37[00:15:40] <jelly> you will probably want to use the firmware images, to get the wired ethernet working during the installer run
38[00:15:44] <jelly> !firmware images
39[00:15:44] <dpkg> Unofficial <live> system and <netinst> and DVD images containing non-free Debian <firmware> packages are available from replaced-url
42[00:16:29] <jelly> oh good. I thought those bnx2x need fw to work properly
43[00:17:27] <Sora> oh they do for the flexlom
44[00:17:28] <jelly> never installed linux on a controller in hba mode, it should in theory be very similar to installing on a machine with sata controller and disks
45[00:17:32] <Sora> but i am using the on-board for installing
46[00:17:45] <Sora> ok my existing system uses legacy so legacy it is
196[02:04:15] *** Quits: Numero-6 (~Numero-6@replaced-ip) (Quit: << - Qui etes vous ? - Je suis le nouveau numero 2 - Qui est le numero 1 ? - Vous etes le numero 6 - Je ne suis pas un numero ! Je suis un homme libre!! >>)
230[03:14:13] <b_jonas> hi. I'm listing packages with aptitude, and I'd like to list which ones come from the contrib and non-free sections. I thought that the origin field "%O" would display this, but no, it just says "Debian:10.5/stable [all]" regardless if it's from the main or non-free part. which format can display this?
238[03:19:11] <b_jonas> I generally want to install packages from main only, but there are a few exceptions that I want to install from non-free, so I have to add them to the sources.list, but I don't want to accidentally intsall a package from there without knowing, so I need it in the main view
277[03:43:31] <jmcnaught> I don't have any u, C, or H flags on any of my packages so I don't know how to test if packages in those states would be selected by ~i but that would be my guess.
278[03:43:56] <b_jonas> yes, nor do I want to get a package in a broken state to test
279[03:44:11] <b_jonas> I could uninstall a package without purging to test that at least if I wanted
280[03:44:16] <b_jonas> well, I'll probably leave this for later
287[03:48:25] <jmcnaught> You're welcome, I learned something out of the deal :)
288[03:49:29] <b_jonas> and what I learned is that this channel is good for questions about package management itself. obviously it can't help about every software problem on a debian computer, so I'm never sure what questions I should bring here.
289[03:50:16] <mi11k1> b_jonas, legit questions
290[03:50:35] <mi11k1> anybody familiar with gmediarender?
292[03:50:53] <jmcnaught> If it's software in Debian I think you can ask here. Worst that could happen is someone recommends a more specialized channel.
293[03:51:18] * dvs puts down the rotten tomatoes
294[03:51:46] <b_jonas> most of the software that I use is in debian. for many, I already know where to ask. but not for everything.
295[03:52:20] <mi11k1> b_jonas, try to do a google search first
296[03:52:31] <mi11k1> usually you can get what u need there
297[03:52:38] <b_jonas> mi11k1: sure
298[03:52:46] <mi11k1> if im stumped, i bug ppl heree
299[03:52:46] <jmcnaught> Google has wrong answers, so does IRC though.
300[03:53:03] <mi11k1> if you google, stay away from ubuntu stuff
301[03:53:19] <mi11k1> eventually youll find a p[lace where theres valid info
302[03:53:25] <alex11> or any old advice in general
303[03:53:37] <mi11k1> actually i lied.
304[03:53:41] <mi11k1> i guess first
305[03:53:46] <mi11k1> and guess and guess
306[03:53:50] <b_jonas> mi11k1: I have figured that out. like a decade ago, I installed debian, then a few weeks later I deleted it without backups and installed debian, and I haven't looked back ever since.
307[03:54:15] <mi11k1> wanna know a secret...i use MX
308[03:54:21] <mi11k1> for my desktop
309[03:54:45] <mi11k1> when i have mx questions, i ask them here and dont say its mx
310[03:54:46] <jmcnaught> Before anything I check /usr/share/doc/$package/ especially for a README.Debian. Then man pages. I avoid GNU info pages but will go there if necessary. Then search engines.
311[03:55:13] <mi11k1> or kali even
312[03:55:49] <mi11k1> so, nobody knows gmediarender?
314[03:56:54] <b_jonas> jmcnaught: sure, I know all those usual places to check. also, apart from the html or pdf docs in /usr/share/doc/ , man pages, info docs, and online docs, there's one more place to look for docs: a few programs hide their reference documentation in the executable such that you can access them with something like program-name help --verbose then program-name help subcommand-or-doc-section
315[03:57:24] <mi11k1> or just guess
316[03:57:43] <b_jonas> svn does that most notably, for multiple of its programs, it's evil, I should go through and extract all the help pages it has once and concatenate them to a single searchable text file so
317[03:57:44] <mi11k1> tab works well
318[03:57:47] <b_jonas> but so far I was lazy
319[03:58:21] <derpadmin> b_jonas : or use git?
320[03:58:34] <mi11k1> yah git
321[03:58:41] <mi11k1> get it from git
322[03:59:13] <b_jonas> and as for documentation, I added non-free specifically to install some docs that use the options in the GFDL license so they are in the non-free section of debian; though I will probably install a few other packages from outside of main
323[03:59:17] <derpadmin> huh? I meant, use git instead of svn..
426[06:02:46] <genr8_> i would assume its either a borg daemon that stays running that then launches the task, a systemD service or timer, or the /etc/crontab or /etc/cron.* dirs
648[09:51:34] <RadoS> Moin, when booting with 2 notebooks, on A the services are printed/ visible as they are started, on B it only switches the display and shows the fsck/mount status, but no started services progresses is visible. I _guess_ it's still printed, just not visible. How to investigate this further?
664[10:11:44] <shad0w_> debian testing + xfce + picom is so sweet.
665[10:12:36] <ratrace> RadoS: are you talking about systemd boot output? maybe you have "quiet" in your kernel command line so it's hidden. you can hit when you have post-grub blank screen to show that output, or remove "quiet" from the kernel command line
666[10:14:35] *** Quits: szorfein_ (~ninja@replaced-ip) (Remote host closed the connection)
678[10:19:29] *** Quits: chele (~chele@replaced-ip) (Remote host closed the connection)
679[10:19:31] <genr8_> when they screw you, screw back.
680[10:19:32] <Lope> ratrace, when I try start the VM I get: unable to map backing store for guest RAM: Cannot allocate memory
681[10:19:47] <aderchox22> Is it possible to preseed the network configurations except only the hostname ?
682[10:19:50] <shad0w_> debian is the first distro where suspend works out of the box for me on my MBP
683[10:19:58] <Lope> ratrace, it seems like there's one more step I might need to take?
684[10:20:00] <shad0w_> and i've tried about them all.
685[10:20:32] <Unit193> shad0w_: It might be worth mentioning that testing support is on OFTC in #debian-next, though. Heh, amusingly I have suspend issues. :P
687[10:20:45] <Lope> ratrace, I see people mentioning hugeadm, but I don't have that nor is it found in the debian repos.
688[10:20:53] *** Quits: yonder (~yonder@replaced-ip) (Remote host closed the connection)
689[10:20:53] <ratrace> Lope: sure. this is what I do to pre-reserve (12GB) hugepages for my VMs: replaced-url
690[10:21:40] <ratrace> Lope: problem with hugepages (esp 1GB ones) is that the kernel needs contiguous memory for them. it's very likely that you might not have it even if you try to reserve few GB less than your total physical RAM
691[10:22:05] <genr8_> reboot
692[10:22:44] <shad0w_> Unit193: Q1. what does oftc means lol
693[10:22:44] <ratrace> Lope: hugepages reserved and used like that are _not_ available to your host. also, with qemu-system at least you have to point it to the reserved region, eg. with -m ${MEMORY}G -mem-prealloc -mem-path /dev/hugepages1G (or wherever you mounted them)
694[10:22:52] <Unit193> !oftc
695[10:22:52] <dpkg> OFTC is the Open and Free Technology Community, a support/collaboration service. They have an IRC network: irc.oftc.net. You may (or may not) be connected to OFTC's network. replaced-url
711[10:24:31] <ratrace> (don't have to = you can chown directly as VM being the owner)
712[10:25:11] <ratrace> Lope: all this si supposed to be automagic on boot if you set appropriate sysctls etc.. but I never got that working, so I run explicit reservation like that paste, in a custom service unit on boot
713[10:25:51] <Lope> ratrace, what does shm mean in shm-hugetlb?
714[10:25:51] <ratrace> Lope: you can reserve it and free it back to the host but the biggest problem is fragmentation. once you free it, there's no guarantee there will be enough contiguous RAM to reserve it back
715[10:26:03] <ratrace> Lope: shared memory, the group name is arbitrary, I made it up
716[10:26:21] <Lope> okay
717[10:26:23] <ratrace> Lope: so if you have fragmentation issues, consider using smaller pages like 2MB ones
718[10:26:41] <ratrace> I prefer 1GB because in theory those should offer best performance with least number of TLB lookups
719[10:26:43] <genr8_> the flag is "SHM_HUGETLB"
720[10:26:56] <ratrace> genr8_: flag?
721[10:27:01] <Lope> yeah, I've got 32GB RAM and I run VM's with whole numbers of GB of RAM, so I figured 1GB pages seems like a good diea
722[10:27:04] <genr8_> extension
723[10:27:18] <ratrace> genr8_: what extension? I made that group name up
724[10:27:25] <genr8_> no you didnt :)
725[10:27:31] <genr8_> thats literally what its called in the kernel :)
726[10:27:33] <ratrace> Lope: it's great idea if you can reserve it in advance. but later you might not be able to
727[10:28:00] <ratrace> genr8_: okay I mean it's not a system group or anythin ginstalled by packages. I made it up and yes quite possibly subconsciously due to what you say
728[10:28:01] <jmcnaught> Does using hugepages make a noticeable difference?
729[10:28:22] <genr8_> yea, well, it makes perfect sense
730[10:28:24] <ratrace> jmcnaught: in theory yes, I haven't benchmarked it
731[10:28:50] <Lope> ratrace, I'm not sure your commands are appropriate for me
732[10:28:55] <Lope> because I've got /dev/hugepages/libvirt/qemu
746[10:32:58] <genr8_> It bumps the limits to 128 PiB of virtual address space and 4 PiB of physical address space.
747[10:33:07] <genr8_> Original x86-64 was limited by 4-level paing to 256 TiB of virtual address space and 64 TiB of physical address space.
748[10:33:32] <Lope> ratrace, all I did was: mkdir /dev/hugepages1G; mount -t hugetlbfs -o pagesize=1G none /dev/hugepages1G (from your URL, thank you). then I configured hugetlbfs_mount = "/dev/hugepages1G" in qemu.conf.
749[10:33:33] <genr8_> very interesting subject if you dive into it actually
750[10:33:58] <Lope> and obviously the kernel params and xml adjustment.
751[10:34:11] <ratrace> Lope: right so libvirt does the actual reservation I suppose
752[10:34:21] <ratrace> Lope: what kernel params, btw?
753[10:34:46] <Lope> it seems so. In libvirt you can get technical and specify different memory backing areas and sizes on whatever numa nodes you want etc.
754[10:34:48] <bipul> I have tried using `user data`, i.e to put preseed instruction inside `user data`. but it seems it does not worked out. replaced-url
755[10:35:17] <Lope> but I just told libvirt to use hugepages and let it do it auto. I don't have ryzen yet, using haswell, so I'm not worried about numa nodes at this time.
764[10:36:38] <Lope> oh, so you can do it without kernel params?
765[10:36:43] <Lope> just echoing into that /sys ?
766[10:36:46] <ratrace> yes
767[10:36:59] <genr8_> its better to do it on fresh boot so your address space is not cluttered
768[10:37:06] <Lope> but surely I would need to start my kernel with hugepagesz=1G at least?
769[10:37:15] <ratrace> I prefer it dynamically alocated (because I free them pages when the VM exits)
770[10:37:30] <Lope> ratrace, that's pretty cool, how do you free them?
771[10:37:37] <ratrace> Lope: echo 0 > ....
772[10:37:47] <Lope> easy enough, thanks.
773[10:37:55] <Lope> it seems like a crazy jump 4K 2M or 1G
774[10:38:09] <Lope> I'd prefer 256MiB pages, or something.
775[10:38:13] <ratrace> Lope: I think that kernel option is merely to point at which page is hugepages=N referring to. there's really only two sizes on x86: 2MB and 1GB
776[10:38:35] <ratrace> and both are available on the kernel /sys API regardless of kernel command line options
777[10:38:40] <Lope> okay
778[10:38:48] <Lope> I guess there's little boys and big boys.
790[10:40:58] <Lope> I've got a bunch of old laptops, incl an old surface pro with 4GB RAM soldered in. I feel like an idiot for buying it sometimes. But I often find creative uses for it.
791[10:41:34] <Lope> 4GB is like a DE plus a small ZFS ARC, and nothing else.
792[10:41:53] <ratrace> would make a good router :)
793[10:41:58] <Lope> yeah hahahahah
794[10:42:13] <Lope> a router with a 10" full HD IPS display.
795[10:42:19] <ratrace> network metrics display!
796[10:43:02] <Lope> ratrace, can you recommend any network monitors or OS's etc?
797[10:43:46] <ratrace> I use Munin because I've used it for many many years, but maybe there are better ones now, more modern ones with fancy javascripts graphics instead of images
798[10:43:56] <ratrace> like something based on graphana
799[10:44:04] <ratrace> *grafana
800[10:44:17] <Lope> ok cool, will check it out. My one friend is raving about pfsense.
801[10:44:22] <ratrace> Prometheus ferinstance.
802[10:44:36] <ratrace> !info prometheus
803[10:44:55] <jelly> prometheus and grafana seem nice
804[10:45:06] <ratrace> what, !info no mas triggering?
805[10:45:10] <Lope> what do you think are the attack vectors for meltdown and spectre if you're just doing dev in a VM?
806[10:45:38] <genr8_> compiling malicious code ?
807[10:45:40] <Lope> Since JavaScript is mitigated, I mean pretty much all of the software is trusted.
808[10:45:52] <Lope> I don't compile malicious code :/
813[10:46:47] <ratrace> so why does dpkg not see it for !info ..... bah.
814[10:47:23] <genr8_> ,i prometheus
815[10:47:25] <judd> Package prometheus (net, optional) in buster/amd64: Monitoring system and time series database. Version: 2.7.1+ds-3+b11; Size: 10710.1k; Installed: 46663k; Homepage: replaced-url
816[10:47:31] <ratrace> Lope: meltdown and spectre are infoleak vulns. you'd have to be targeted for some juicy info
817[10:47:44] <ratrace> genr8_: ah judd has that too, mkay
830[10:54:42] <jmcnaught> For what it's worth I have been gaming in a Windows VM with 16GB of memory without hugepages and without CPU pinning and I don't have any noticeable latency. LatencyMon says my VM is "suitable for handling real-time audio and other tasks without dropouts."
831[10:56:14] <shad0w_> Lope: try 8gb soldered on in your laptop lol
845[11:01:27] <ratrace> jmcnaught: in theory you can only have lower latency, even if it's not noticeable to you, with hugepages. less TLB lookups can't be worse
874[11:22:15] <no_gravity> I wonder if I should install ipconfig on it. Or if there is some other tool that tells me my ip.
875[11:22:29] <genr8_> "ip a"
876[11:23:03] <ratrace> genr8_: from what I read it's the limit of 64-bit allocation. it's just virtual memory so it makes sense it's max the size that can possibly be, being virtual
891[11:27:13] <ratrace> genr8_: "In addition, the AMD specification requires that the most significant 16 bits of any virtual address, bits 48 through 63, must be copies of bit 47 (in a manner akin to sign extension). " from here: replaced-url
892[11:27:28] <genr8_> nice
893[11:27:31] <ratrace> genr8_: so I guess AMD spec does it
905[11:34:20] <oxek> does debian announce future release dates of big software upgrades, such as firefox78esr? I'd like to know when it's coming so that I can prepare.
914[11:36:20] <aderchox22> When I press ESC, I get "boot: " and it's ready to type, but when I type auto url=URL the message I get is loading auto: no such file or directory
933[11:43:51] <ratrace> so we found some mozilla forum post that _implied_ 68.12 was last, there's no .13 coming, and 78.3 is teh next upgrade really for Stable
934[11:44:01] <jelly> ratrace, upstream usually upgrades users of their native builds after .2 or .3 of next ESR is out
935[11:44:16] <ratrace> which landed yesterday in sid so I'm guessing a few days before it's in Stable.
936[11:44:26] <ratrace> jelly: right, but there's no explicit calendar stating EOL dates
937[11:44:49] <genr8_> aderchox22, I found the manual: replaced-url
938[11:44:50] <ratrace> or no explicit bit of info stating which .minor version is considered EOL for previous major.
939[11:45:05] <genr8_> "This is enabled by using the Automated install boot choice, also called auto for some architectures or boot methods. In this section, auto is thus not a parameter, it means selecting that boot choice, and appending the following boot parameters on the boot prompt."
947[11:47:51] <oxek> ratrace: there's nothing explicit, but it was always implied that when only one esr release it upgraded and not the other, it means the older one is out of support
948[11:47:57] <oxek> and that has happened now with 78.3esr
949[11:48:04] <jelly> ratrace, and they broke the previous URL replaced-url
952[11:49:59] <genr8_> "The auto boot choice is not yet defined on all arches. The same effect may be achieved by simply adding the two parameters auto=true priority=critical to the kernel command line."
954[11:50:30] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 272 seconds)
955[11:50:39] <ratrace> oxek: not quite. ESR 78 and 68 overlapped for three minor releases: 78.0 .1 and .2, with 68.10 .11 and .12 corresponding releases
957[11:51:15] <ratrace> so there's no explicit info stating "there will be N ovelapping before old ESR EOL's", there's also no explicit statement saying 68.12 is last, etc...
958[11:51:31] <jelly> no, no there isn't
959[11:51:40] <genr8_> what are we gonna do when Mozilla becomes Microsoft
960[11:51:46] <aderchox22> genr8_ hitting tab gives language names! E.g., Hungarian, English, etc. Also there's no "install" option to append the auto line after. (of course I did but I got the same error).
961[11:52:16] <aderchox22> I don't know how I can add parameters to the kernel command line before installing... [cringe]
962[11:52:19] <genr8_> well go into the command line and just add that stuff
963[11:52:24] <oxek> ratrace: that's what I said. While they overlap they are supported, once only one of them gets upgraded (78.3 but no 68.13), then only the newer one is supported
964[11:52:29] <jelly> and those values varied a bit in the past, sometimes there were 3 overlapping. sometimes 4, sometimes they cheated and made versions like 52.2.1esr
965[11:52:42] <ratrace> oxek: okay, but unfortunately that comes with no prior announcement.
966[11:52:52] *** Joins: Immanuel (~Manu@replaced-ip)
967[11:53:18] <ratrace> and given it takes days for Debian to Stabilize a package, there's a window, like it is right now, where a particular CVE exists, is known, and we're wide open.
968[11:53:34] <genr8_> Its a grub thing, press E or C, or something
970[11:53:51] <genr8_> read the instructions on the bottom
971[11:54:31] <oxek> i mean, there's no announcement that 80 is no longer supported once 81 is released either - it is assumed that you upgrade to the newer release once your release no longer gets upgrades.
972[11:55:10] <ratrace> now, if mozilla did something like openssl does, "78.3 coming up, there's high-impact CVEs, there won't be 68.13, heads up!", then distros could do coordinated releases in a timely fashion
973[11:56:02] <genr8_> i lost trust in mozilla this year
974[11:56:07] <aderchox22> genr8_ you know I've not installed the Debian yet, right? I don't get the Grub yet I guess. I only get a menu which asks me whether I want the Debian Installer, the Debian Graphical installer, or booting up the Debian Live and two other irrelevant options
975[11:56:10] <ratrace> and debian really has this dumb policy of waiting until old esr expires....
1032[12:26:38] <genr8_> The following command (available from the whois package) can be used to generate a SHA-512 based crypt(3) hash for a password: mkpasswd -m sha-512
1033[12:27:12] <genr8_> I guess it does allow plaintext passwords too:
1034[12:27:22] <genr8_> # Root password, either in clear text
1043[12:28:37] <genr8_> if you want plaintext instead of crypted, you need to use Both of those 2 lines instead of the d-i passwd/user-password-crypted password [crypt(3) hash]
1061[12:52:10] <genr8_> yes. but it doesnt need to be looked up. generate a SHA-512 based crypt(3) hash for a password: mkpasswd -m sha-512
1062[12:52:57] <genr8_> man 3 are the dev man pages and my install doesnt have them at least, i checked that
1063[12:53:00] <aderchox22> Right thank you.
1064[12:53:38] <aderchox22> Mine has it but it's "programmingly", the cli way (mkpasswd) is easier.
1065[12:54:17] <genr8_> its a C function and that program implements it. it also specifies that POSIX doesnt require a specific hash algorithm, it just has to be agreed upon. SHA-512 in this case.
1102[13:13:35] <sponix2ipfw> genr8_: doing it for security?
1103[13:14:16] <genr8_> im still investigating kernel stuff as we speak. I did all the work on 4.19.132 cause thats what comes from Buster Apt-get. then 2 days ago, I found out that 4.19.146 came out on the debian dev gitlab. Now Yesterday 147 came out on kernel.org
1116[13:22:44] <genr8_> Im trying to figure out a better way to edit these config files. I kinda just winged it, taking the 200KB /boot/config-4.19.0-10-amd64 and starting with that, and turning it into a 130KB result
1117[13:22:49] *** Quits: frostschutz (~frostschu@replaced-ip) (Remote host closed the connection)
1126[13:25:49] <genr8_> I do use menuconfig, But menuconfig operates on 1 file at a time. I want Diff'ing and A/B'ing, and not having to press "Help"/Escape everytime i want to read the documentation on the item
1127[13:26:52] <genr8_> the 3 .config files debian uses are KCONFIG='debian/config/config debian/config/kernelarch-x86/config debian/config/amd64/config'
1128[13:27:22] <genr8_> 172K, 40K and 4K respectively
1132[13:28:31] <jelly> presumably the last one wins?
1133[13:28:53] <genr8_> i believe so. (forgot what i read)
1134[13:29:24] <genr8_> so those last 2 are important architectural options, but thats kinda what im changing.
1135[13:30:22] <genr8_> so only changing the first one means stuff will just get overrode. and waiting for them all to combine, and then editing it, .... somehow doesnt feel right.
1137[13:31:59] <genr8_> I want a better "menuconfig" kernel utility that knows how to operate on multiple files, and has 2 identical sections A / B and knows how to diff, and also takes in the documentation blurbs from the source tree
1138[13:32:36] <genr8_> otherwise doing everything manually is gonna suck. its too tedious.
1139[13:33:20] <genr8_> I wanna be precise about this and come up with "patch" files that have comments for everything I did like "I Disabled this option and heres why"
1140[13:35:09] <genr8_> We need a better "menuconfig" in 2020
1150[13:48:21] <genr8_> I would document what I figured out, but im sure it would change before anyone cared enough to find it and be helped by me deciphering which script calls which build target and when it deletes a directory and why stuff is being built without asking for it, and what command options to use and env variables to set
1153[13:50:32] *** Quits: kmikita (uid461814@replaced-ip) (Quit: Connection closed for inactivity)
1154[13:50:59] <genr8_> I was basically told "just run make deb-pkg" And of course that failed immediately due to patches. then I figured out theres a "dpkg-buildpackage" perl script that knows how to run in the kernel. That also failed halfway through. Then I re-ran that script and it deleted my entire "build/" directory and erased all my progress of the last 8 hours. (then i found out theres a -nc "no pre-clean") command
1155[13:52:34] <genr8_> not to mention all the scripts just build a ton of extra stuff like Docs, Headers, Debug tools, re-packaging up the source tree (in a single-threaded xz instance - actually I might submit a patch for that first)
1161[13:55:35] <genr8_> oh and speaking of multi-threaded, the Debian Kernel Handbook manual lists commands that only build in single thread mode, and telling it to run in parallel was not so simple
1162[13:56:22] <epitamizor> just need to edit make's build options config file
1163[13:56:56] <genr8_> it has provisions for that, its just not explained how to use them
1165[13:58:05] <genr8_> and if you do the obvious thought and run with "-j6" from your initial "debian/rules.gen" command, the different makefile steps end up stepping on itself and sends itself repeating into an infinite loop on the checking modules and trimming unused kernel symbols step.
1167[13:59:55] <genr8_> i havent figured out the exact cause of why yet, I presume the next make job launched deletes the progress of the job before it before it can complete
1172[14:01:29] <RadoS> ratrace, whatever is responsible for printing things like "starting ssh deamon...", I guess it is systemd. I'll have a look if the 2 systems have different boot params despite being installed the same way.
1173[14:01:47] <genr8_> it DOES work if you do everything in the exact order the scripts want, all at once, but the process is so long that I had no choice but to start customizing build scripts I didnt understand
1178[14:04:01] <genr8_> Its made doubly confusing because theres the kernel makefiles we all know and love (make menuconfig, make vmlinux, make modules) , and then theres the debian scripts Which are just ridiculous imo
1183[14:05:09] <genr8_> and the documentation does NOT cut it
1184[14:09:42] <genr8_> i'd consider myself an advanced user in other ways, and I know how to use the kernel makefiles, but with the debian ones combined, its been 4 days so far, and ive only built 3 kernels.
1190[14:13:10] <genr8_> the learning curve is made worse because of CPU intensive long tasks, and build progress getting deleted, or sequences of makefile steps that only work if you run all of them in order
1191[14:13:20] <jelly> genr8_, if you can document things a bit perhaps we can write a simpler howto to put on wiki.d.o
1192[14:13:42] <jelly> (and by we I mean mostly you)
1193[14:15:45] <genr8_> Ill consider it, but why does it have to be me? Ive been in this situation before, and my suggestions were denied for official release because "Oh you did that wrong, you just have to do this other thing we didnt mention instead".
1197[14:17:52] <genr8_> I will write my findings, but if ive only been at this 4 days, theres a good chance I picked the bumpiest road to victory
1198[14:18:56] <ratrace> but the spoils are that much sweeter!
1199[14:19:48] <genr8_> a lot of the "findings" were just "Investigate which scripts are being run, examine the command lines, the Env variables, and the parameters. Figure out the order things are being run, which build step touches which directory" kinda thing
1201[14:22:13] <genr8_> That should be documented better, but these things were clearly made by smarter people than me, they just put all the work into making the scripts work rather than documenting why
1218[14:30:08] <epitamizor> try working on a project for years then have some newbie tell you your project is failing
1219[14:30:27] <hwm4rgs> not an excuse for that kind of behaviour.
1220[14:30:43] <genr8_> of course I can anticipate they're gonna say "just use our kernel, our only job is for us to make the kernel for YOU. It works fine. If you wanna make your own kernel you're on your own"
1241[14:47:00] <genr8_> yeah, must be cause I have more ways to relate to it in daily life.
1242[14:49:31] <codedmart> I am a long time arch user. Have used ubuntu here and there over the years but mostly as a server and never for more then a week on my laptop/desktop. I have been on arch because of a craze of wanting the latest and greatest all the time. But I definitely hit bumps here and there. So my main question is how do you use debian which allowing for some of your applications to be at the latest version. Like thunderbird, chromium, firefox, gimp
1243[14:49:32] <codedmart> , etc? Is flatpak the way to go?
1244[14:50:02] <codedmart> I am really considering debian to get a more stable base system, but would still like some of my everyday apps to be newer.
1245[14:50:03] <genr8_> you can give up on that mentality
1248[14:50:21] <dpkg> Shiny New Shit Syndrome is a serious disorder, which usually breaks out into an epidemic every time something new is released. If you have SNS, ask me about <backports> and <ssb>; these are better options than upgrading to <testing> because it is a <moving target>.
1249[14:50:39] <genr8_> !backports
1250[14:50:39] <dpkg> A backport is a package from a newer Debian branch, compiled from source for an older branch to avoid dependency and <ABI> complications. replaced-url
1251[14:50:53] <fireba11> lol
1252[14:50:57] <genr8_> !debian-backports
1253[14:50:57] <dpkg> backports.debian.org (formerly backports.org) is an official repository of <backports> for the current stable (see <buster backports>) and oldstable (<stretch backports>) distributions, prepared by Debian developers. Ask me about <backport caveat> and read replaced-url
1254[14:51:03] <codedmart> Nice that is funny.
1255[14:51:07] <ratrace> nothing wrong in wanting or needing _some_ applications to be latest. browsers especially.
1256[14:51:07] <codedmart> I have sns for sure.
1257[14:51:11] *** Quits: pvdp66556 (~pvdp@replaced-ip) (Remote host closed the connection)
1258[14:51:31] <H4ndy> Some softwares also offer their own repository you can use to get specific or newer releases
1262[14:51:48] <ratrace> codedmart: you can install snaps if you like snaps. snapd works on debian, albeit you might need to fiddle with PATH to get snap specific binaries in it
1263[14:52:20] <codedmart> H4ndy is that frowned upon or a bad idea?
1264[14:52:39] <codedmart> ratrace I have never used snaps or flatpaks. Is there performance issues with either?
1265[14:52:58] <ratrace> codedmart: only startup times for snap'd applications are somewhat longer
1266[14:53:22] <ratrace> third party repos are generally bad idea unless it's a known well behaving repo, like for example pgdg
1268[14:53:28] <H4ndy> codedmart: I don't know, don't care. I use it for some server software to get releases directly from the vendor or a well-known packager
1273[14:55:00] <codedmart> Yeah I guess I should just dive in and see. I do need to let go of the sns a bit though. Even though I have mostly been fine and had good experieces with Arch. I have definitely had some down time or moments of brokeness.
1274[14:55:24] <H4ndy> I don't use Debian on any desktop environments, only headless and packages are a bit easier to handle because there are no graphical dependencies (GUI, window manager, drivers, etc.) that can easily break. So I had 0 problems with external repos so far.
1276[14:56:08] <codedmart> Really want to try and let that go a bit. And I don't particularly care for ubuntu. Nothing against it, but debian has always seemed like a solid/stable system to run.
1277[14:56:23] <genr8_> good
1278[14:56:40] <genr8_> everything you get outside of the debian repos pokes another hole in your machine. sooner or later it may be all holes and you'll say "guess im just gonna reinstall now, just like I do on Windows"
1279[14:57:08] <genr8_> and i keep having to bring this up. The Arch AUR serves a purpose, but thats technically untrusted code.
1320[15:17:04] <genr8_> "Flatpak and Snap" acts like theyve solved it all <--- all theyve done is remove user agency and your control over what gets run and said "Trust us everythings fine". (Hint: it may or may not be fine)
1321[15:17:09] <ratrace> there's just one giant glaring issue with that chroot break .... it's experimental and useless in real world scenarios
1322[15:17:25] *** Quits: dez (uid92154@replaced-ip) (Quit: Connection closed for inactivity)
1327[15:17:53] <epitamizor> ofcourse, anything that is documented is not relevant when it comes to security
1328[15:17:55] <ratrace> once chrooted, it's rather very hard to break out without a vuln (and assuming the chrooting process (not chrooted) didn't mess up)
1336[15:21:37] <greycat> For reference, in case it hasn't been brought up yet, the Linux man page chroot(2) describes one way to break out of a chroot.
1339[15:22:32] <ratrace> greycat: that's the way explained in the link above and requires messing and moving with directories by the privileged, chrooting process
1340[15:22:35] <greycat> manpages-dev package
1341[15:22:45] <epitamizor> unless its a closed box, then its going to be open for exploit
1342[15:23:19] <genr8_> well its not perfect, but its certainly gonna help.
1343[15:23:43] <ratrace> like.... move the directory with an open fd given to chrooted process, ousside of the chroot. well, duh! you just.... moved..... it...... ousside..... wth.....
1345[15:24:47] <epitamizor> its almost like a cat/mouse game, the cheese will move and the mouse will go find it, but the cat will be in the area sooner or later
1346[15:25:17] <genr8_> doesnt mean the mouse should give up and shout "HEY IM A MOUSE"
1347[15:25:24] <ratrace> proper security is alwyas defense in depth. no single tool, like chroot, is the one and only tool to use
1375[15:33:06] <BadPractice> hey, i want to use my pocket book 632 with calibre. When i pc link my file browser lets me acess the memory but calibre does not recognize the device. I have found a post of the devs that that could be cause from the file browser auto connecting via mtp (which makes it impossible for calibre to do so). Im using a standard Gnome enfironment using X11. Is there a way to temporarely disable mtp device discovery?
1376[15:33:16] <genr8_> Im sure you're an expert on this, so why dont you tell us what you do.
1379[15:36:14] *** Quits: lesless (~lessless@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1380[15:36:46] <cipherize> BadPractice: You should be able to add a udev rule to tell it to ignore the device, so that MTP is available when calibre tries to connect.
1381[15:37:58] <cipherize> Ignore the device specifically for mtp purposes.
1382[15:38:20] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
1383[15:39:31] <genr8_> my point was "least privilege is not handled for us." Doing something extra to protect yourself from rogue programs sending tentacles and tendrils into every back orifice of your computer, should be considered mandatory now. Everyday programs do not deserve to be trusted, I see too many people let them run wild, even on linux.
1384[15:39:41] <BadPractice> cipherize, has been a while till i made a udef rule. A little guidence would be nice
1385[15:40:00] <cipherize> BadPractice: Sure. Can you look at the output of lsusb and find the pocket book?
1386[15:40:16] <genr8_> "Rules for rules" replaced-url
1387[15:40:24] <jelly> genr8_, but that takes resources and energy, and people maybe don't like spending those without a clear goal
1388[15:40:38] <jelly> and it's very hard to say "this system is now 70% more secure"
1389[15:40:49] <BadPractice> cipherize, already on it
1390[15:40:57] <cipherize> jelly: Reject 70% of all login attempts. ;)
1391[15:41:04] <genr8_> then the devs should have put more effort in to make a less toxic program
1392[15:41:32] <greycat> *plonk*
1393[15:41:33] <jelly> genr8_, will you pay the devs and provide specifications?
1394[15:41:41] <genr8_> I blame terrible devs and terrible corporations who give 0 shits. the user is bombarded and overwhelmed. duplication of work.
1397[15:42:43] <genr8_> I dont even have enough resources to protect myself. But i will provide more information if people care.
1398[15:43:07] <genr8_> I dont get why I have to be the one. I'm nobody
1399[15:43:12] <jelly> I managed to convince the company where I work to support a security project with hard cash, and then I use their software to make pieces of infra a bit more resilient
1484[16:18:27] <ahabchi> same for me. Actually CUPS is not working for me
1485[16:18:44] <jelly> genr8_, you are allowed to stay quiet when you have no constructive approaches to suggest for a particular issue
1486[16:19:01] *** debhelper sets mode: +l 1175
1487[16:19:02] <ahabchi> when i add the printer to network i could find it using print settings app ut when trying to print the job gets stopped for no reason
1491[16:19:45] <ahabchi> i used to print with usb but when i upgraded hplip it is no longer working
1492[16:19:54] <b1ack0p> is debian 9.13 still supported?
1493[16:21:08] <jelly> !stretch
1494[16:21:08] <dpkg> Stretch is the codename for the current <oldstable> release, Debian 9, released 2017-06-17. "Stretch" is the rubber octopus in Toy Story 3, see replaced-url
1495[16:21:21] *** Quits: dreamon (~dreamon@replaced-ip) (Remote host closed the connection)
1496[16:21:40] <jelly> b1ack0p, not by debian, but Debian LTS supports a subset of packages and architectures
1497[16:21:45] <jelly> !stretch-lts
1498[16:21:45] <dpkg> Security support for Debian 9 "stretch" from the Debian Security Team ended July 6 2020. The <LTS> team will continue to provide limited security support for some architectures and a subset of packages until June 30 2022 (total 5 year life). See replaced-url
1499[16:21:48] <jelly> there.
1500[16:21:51] <b1ack0p> huh
1501[16:22:01] <b1ack0p> so should i pay for lts? :p
1591[16:37:54] <n4dir> ask people who know about such, if my utter feeling that systemd is more ressource intensive than other inits is correct (it probably isn't ...)
1593[16:38:15] <b1ack0p> n4dir: well it is ok i dont expect much on this laptop
1594[16:38:50] <n4dir> yeah, if it works for you, no need to look into such. I also changed small things, like terminal emulators with little ressource usage (i used st)
1595[16:39:00] <n4dir> but as you said: xfce works just fine too
1596[16:39:45] <b1ack0p> yes running xfce now
1597[16:40:18] <n4dir> fluxbox always works like a charme. And it is pretty intuitive (compared with say openbox)
1598[16:40:34] <hmuller> i like i3wm. going to experiment with sway soon.
1606[16:42:11] <n4dir> to get it of my chest: after the 3 laptops were gone i decided to put FreeBSD on the iMac (as debian was kinda slow). That went south too, and i can't install anything else anymore. When it rains it seems to pour. Oh my ...
1630[16:54:46] <Franciman> I was thinking about installing debian testing on my pc. I read, though that sometimes there can be delays in security updates
1635[16:55:18] <Franciman> while this would not be a problem for stable and experimental
1636[16:55:28] <dvs> !debian-next
1637[16:55:28] <dpkg> #debian-next is the channel for testing/unstable support on the OFTC network (irc.oftc.net), *not* on freenode. If you get "Cannot join #debian-next (Channel is invite only)." it means you did not read it's on irc.oftc.net. See also replaced-url
1638[16:55:32] <Franciman> but do you think that it's safe to use debian testing on a proper pc?
1639[16:55:37] <Franciman> oh DRAT
1640[16:55:42] <Franciman> thanks dvs
1641[16:55:46] <dvs> np
1642[16:56:21] *** Quits: grobi (~rtng@replaced-ip) (Remote host closed the connection)
1713[17:41:51] <oxek> can someone help me find the documentation of what the 'sw' option does in /etc/fstab? I checked manpages of fstab, mkswap, swapon, debian wiki, and did not find what I am looking for.
1714[17:42:04] <oxek> the 'sw' option only applies to swap
1715[17:43:53] <cipherize> oxek: I want to say that it's a carry-over from BSD.
1716[17:45:39] <oxek> that's indeed possible.
1717[17:45:40] <cipherize> In fact, I believe that, in Linux, those have no effect.
1718[17:45:58] <cipherize> The only mount option that I am aware of that has any function for fstype "swap" is "pri="
1719[17:46:14] <oxek> my main issue is that I don't know if I should apply the 'sw' option to swapfiles as well as swap parititons, or whether I should just use 'default' for the option column.
1720[17:46:41] <oxek> and of course the 'defaults' option is not documented anywhere for the 'swap' type
1721[17:46:50] <cipherize> Because I don't believe it applies.
1723[17:47:15] <oxek> cipherize: debian-stable default installation creates a swap partition that has the 'sw' option in /etc/fstab
1724[17:47:17] <cipherize> The default mount option for 'swap' is ... nothing. It allows the kernel to assign a pri value, and swap implies 'auto' unless 'noauto' is specified.
1745[17:57:32] <cipherize> Basically, you want to ensure that the space for the swapfile is fully, actually allocated, not just preallocated. That's what -z is for.
1746[17:59:02] *** debhelper sets mode: +l 1169
1747[17:59:45] <lle-bout> looks like web package information debian website is down - e.g. replaced-url
1778[18:11:41] <genr8_> i must be doing something right, cause it looks normal to me.
1779[18:12:35] <sney> haha. yesterday. time flies. assuming everything goes normally, it'll be in testing late next week, and backports shortly after. replaced-url
1795[18:22:24] <jhutchins> RedHat preserves an install log that can tell you when the system was built. Does Debian have some sort of file that will show by it's date when the system was built or upgraded?
1798[18:24:18] <JyZyXEL> kernel prints the build info into dmesg during startup?
1799[18:24:32] <JordiGH> Uhm, I have some .script and .time files in my root homedir that I created during installation because the release notes told me to.
1800[18:24:43] <JordiGH> But it was a manual thing. It's a full log of the upgrade process.
1801[18:24:53] <JordiGH> I suppose not at all what you're asking.
1806[18:29:11] <jsync> (Cool people are "Cool". 😉) "Here Here" sounds like stuttering. Troll-Name-Calling is a Crime. How were "right" & "wrong" considered amidst a coherency test?
1807[18:29:31] *** Quits: chiluk (~quassel@replaced-ip) (Quit: No Ping reply in 180 seconds.)
1808[18:29:55] <genr8_> Upgraded = timestamp of /etc/debian_version (and contents is the build number - 10.5)
1819[18:35:17] <jhutchins> I have a speech-dispatcher log that's the oldest file in /var/log on this laptop, but on my server nothing seems to be as old as I know the hardware to be.
1826[18:37:43] <jhutchins> genr8_: Yeah, but /var/log/speech-dispatcher is older.
1827[18:38:13] <genr8_> thats not a valid pick of whats older
1828[18:38:38] <jsync> ("Dance of the Clairvoyants")
1829[18:39:18] <Mazhive> i have a question i am working on a xinit/xstart run 1 program from boot into a specific user iuse the .bashrc to start the program i wander if the output from a xterm could be blinded i get a whit console starting up my program for a few seconds and confirming it running and the program goes in fullscreen covering the xterm can i use xterm arguments to minimize or run it in the background
1830[18:39:23] <genr8_> unless you did something ridiculous like re-create your whole root filesystem, it should be easy to find
1831[18:39:54] <genr8_> the symlinks in / should not have changed.
1832[18:39:57] <joepublic> Mazhive, wmctrl can manipulate windows for you
1838[18:41:14] *** Quits: jsync (~nosaj@replaced-ip) (Quit: It's my role, in this life, to express my perspective. It's the role of others to ensure that I, as well as others, actually are safe in doing such.)
1866[18:51:08] <Mazhive> openbox true ... yet another step to account with..maybe someone can look into my tutorial i wrote i need feed back to adress my needs and problems
1867[18:51:32] <Mazhive> eventually from this tutorial i intend to auto mate this
1881[19:00:21] <no_gravity> Hello! The screen of my PinePhone with Mobian randomly wakes up. Any ideas how to debug it?
1882[19:00:23] <Mazhive> it is my first attempt to make things easier for people who used to know a os (Motorola 680x0) which is not "popular" by the gross people, this script/tutorial is mainly goaled on enthousiast still wanting to do more with it by giving the community the tutorial/script
1907[19:15:46] <jhutchins> mtnman: wicd is prety much deprecated. At one point it worked better than Network Manager, but NM has long since surpassed it.
1912[19:18:44] <mtnman> i asked there (box dual boots debian/devuan) they said wicd is deprecated in next release. suppose i will try networkmanager thanks.
2027[21:08:47] <tmroland> in terms of memory management on linux, disk blocks are buffered and cached in memory pages (slabs) which make up the cache right?
2046[21:25:25] <greycat> Is there any specific reason *why* you want a higher-numbered firefox-esr package? How many things is it going to break?
2047[21:25:58] <NetTerminalGene> ten version bigger than the old one
2048[21:26:19] <NetTerminalGene> it must have a lot of impovements
2049[21:26:25] <greycat> *plonk*
2050[21:27:26] <greycat> I keep hoping that when I say things like "the sum of your version numbers is not your score; it does not determine your worth", that I am exaggerating. Alas.
2051[21:27:28] <NetTerminalGene> i think it has revamped addressbar
2052[21:27:54] <NetTerminalGene> and maybe vaapi accel
2053[21:27:57] *** Joins: Kel (~KelMonsta@replaced-ip)
2073[21:33:26] <EdePopede> acceleration is a nice thing to have (or probably rather would be in my case *g*), and then imgur, and as it was mentioned, also other sites.
2075[21:33:57] <EdePopede> an idea which version is minumum now and if there's a difference to upstream?
2076[21:34:35] <EdePopede> NetTerminalGene: seems they changed something breaking older browsers. at least my old firefox and what i've read before 68 in general?
2077[21:34:45] <EdePopede> <JordiGH> genr8_: it's possible to opt out of imgur's new design, but yeah, the new design just shows nothing on Firefox 68
2078[21:34:50] <greycat> The other day, we had a discussion in here, where we were trying to figure out when support ends for ESR 68. Turns out, someone found a web site where they said ESR 68 support will end when 78.3 is released, and the release date of that was 2 days ago (allegedly).
2080[21:35:30] <greycat> imgur lost my support when they demanded that I log in before I could upload an image. Nope.
2081[21:35:40] <EdePopede> i always had issues keeping an eye on the relevant mozilla pages, they like to redesign their site for every new EST it seems
2082[21:35:56] <EdePopede> i remember it, didn't they take it back later?
2083[21:36:18] <ratrace> imgur lost _my_ support when they started requiring resources from sites firefox and my dns blacklist were blocking.
2085[21:37:46] <EdePopede> and yeah, most of the modern/commercial web is broken in that regard these days. youtube shouts at me every time i return there to login, pinterest lets me scroll down maybe a page, quora doesn't allow f'up reading for years already.
2086[21:38:54] <EdePopede> right, and even wikipedia seems to be going mobile first now. i want the 90s back!
2087[21:40:53] <NetTerminalGene> i wonder if you guys use firefox's default scrolling step size?
2088[21:41:17] <NetTerminalGene> default one is very small step
2126[21:50:21] <genr8_> everyone keeps saying it doesnt.
2127[21:50:25] <genr8_> it does.
2128[21:50:26] <EdePopede> wtf, here too now. didn't for weeks until just minutes ago.
2129[21:50:37] <EdePopede> oh, stop.
2130[21:51:01] <EdePopede> a lot of the images on the startpage just show the broken image icon in their rectangl
2131[21:51:05] <NetTerminalGene> genr8_: why do you use privacy badger with ublock?
2132[21:51:38] <genr8_> idk
2133[21:51:51] <oxek> EdePopede: what's the value of layout.css.resizeobserver.enabled?
2134[21:52:03] <genr8_> you should ask me why does PrivacyBadger still show that its blocking stuff, if its not needed ?
2135[21:52:14] <EdePopede> browsers are so complex these days, and w3c/whatwg are releasing so fast it may be easier maintaining an OS than a browser
2136[21:52:31] <n4dir> lol.
2137[21:52:42] <EdePopede> oxek: i'll check, may be i don't have it in this version yet. not running the default profile atm.
2138[21:52:53] *** Quits: ledeni (~ledeni@replaced-ip) (Remote host closed the connection)
2139[21:53:02] <oxek> default is false on debian, set it to true to fix imgur
2140[21:53:48] <genr8_> mine is false... still works. every image on frontpage shows up.
2141[21:54:09] <oxek> genr8_: but do the actual images load if you click them?
2142[21:54:11] <ratrace> genr8_: but what if oyu click on a post?
2143[21:54:33] <genr8_> ah. yes its broken then
2144[21:55:49] <genr8_> well the layout.css.resizeobserver.enabled works. thanks for figuring it out
2145[21:56:04] <EdePopede> oxek: nope. had to add it, but to no effect. i'll check for the version at which it starts to work ;)
2146[21:56:38] <oxek> genr8_: mozilla flipped it to default true in firefox 69, so imgur figured it has been 11 months since that release so it was safe to sue
2163[22:00:37] <EdePopede> or the old story with the reflecting displays. sure, all this LOOKS good in a shopping window, but the moment you have to use it all you want is your money back.
2165[22:01:55] <EdePopede> greycat: it's like i refuse to call certain... products press or the people writing for them journalists. this would actually be an offence against people deserving that title.
2167[22:05:32] <NetTerminalGene> new version firefox has gray padlock >.<
2168[22:07:57] <EdePopede> always a good idea to change the look of interface elements holding relevant information not meant to interrupt the workflow by investigating them
2235[22:46:21] <genr8_> Ill touch what I want. no harm anyway, new version comes in 3 days
2236[22:47:03] <oxek> genr8_: the upgrade does not delete any extra files you might have added to the /usr/lib/firefox dir, and firefox can still load those files
2237[22:47:47] <oxek> on apt purge you should get a warning though that the dirs are not empty hence not deleted
2238[22:47:57] *** Quits: fflori (~fflori@replaced-ip) (Remote host closed the connection)
2239[22:48:14] <oxek> I had to do a apt purge of firefox recently to reinstall it, and it was not trivial
2248[22:54:08] <nvz> I've only glanced in twice today and didnt really /read/ anything per se and this is the 2nd time I've noticed mention of firefox maintainers
2257[22:55:54] <oxek> joepublic: you could try emailing them team+pkg-mozilla@tracker.debian.org
2258[22:56:12] <KOLANICH> oxek: I just want the official package be built with some flags. I mean that one of the features (or a lack of antifeature, because it is also available in regular editions, just reserved fkr Mozilla exclusive use) reserved by Mozilla for devedition is missing.
2259[22:56:47] <greycat> so, file a wishlist bug
2260[22:56:55] <oxek> KOLANICH: what flag might that be?
2261[22:59:34] <KOLANICH> oxec: some of the ones enabling MOZ_DEV_EDITION macrodef
2293[23:12:22] <alex11> though firefox esr is going to come out in debian soon as 70 something or other
2294[23:12:55] <alex11> 78 i guess
2295[23:13:02] <alex11> which is not what you want :P
2296[23:13:15] <jim> it's one behind :)
2297[23:13:16] <KOLANICH> alex11: I cannot distribute my addon to people saying "don't install firefox from packages, install from Mozilla"
2298[23:13:33] <oxek> KOLANICH: direct them to the flatpak version of firefox
2299[23:13:51] <KOLANICH> oxek: flatpack and snap is shit
2300[23:13:59] <KOLANICH> *are
2301[23:14:28] <nvz> jim: it seems to be being held up by transitions that effect numerous other packages
2302[23:14:44] <oxek> KOLANICH: then you're gonna have to direct users to the mozilla download pages, tell them not to get the packages and instead get the tars, and to extract them in correct locations and setup links
2303[23:15:14] *** Quits: ElNomReal (~E@replaced-ip) (Remote host closed the connection)
2304[23:15:53] <KOLANICH> oxec: is there any reason not to distribute the binaries with the feature enabled?
2305[23:15:54] <jim> oh ok
2306[23:16:08] <KOLANICH> *the official binary packages
2315[23:20:45] <KOLANICH> oxek: Mozilla imposes too much policy in their decisions. Their decision is that the API to set about:config prefs won't be exposed. In fact their course is on disabling about:config, they have done it in Fenix (and dropped Fennec). We have to counter.
2319[23:21:52] <JordiGH> What are we gonna have instead?
2320[23:21:52] <oxek> JordiGH: they did that on android
2321[23:21:59] <JordiGH> wtf wtf wtf
2322[23:22:00] <JordiGH> why
2323[23:22:15] <oxek> KOLANICH: we're safe for another year, ff78esr is nearly here
2324[23:22:44] <KOLANICH> WebExtensions ex0eriments is a semi-internal very poorly documented unstable stuff. But at least it is a way to add an own API without forking and building Firefox ourselves (which is very long and resource-consuming).
2325[23:22:52] <KOLANICH> *experiments
2326[23:23:03] <oxek> we have a year to find a different browser if mozilla does something anti-user that would strongly go against the debian social contract
2327[23:23:26] <pingfloyd> mozilla's decision makers have been too many hard drugs
2328[23:23:49] <oxek> JordiGH: telemetry showed that no one used about:config on android
2329[23:24:05] <JordiGH> oxek: So that means there's no reason to do anything about it.
2330[23:24:06] <pingfloyd> seems like it has become a trend with mozilla to make terrible decisions
2352[23:27:57] <pingfloyd> JordiGH: yeah, it's a pretty big support burden
2353[23:28:04] <pingfloyd> JordiGH: and firefox is such a fast moving target
2354[23:28:31] <JordiGH> Firefox is bigger than Debian.
2355[23:28:33] <JordiGH> Bigger than Jesus.
2356[23:28:36] <JordiGH> It's a huge beast.
2357[23:28:45] <KOLANICH> oxek: Mozilla is doing anti-user activity about 10 years. Noone can counter because developing a modern web browser is very expensive. Chromium is subsidized entiry by google and was created with the sole purpose - to take over Firefox, gec tontrol over the ecosystem and turn the ecosystem the way beneficial for Google
2358[23:28:58] <oxek> pingfloyd: it really isn't that fast-moving. The only real change between 68esr and 78esr is flipping some options (that already existed in 68esr) and changing the url bar.
2359[23:29:04] <pingfloyd> it all kind of started to go downhill around the time quantum came out
2360[23:29:28] <JordiGH> oxek: Well, I assume imgur works in 78esr
2361[23:29:37] <oxek> JordiGH: imgur works in 68esr too
2362[23:29:39] <JordiGH> So I assume that's more than options.
2363[23:29:41] <pingfloyd> some good addons were lost forever too because of that
2364[23:29:46] <JordiGH> Only if you opt out of the imgur beta.
2365[23:29:56] <oxek> JordiGH: no, it works even without that
2366[23:30:12] <JordiGH> Oh? How do you make imgur not show a blank grey page in 68esr?
2379[23:33:33] <KOLANICH> oxek: I know only 3 forks that got any significant traction. 1 TorBrowser. Centered around privacy. Follows mainstream firefox. 2 PaleMoon. Centered aroun non-supporting webexts and all modern stuff. Ideologically wrong in the sense that Webexts is a very right approach, the wrong what Mozilla has done was to impose politics into them. Refusing to use webexts is a big step back. 3 IceCat - centerd around FOSS license enforcement.
2380[23:33:41] *** Quits: treeview2 (~treeview@replaced-ip) (Remote host closed the connection)
2387[23:36:37] <KOLANICH> jhutchins: About Debian support. I think it is necessary to enable some devedition features in the `firefox` package built and shipped by debian.
2388[23:37:12] <greycat> Then file a wishlist bug.
2389[23:37:42] <jhutchins> This is not a philosophy channel.
2390[23:37:42] <KOLANICH> greycat: it has already been filed.
2396[23:39:37] <pingfloyd> jim: I just create a symlink to it in ~/bin
2397[23:40:11] <pingfloyd> jim: and then have ~/bin in your path. Or alternative a symlink in /usr/local/bin.
2398[23:40:29] <jim> pingfloyd, ok... which one of /opt/firefox? /opt/firefox-bin?
2399[23:40:41] <pingfloyd> jim: to /opt/firefox/firefox
2400[23:40:49] <KOLANICH> jhutchins: I am not about philosophy here. I just want to know if is there any real (renaming firefox to iceweasel is OK IMHO, the benefits outweight the problems caused by renaming, I also think that Mozilla may want not to demand renaming because it would mean further drop 0f Firefox market share) reason not to enable the feature.
2401[23:40:56] <jim> oh ok thanks
2402[23:41:12] <pingfloyd> jim: there should be the firefox directory under /opt
2408[23:43:23] <pingfloyd> jim: one thing I further do is find /opt/firefox/ ! -user root; find /opt/firefox ! -group root (and there's actually a reason for this)