:94ec:6dc2:1365:382e) |
06:26.13 | *** join/#devuan Diagon (DiagonalAr@gateway/vpn/protonvpn/diagonalarg) |
06:30.06 | *** join/#devuan jotaxpe (jotaxpe@gateway/vpn/protonvpn/jotaxpe) |
06:34.16 | *** join/#devuan Jjp137 (~Jjp137@2603-8000-2a01-3406-92fb-a6ff-fee2-6861.res6.spectrum.com) |
07:35.09 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
07:37.24 | *** join/#devuan Joril (~joril@217.194.188.145) |
07:37.27 | *** join/#devuan Inepu (~Mithrandi@137-217-234-46.wifi4all.it) |
07:57.30 | *** join/#devuan arnoldoree (~arnoldore@113.210.123.25) |
08:03.27 | *** join/#devuan ar3itrary (~hacker@2a03:4000:6:8177:2::1) |
08:26.21 | *** join/#devuan mro_name (~mro_name@2002:5cc3:89f8:0:889c:3c42:21ea:ec70) |
08:50.48 | *** join/#devuan Pali (~pali@Maemo/community/contributor/Pali) |
08:57.54 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
09:25.30 | *** join/#devuan mro_name (~mro_name@2002:5cc3:89f8:0:2d7a:adbf:c007:d96) |
09:28.15 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
09:30.54 | *** join/#devuan GoatAvenger (~GoatAveng@unaffiliated/goatavenger) |
09:45.30 | unixbsd | does the Linux kernel of devuan on AMD64 (testing or sid) read the UFS A6 FS format of openbsd? |
09:53.40 | *** join/#devuan chomwitt (~quassel@2a02:587:dc2f:b800:94ec:6dc2:1365:382e) |
10:09.42 | *** join/#devuan user282069 (user@2a01:7e00::f03c:92ff:fe7f:456d) |
10:10.15 | *** join/#devuan user282069 (user@2a01:7e00::f03c:92ff:fe7f:456d) |
10:13.51 | *** join/#devuan user282069 (user@2a01:7e00::f03c:92ff:fe7f:456d) |
10:20.30 | gnarface | unixbsd: yes, but the module might not have auto-loaded for you |
10:20.58 | gnarface | oh, well i dunno about "A6" specifically, whatever that is |
10:21.09 | gnarface | i would say modprobe ufs and try it |
10:21.10 | *** join/#devuan rsx (~rsx@ppp-188-174-156-175.dynamic.mnet-online.de) |
10:30.46 | *** join/#devuan cocoadaemon (~foo@60.ip-51-91-255.eu) |
10:31.11 | *** join/#devuan Volk (~Volk@a95-94-47-32.cpe.netcabo.pt) |
10:57.23 | *** join/#devuan KREYREN (~kreyren@gateway/tor-sasl/kreyren) |
11:39.23 | *** join/#devuan Captain4LK (~Captain4L@p200300edbf0a4434f6de0000010b9517.dip0.t-ipconnect.de) |
12:10.24 | *** join/#devuan mro_name (~mro_name@2002:5cc3:89f8:0:385a:b2fa:f85f:7f3e) |
12:20.47 | *** join/#devuan xrogaan (~xrogaan@unaffiliated/xrogaan) |
12:29.09 | *** join/#devuan Diagon (~DiagonalA@unaffiliated/diagonalarg) |
12:30.25 | *** join/#devuan Diagon_ (DiagonalAr@gateway/vpn/protonvpn/diagonalarg) |
13:00.26 | *** join/#devuan Captain4LK (~Captain4L@p200300edbf0a4434f6de0000010b9517.dip0.t-ipconnect.de) |
13:16.51 | *** join/#devuan gnu_srs2 (~srs@host-95-192-24-16.mobileonline.telia.com) |
13:19.16 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
13:38.11 | *** join/#devuan bru (~bru@181.31.228.135) |
13:45.14 | *** join/#devuan Kizano (markizano@2600:3c00::f03c:92ff:fe86:b66e) |
13:47.47 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
13:49.59 | *** join/#devuan Kruppt (~Kruppt@50.111.30.10) |
14:01.08 | *** join/#devuan arnoldoree (~arnoldore@113.210.102.79) |
14:03.39 | *** join/#devuan bru (~bru@181.31.228.135) |
14:04.27 | *** join/#devuan targz (~Thunderbi@unaffiliated/targz) |
14:05.29 | apollo__ | how would I go about finding out what processes are altering /proc/sys/vm/{dirty_background_bytes,dirty_bytes} and their friends? about once a week or so they get set to 0 which makes a mess for usb thumbdrives. |
14:06.07 | gnarface | fuser maybe? |
14:13.56 | apollo__ | gnarface, that would works if something was writing to the files continuously. but it just exits if nothing is accessing the file /right now/. |
14:15.50 | apollo__ | i mean, i could set up a cron to check the values every minute (i already have it set to write sane values in rc.local on boot), but that seems a hopeless throw-your-hands-in-the-air workaround rather than a i-know-whats-happening-here solution |
14:17.10 | gnarface | hmmm, maybe check out systraq or tripwire |
14:17.52 | gnarface | there's also inotify or dnotify but i think those are the type of things you gotta write code to use |
14:38.00 | *** join/#devuan m4rley (~m4rley@207.148.96.120) |
14:38.31 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
14:39.12 | *** join/#devuan drmbls (~marius@78-56-83-14.static.zebra.lt) |
14:49.56 | apollo__ | gnarface, thanks for the hints, checking... |
14:59.39 | DPA | Maybe auditd could help with that. |
15:09.10 | *** join/#devuan twiggy (~Twig@2.218.39.125) |
15:14.38 | *** join/#devuan DashiePie (~Rawr@c-73-152-100-71.hsd1.wv.comcast.net) |
15:15.36 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
15:16.54 | *** join/#devuan yae7nae4 (~yae7nae4@ezup.dev) |
15:21.15 | mason | unixbsd: Worst case, you can run a VM that can access your UFS. Remember that UFS is years past meaning just one filesystem, so universal support is harder and harder. |
15:22.33 | *** join/#devuan unixman (~aunixman@unaffiliated/eracc) |
15:35.52 | *** join/#devuan IoFran (~Thunderbi@189.237.102.209) |
15:37.19 | *** join/#devuan twiggy (~Twig@2.218.39.125) |
15:37.56 | *** join/#devuan Akuli (~akuli@82.128.159.204) |
16:04.00 | *** join/#devuan aperifons (~aperifons@99-90-85-53.lightspeed.sndgca.sbcglobal.net) |
16:14.08 | *** join/#devuan sunshavi (~user@190.42.230.123) |
16:21.13 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
16:26.18 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
16:30.25 | *** join/#devuan xinomilo (~xinomilo@gateway/tor-sasl/xinomilo) |
16:40.24 | *** join/#devuan cocoadaemon (~foo@60.ip-51-91-255.eu) |
16:40.25 | *** join/#devuan phillipsjk256 (~phillipsj@107-190-70-111.cpe.teksavvy.com) |
16:41.22 | *** join/#devuan Atari-Frosch (~frosch@atari-frosch.de) |
16:49.20 | *** join/#devuan AnimaInvicta (~AnimaInvi@ber27-1-78-230-215-86.fbx.proxad.net) |
16:56.12 | *** join/#devuan twiggy (~Twig@2.218.39.125) |
17:40.10 | *** join/#devuan IoFran (~Thunderbi@189.237.102.209) |
17:52.03 | *** join/#devuan Captain4LK (~Captain4L@p200300edbf0a4434f6de0000010b9517.dip0.t-ipconnect.de) |
17:55.01 | *** join/#devuan alexandros_tab (~quassel@unaffiliated/alexandros-c/x-1684531) |
17:55.08 | *** join/#devuan alexandros_c__ (~quassel@unaffiliated/alexandros-c/x-1684531) |
17:57.44 | *** join/#devuan sunshavi (~user@190.42.230.123) |
17:57.58 | *** join/#devuan qin (~qin@unaffiliated/soalokin) |
18:14.45 | *** join/#devuan petzi (~petzi@p578b3438.dip0.t-ipconnect.de) |
18:17.41 | *** join/#devuan aperifons (~aperifons@108-248-102-130.lightspeed.sndgca.sbcglobal.net) |
18:18.36 | *** join/#devuan Twiggy3 (~Twig@2.218.38.157) |
18:40.30 | *** join/#devuan Twiggy3 (~Twig@2.218.38.157) |
18:55.23 | *** join/#devuan DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
19:08.40 | *** join/#devuan DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
19:16.08 | *** join/#devuan smpl (~smpl@unaffiliated/smpl) |
19:34.11 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
19:38.46 | *** join/#devuan cocoadaemon (~foo@60.ip-51-91-255.eu) |
19:49.03 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
19:53.55 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
20:05.57 | *** join/#devuan shibboleth (~shibbolet@gateway/tor-sasl/shibboleth) |
20:08.25 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
20:11.34 | unixbsd | Hello, which software allows to modify a PDF documents, with adding Gifs, Png,... or Text on it, to be annotated? |
20:17.11 | djph | unixbsd: generally one would write some form of sourcecode document (e.g. LaTeX or LO) and then let that compile the PDF |
20:18.27 | unixbsd | I would like to add little text on the document, on each page. Of courses, doing with latex, would be nicely possible. I guess to add a y,x Text on it, at given fixed absolute position might be possible somehow. |
20:19.29 | djph | I mean, there are also paid-for editors if the only "source" you have is an already-compiled PDF. |
20:20.15 | specing | unixbsd: mupdf, probably |
20:20.19 | unixbsd | Sure, I have only a source : PDF (compiled, and only this). |
20:20.45 | unixbsd | mupdf has not editing capability, it is only a viewer. |
20:22.22 | onefang | LibreOffice, maybe GIMP. |
20:23.36 | unixbsd | I didnt know that libreoffice would edit PDF documents (here about 30-40 pages). |
20:24.43 | onefang | I vaguely recall that it does. |
20:25.56 | unixbsd | actually your idea to use pdfpages{ page ... = 1,... 30} + use x,y editing thing on the pdf (each pages), then run pdflatex might be interesting output. |
20:29.03 | unixbsd | it is however bit hard, I cannot recall how to place text or png at a given position using tex |
20:29.45 | *** join/#devuan salsbury (~kurgan@bl8-22-58.dsl.telepac.pt) |
20:32.10 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
20:52.22 | *** join/#devuan aperifons (~aperifons@99-90-85-53.lightspeed.sndgca.sbcglobal.net) |
21:07.53 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
21:08.11 | *** join/#devuan shibboleth (~shibbolet@gateway/tor-sasl/shibboleth) |
21:08.13 | *** join/#devuan furrywolf (~furrywolf@172.58.92.140) |
21:10.19 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
21:11.17 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
21:14.56 | *** join/#devuan aitor (~aitor@30.85-87-182.dynamic.clientes.euskaltel.es) |
21:16.07 | *** join/#devuan aitor_ (~aitor@30.85-87-182.dynamic.clientes.euskaltel.es) |
21:22.13 | *** join/#devuan ac_laptop (~ac_laptop@186.2.247.129) |
21:22.41 | aitor_ | unixbsd: i did something in the past using the haru library |
21:22.53 | aitor_ | http://libharu.org/ |
21:23.45 | rwp | unixbsd, I have a client using a commercial bit of software for filling in parts like gifs and things. But recently I was looking for a replacement and I think pdftk will do the same thing. |
21:24.18 | rwp | I have used pdftk a lot for other things and it is on my short list to go investigate using it to fill in pdf forms and to insert graphs and things. |
21:24.48 | rwp | aitor_, I haven't heard of haru before but I am adding it to my list to go look at! |
21:25.27 | rwp | Is haru packaged? One advantage of pdftk is that it is already packaged and available for installation. |
21:26.26 | rwp | It looks like libharu is all about generating pdf files. Is that right? My need, different from unixbsd I am sure, is to generate a pretty report in pdf format with portions filled in. |
21:26.30 | aitor_ | nope, as far as i know, but i did it in the past |
21:26.53 | aitor_ | libharu doesn't support reading and editing existing pdf files so far, but you can generate a new one and combine the result with an existing one using the pdftk toolkit |
21:27.46 | aitor_ | i did .deb packages a few years ago, and started working on some C libraries |
21:28.24 | aitor_ | c++, better said, but i abandoned the work |
21:30.33 | aitor_ | it might be interesting to take it up again |
21:31.47 | rwp | PDF has this concept of tagged fields that then can have objects stuffed into them later. |
21:31.53 | rwp | This allows someone relatively unskilled to generate a PDF template and then have it filled in later such as in a web page generating a report. |
21:32.51 | rwp | But almost everything about PDFs are ten times harder than they should be because of the Adobe nature of PDF. I wish there were something better. Like plain text! |
21:32.56 | unixbsd | onefang: I have modified the mupdf project, compiled it, and now it can show the position x,y. https://gitlab.com/openbsd98324/mupdf |
21:33.41 | unixbsd | then, I run pdlatex with the package: texblock* so that I can place at x,y the box or text, or even imaage on the top of pdf (pdfpage or wallpaper background pdf). |
21:34.02 | aitor_ | unixbsd, that's interesting! |
21:34.20 | unixbsd | really,? It is just little bit of C and tex. nothing else. |
21:35.07 | aitor_ | little things give ideas :) |
21:35.21 | *** join/#devuan qin (~qin@unaffiliated/soalokin) |
21:50.47 | *** join/#devuan nil21 (~ricardo@139.47.5.81) |
21:55.31 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
21:56.24 | unixbsd | aitor_: please find herewith the little source code, url: https://termbin.com/mbbf since it interests you. |
21:57.20 | aitor_ | thanks, i'll take a look |
22:08.05 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
22:10.33 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
22:10.40 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
22:14.08 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
22:19.59 | *** join/#devuan gnu_srs1 (~srs@host-95-192-179-251.mobileonline.telia.com) |
22:20.51 | *** join/#devuan aqu4bot (~aqu4bot@unaffiliated/subsen/bot/aqu4) |
22:26.17 | *** join/#devuan chomwitt (~quassel@2a02:587:dc0f:f400:94ec:6dc2:1365:382e) |
22:31.17 | aitor_ | rwp: *nope, as far as i know, but i did it in the past* <-- I was referring to the existence of the packages. haru is not packaged |
22:31.49 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
22:31.58 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
22:31.58 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
22:46.48 | *** join/#devuan Acacia (~Acacia@unaffiliated/acacia) |
22:48.25 | *** join/#devuan ricardo (~ricardo@139.47.5.81) |
22:48.47 | *** part/#devuan ricardo (~ricardo@139.47.5.81) |
22:49.03 | *** join/#devuan nil21 (~ricardo@139.47.5.81) |
22:54.30 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
22:55.41 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
23:09.49 | *** join/#devuan KREYREEN (~kreyren@gateway/tor-sasl/kreyren) |
23:09.54 | *** join/#devuan Uberius (~uberius@gateway/tor-sasl/uberius) |
23:09.55 | *** join/#devuan GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) |
23:19.29 | *** join/#devuan alexvp (~alexvp@51.194.52.176) |
23:42.05 | *** join/#devuan rapha (~rapha@unaffiliated/rapha) |
23:42.09 | rapha | hi all! |
23:42.51 | rapha | is there any reasonably sane way to get ruby and ruby-dev 2.7 from chimaera installed in beowulf? or would i be better off just dist-upgrading entirely? |
23:49.19 | onefang | Did you check if it's in beowulf-backports? |
23:49.59 | fsmithred | it's not |
23:52.49 | rapha | so i'm left only with uninstalling the package and building from source= |
23:53.29 | aitor_ | ruby-defaults 1:2.7+2 requires debhelper-compat (=12) |
23:53.31 | fsmithred | rapha, sometimes packages from testing will work in stable, but often they do not. |
23:54.06 | aitor_ | as builddep, i mean |
23:54.34 | fsmithred | did you try to install it and run into problems? |
23:54.35 | rapha | i tried putting the chimaera sources into sources.list but then immediately it wanted to upgrade half the system |
23:54.49 | fsmithred | yeah, you need to pin it to a lower priority |
23:54.53 | fsmithred | it=chimaera |
23:55.10 | rapha | oh? "pin it"? |
23:55.43 | fsmithred | yeah, you can set the priority lower, so it does not automatically install everything from chimaera. |
23:56.02 | fsmithred | you have to specify that you want a package from there |
23:56.24 | unixbsd | aitor: it works fine. I guess that I will add a little fltk frontend on top of mupdf to add content on pdf (over pdflatex). |
23:56.24 | fsmithred | or else just download the packages you need and install them with dpkg |
23:58.28 | aitor_ | unixbsd: pdftk? |
23:59.10 | fsmithred | ruby-dev ruby2.7-dev libgmp-dev libruby2.7 libgmp10 (= 2:6.1.2+dfsg-4), libgmpxx4ldbl (= 2:6.1.2+dfsg-4) |
23:59.19 | unixbsd | too heavy |
23:59.25 | unixbsd | I prefer to make my own mupdf |
23:59.43 | fsmithred | ok, now it's starting to look more complicated. rapha I didn't keep going with the dependencies. |
23:59.48 | unixbsd | anyhow there is no pdf editor today, so fltk + mupdf is best bet. |