5[00:06:28] <sponix> is there an easily established method to backport from "experimental".. Or do I need to give it a bit until it makes it into Sid ?
6[00:06:46] <sponix> I'd like to run kernel 5.10 with zfs, and Nvidia 450 if possible
7[00:07:19] <sney> experimental packages will generally install on sid or testing, provided they aren't completely broken
8[00:07:31] <sney> (there is no guarantee of any quality wrt experimental)
9[00:07:34] <dvs> !v linux-kernel-amd64
10[00:07:41] <dvs> ,v linux-kernel-amd64
11[00:07:42] <judd> No package named 'linux-kernel-amd64' was found in amd64.
12[00:08:00] <sponix> dvs: I know 5.9 is the latest in backports for buster
16[00:08:11] <sney> there are some patches for zfs to compile on 5.10 but you should know that openzfs does not consider it supported at all, yet
17[00:08:24] <sney> this includes 2.0.0
18[00:08:36] <sney> and finally,
19[00:08:39] <sney> !debian-next
20[00:08:39] <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
31[00:11:17] <sponix> I should probably just wait for zfs, nvidia, and kernel backports all to support 5.10 before I make the jump. Got a bit excited and went to early :)
32[00:11:20] <sponix> !sns
33[00:11:20] <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>.
34[00:11:25] <sney> backporting from experimental to stable is basically the same procedure as backporting from sid to stable. (again with the caveat that there is no guarantee of any quality in experimental.) OR you could use bullseye and just install the packages.
35[00:14:29] * sney uses bullseye, with zfs and nvidia, and is still on 5.9 for the moment
36[00:14:56] *** Quits: mirrorbird (~psutcliff@replaced-ip) (Remote host closed the connection)
58[00:22:21] <sponix> I just sometimes get the SNS :)
59[00:22:38] * dvs cures sponix
60[00:22:54] <ratrace> hmm look at this... it even says that 5.10 is blocked because it breaks, among other things, zfs ... replaced-url
61[00:22:59] <sney> out-of-tree module breakage is a pretty normal thing to deal with on testing anyway. this is why we have apt-listchanges and apt-listbugs and the ability to pay attention
62[00:23:00] <ratrace> so how did bullseye get 5.10 then.
63[00:23:47] <ratrace> sney: bah
64[00:23:54] <sney> probably the same way it did when 5.9 broke nvidia and virtualbox
67[00:25:39] <\\Mr_C\\> i keep getting an error with a clean install of debian -bash: mkdiskimage: command not found
68[00:25:42] <ratrace> I used to have the bpo'd kernel on stable until recently, but I've switched to tracking kernel.org's 4.19 instead as 5.9 EOL'd. 5.10 has a nasty btrfs regression.
69[00:25:46] <\\Mr_C\\> i installed the latest syslinux
70[00:25:52] <\\Mr_C\\> any ideas?
71[00:25:53] <sney> ,file mkdiskimage
72[00:25:58] <judd> Search for mkdiskimage in buster/amd64: syslinux-utils: usr/bin/mkdiskimage
103[00:45:10] *** Quits: Jad (~Nashmi@replaced-ip) (Quit: Benefits I derive from freedom are largely the result of the uses of freedom by others, and mostly of those uses of freedom that I could never avail myself of.)
116[01:06:55] <sney> struck by curiosity, I took the risk, and can happily report that 5.10.0-1 and zfs-dkms 2.0.1-1~exp1 are running just fine on bullseye. so I'm sure buster-backports will follow before *too* long.
129[01:16:43] <zykotick9> heh... sney "report that 5.10.0-1 and zfs-dkms 2.0.1-1~exp1 are running just fine" <- so what you're sayin' is "I didn't just have to downgrade my kernel in bullseye" ;) install zfs-dkms from experimental is the workaround, I take it.
130[01:17:06] <ratrace> I also use LUKS which is suffering a LOT due to meltdown and spectre mitigations. phoronix ran a test, and inspired by it, I tried to reboot a test machine with mitigations=off and indeed.... 2x difference
132[01:17:46] <ratrace> _thankfully_ still outside of sata3 bandwidth saturation, I gots strong CPUs :) but eh.... that plus btrfs atop of it ..... 5.10 is a disaster for me, right now.
137[01:18:08] <sney> zykotick9: I'm using 2 different out-of-tree modules so I always check the weather before I upgrade the kernel. but sure, same idea
139[01:19:05] <ratrace> sney: but it's pretty safe on stable. debian will err on the side of "KILLALL PERFORMANCE HERPDERP!" before breaking zfs.
140[01:19:27] *** Quits: tagomago (~tagomago@replaced-ip) (Remote host closed the connection)
141[01:19:29] <ratrace> like last year, 0.7.x was patched to disable SIMD :)
142[01:20:02] <Souler> Welcome. Which gentlemen or ladies would help me for priv? It is specifically about installing drivers for the NVidia Quadro k2100m graphics card on the ThinkPad W541. I'm not sure, but I think the graphics should work better. You would have to check if everything is as it should be.
143[01:20:20] <ratrace> ahead of 10.1 or 10.2 release ... can't remember now :: because thanks to GKH .... the political patches were backported to LTS kernels in the middle of their releaes cycle..... FUN FUN FUN FUN
144[01:20:31] <sney> Souler: that's not how things work here, just give details in the channel and we can help
145[01:20:48] <sney> !paste
146[01:20:48] <dpkg> Do not paste more than 2 lines to this channel. Instead, use for text: replaced-url
147[01:21:24] <ratrace> (last year .... 2019 ... my brain is still in 2020)
190[01:37:07] <dpkg> Binary-only firmware for the nouveau <DRM> driver is packaged for Debian as firmware-misc-nonfree. Without this package installed, poor 2D/3D performance and/or missing video features in the <nouveau> xorg driver is commonly experienced. To install, ask me about <non-free sources>.
193[01:37:48] <ratrace> Souler: that only shows apparmor profile being loaded. what's the actual problem you're having? "I think [gpu] should work better" is very vague and subjective
195[01:38:31] <ratrace> if you have poor game performance, you'll need the proprietary driver. nouveau is still missing the only firmware that matters :)
211[01:47:39] <seanthegeek> EdePopede tomreyn sney dvs I'm finally back. Thanks for testing from your ends. I just switched to a mobile hotspot for troubleshooting, and apt worked. No idea why it's failing on my local network. I've never had this issue before. It persists even after clearing DNS cache on my gateway and local system. Really weird.
212[01:49:05] <sney> seanthegeek: did you try non-cdn mirrors? the fastly cdn sometimes behaves strangely, particularly if your ISP is doing any cache/proxy tomfoolery behind the scenes
227[01:56:06] <Souler> I restarted the computer and now I am upgrading. I saw that he wants to install drivers from ati and radeon. I'm not sure if it's good. It will download, install and then see if Blender's performance is better.
234[02:00:17] <jmcnaught> Probably the "Possibly missing firmware…" messages during initrd creation.
235[02:01:54] <ratrace> bah.... seems like we won't see PHP8 in Bullseye....... :(
236[02:02:46] <ratrace> been reading through the transitions and related blockers bug reports.... and not just that; PHP software just ain't yet ready for 8
246[02:12:36] <Souler> I checked the efficiency of video editing in Blender. Performance is the same after the system upgrade. Should I do something else besides installing firmware-misc-nonfree? Maybe I should load some mod?
248[02:14:26] <ratrace> Souler: maybe you just need a stronger computer and/or GPU
249[02:15:06] <ratrace> Souler: but eh, if that's nvidia GPU (you mentioned ATI at some point too), you could try the proprietary driver instead of nouveau
258[02:19:50] <sney> really? the gpu in my plex instance is a GK107 and that's supported by nvidia-current, this should be close if not the same
259[02:19:53] <Souler> This is a new computer that I bought a few days ago. I just wondered if it could work even better than it works now. Currently he is a bit stuck as there are transitions in video editing. I thought I didn't have the correct drivers. Maybe the blender has too high requirements. I don't know how to approach this.
260[02:20:13] <ratrace> sney: nvidia . com says 319 for K2100M
261[02:20:35] <ratrace> oldest we have is nvidia-legacy-340xx
271[02:23:26] <sney> Souler: make sure you install the linux-headers-amd64 package too
272[02:23:42] <ratrace> zykotick9: how do you search for it exactly? the nvidia search tab is broken for me (resets the dropdown if I select Quadro from the list), so I tried via google, and 319 is all I got
276[02:26:41] <ratrace> zykotick9: that's some other search form then. that one works for me too, and actually lists 460 as well. thanks for correcting me
277[02:28:45] <ratrace> actually....no, it doesn't list K2100M at for newer drivers
294[02:37:17] <dpkg> Your screen is in front of your face. #debian cannot see it. Please give as many details as you can about the problem. It wastes our time and yours when we have to guess what is on your screen. See <context>, <what>
345[02:56:48] <jennie> thats when i installed ubuntu , when i asked about brightness issue in debian
346[02:56:55] <ratrace> ANYWAY .... this channel is usually busy and populates. there may be pockets of silence. so, if you have an issue, ask away and have a bit patience please.
352[02:57:34] <sney> #debian is staffed by volunteers, so there isn't always someone here. but I'm here all the time (thanks covid) and I haven't seen you either. maybe just an inconvenient time zone. :<
353[02:57:51] <jennie> no, as I insatlled ubuntu, but if someone is willing to resolve, it wont take more than 10min to install debian back
354[02:57:53] <ratrace> the channel is fairly busy
355[02:58:13] <ratrace> jennie: you'll have to come with actual problem on actual debian and with ready logs and pastes.
358[02:59:37] *** Quits: OEIRAS (~OERIAS@replaced-ip) (Remote host closed the connection)
359[02:59:37] <ratrace> but even then.... there's no guarantee of any response. if someone knows, they'll pipe up. sometimes you have to be persistent and repeat your question as people come and go, and timezones change
360[02:59:57] <jennie> any other platform to ask questions than this channel?
361[03:00:11] <ratrace> there's debian channels on OFTC
362[03:00:15] <sney> #debian on oftc, various unofficial forums
363[03:00:21] <ratrace> mailing lists....
364[03:00:28] <sney> ^
365[03:01:10] <sney> if you aren't inclined to wait around on irc for an answer, a ml or forum may be more suited anyway
374[03:04:11] <jennie> i like it thast why i dont want to stick with ubuntu
375[03:04:52] <ratrace> jennie: again there is no guarantee for anything. maybe someone can help you right away, maybe not. no idea. personally, I'm not desktop oriented, I'm way more useful for servers, filesystems.
378[03:06:34] <ratrace> (though I can help with desktop-y stuff that I personally use, like the nvidia issue someone had before you .. screen brightness ain't among those, I don't use a DE, and don't alter the brightness)
380[03:08:00] <Souler> So this is: nvidia-detect told me that the nvidia 418 tesla driver would still be working. I have installed. After its installation, the blender started working, but the performance is as it was at the beginning. I decided to remove everything from nvidia for now and keep only novoveau. I think I installed too much at once. Write me step by step what I should install.
384[03:09:53] <ratrace> Souler: 418, so that's the nvidia-driver from buster/non-free ? that's it then, if that fixed your issue. there's nothing more you can do to improve performance, with gpu drivers alone
396[03:21:47] * unborn "Write me step by step what I should install" - uhm no 'please...' word? oi Souler give me million of euro so I could live forever. whats wrong with this world
398[03:27:08] <unborn> ratrace: Im sitting here couple of years - past month or two no recall of jennie no slow support at all. this channel is great ! and I think some people are too nice. anyway - sorry for making noise but millennial's without manners piss me off.
403[03:29:42] <sney> nice, a non-sequitor shot at millennials when we're the last generation of die hard irc users. I'll try to forget that next time you have a question.
406[03:31:17] <unborn> sney: I am in theory millennial as well :) but those very early born expect too much for free everywhere.
407[03:31:22] *** Quits: H-var (H-var@replaced-ip) (Remote host closed the connection)
408[03:31:48] <sney> I'm 36 and I've lost most of my hair, and I still can't escape the attitude that I'm somehow a child with no manners. no thanks. change your approach.l
410[03:33:23] <Souler> So yeah: I removed everything from nvdia, restarted my computer, installed nvidia 418 tesla driver. Everything works, but I haven't noticed any great improvement. Are there any other packages from nvidia that we will need and possibly improve performance? NVidia Settings or something else?
412[03:34:24] <unborn> sney: well you are my age.. am 40 badly - I wont change my attitude - knowledge takes decades and I see no reason for rude people too lazy to learn to just get what they want just because they want it.. decades are not free even open source is not free, someone had paid for it with his own time.
413[03:35:04] <sney> Souler: it sounds like you just have an old gpu, and you've done the best you can with it.
416[03:40:42] <cybercrypto> Thats quite interesting... I am using irc a bit after BBS got 'killed' and I like it. Nowadays, it is quite common that people ask for other 'support channels' for speed of course... and I think it is not only yong people. :-)
418[03:43:56] <Souler> sney: Maybe you are right. From what I can remember, Blender always performed a bit worse for video editing. Maybe I got everything I could out of the graphics and the program itself is simply not efficient enough. Most likely, I will change my video editing program and that will be the end of it. Do you recommend any good video editing program? I'm shooting a music video soon and I should be keeping an eye on something by now.
419[03:44:33] *** Quits: milkt (~debian@replaced-ip) (Remote host closed the connection)
420[03:44:33] *** Quits: filohuhum (~filohuhum@replaced-ip) (Remote host closed the connection)
423[03:45:36] <sney> Souler: I use kdenlive for video editing, I hear openshot is good. That stuff will generally be smoother than 3d modelling with older hardware too.
430[03:52:24] *** Quits: blkmanta (~quassel@replaced-ip) (Quit: Escaped into the void.)
431[03:54:44] <Souler> I also think Kdenlive is good, I used it the most. Has a lot of plugins. What only irritated me is that there are few paths in it. Unless I can't add more.
721[09:53:49] <shtrb> I just don't understand why my speed (even with speedtest-cli goes now to 10 Mbit , when using the same cable on a windows machine from WSL give me 60 MBit)
750[10:26:10] <mathsaey> Hey everybody. I'm trying to set up a raspberry pi (running raspbian) as a bridged AP. I've managed to do this on my home network without issues. However, when I plug in the pi on the network where I actually want to use the pi as AP, nothing works. The pi is assigned an ip by the router (I can ssh to it), but it does not seem to have a correct gateway set up.
751[10:27:14] <mathsaey> I'm guessing there is some difference in dhcp setup between the local network and the target network. I don't know that much about networking, so if anybody could give me some pointers I'd be very grateful :).
753[10:30:01] <aminvakil> mathsaey: you should be more specific on changes, what are you trying to do? what do you expect? what do you see instead? and ask it in its proper channel :) i'm not sure debian is the one
755[10:31:43] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
756[10:33:51] <mathsaey> Ah, I figured #debian would be fine since most of this is not raspberry pi specific. But I'll look around for a raspbian specific channel :). Besides that, I'm not 100% sure which additional information you are asking for. Running ping 1.1.1.1 on the pi returns "network is unreachable", while everything worked fine at home (i.e. the pi had internet and any devices connected to the pi over wlan0 had
757[10:33:53] <mathsaey> internet as well).
758[10:34:20] <mathsaey> But anyway, I'll look around for a more suitable channel, thanks for replying and telling me!
759[10:37:06] <aminvakil> i don't think your question is specific to the os you're running, and it's purely network-related.
760[10:37:49] <aminvakil> also for network troubleshooting don't start with pinging 1.1.1.1, first see if you can reach yourself (localhost), then gateway, then outside of gateway
761[10:39:57] <mathsaey> I should have mentioned that, pinging localhost and the gateway work with no issues. Since dns is down I just used 1.1.1.1 to see if I could reach anything outside my network, which I can't.
763[10:40:29] <mathsaey> And right, I asked here since I figured the packages and the default configuration is basically the same between raspbian and debian.
767[10:44:53] <quadrathoch2> mathsaey nope, the raspberry pi foundation do change a few things, so we in #debian have no clue which of those changes are. that's why it makes more sense to ask them
772[10:46:28] <dpkg> Raspberry Pi OS (previously called Raspbian) is a distribution <based on Debian> made specifically for the <Raspberry Pi>. Raspbian is not Debian and it is not supported in #debian. Please use #raspbian (or #raspberrypi) on irc.freenode.net for support. replaced-url
776[10:58:05] <freem> Hello. Got an "interesting" behavior at this boot: acpid was flooding "inotify fd eof encountered" into logs, making CPU run at 100% (and the log daemon, of course). Restarting it gives:
777[10:58:05] <freem> starting up with netlink and the input layer
778[10:58:05] <freem> 27 rules loaded
779[10:58:05] <freem> waiting for events: event logging is on
787[11:05:05] <freem> got a problem since previous boot with acpid-2.0.31 (in debian 10). It spits "inotify fd eof encountered" to the logs, using a full core. When I restart it, the problem "disappear" and gives replaced-url
788[11:05:46] <freem> I don't think I've updated my system yesterday
822[11:27:37] <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
829[11:33:27] <themill> [sr]: from the changelog, the init script was buggy, didn't work, and couldn't be fixed. If you're running sid, you really really should be reading the changelogs and also NEWS.Debian for the packages, both of which mention this change.
830[11:33:39] <themill> !apt-listchanges
831[11:33:40] <dpkg> apt-listchanges is a tool that displays NEWS.Debian file contents (default) and/or the changelogs of Debian packages before upgrading them (dpkg-reconfigure apt-listchanges). It is recommended for users of stable to install the apt-listchanges package, to see what is happening to their systems as security fixes are applied; it should be made _mandatory_ for users of testing and unstable. Also ask me about <d-d-a>, <apt-listbugs>.
838[11:41:51] <shtrb> Please correct me if I'm wrong , in case of (hardware) raid 5 single disk failure , there filesystem nor any running application should see any effect at all.
841[11:42:14] <shtrb> The FS would have an inidication ?
842[11:42:31] <petn-randall> shtrb: nope
843[11:42:33] <Mr_Queue> will march right along until you lose another one and then everything is tits up.
844[11:42:53] <Mr_Queue> shtrb: ^^
845[11:43:33] <shtrb> petn-randall, do you mean I was originally wrong (that the filesystem will have an indication ) or that I was correct originally that the fs should not have any indication at all
846[11:43:56] <Mr_Queue> You will notice when you replace the drive an the array rebuilds.
847[11:44:17] <petn-randall> shtrb: RAID and the filesystem are completely independent. The filesystem will notice *nothing* about a disk fail in RAID5. That's intended.
848[11:44:42] <petn-randall> shtrb: You will however notice a disk fail, since you're hopefully monitoring the health of the RAID array.
849[11:45:11] <shtrb> petn-randall, thank you !
850[11:45:15] <Mr_Queue> SMART can email you, or use something like Zabbix to watch the arrays..
851[11:45:56] <Mr_Queue> Something to think about.... The I/O thrashing that goes on rebuilding an array can drop more drives from the array.
854[11:46:53] <petn-randall> Mr_Queue: That would be a serious design error, and something I haven't seen in neither HW nor SW RAID.
855[11:47:15] <ratrace> shtrb: 2021 called, they're asking why are you not using software raid yet ;) :)
856[11:47:44] <petn-randall> ratrace: SW RAID doesn't have writeback cache :P
857[11:48:29] <Mr_Queue> petn-randall: If you have an array with 5 drives with 60,000 hours on each because that when you built it... If you lose a drive, then thrash the rest rebuilding the new on in...... You are right there in the danger zone.
858[11:48:41] <petn-randall> Btw, you can set I/O priority, and bandwidth limits for rebuilding on most RAID.
859[11:48:51] <ratrace> petn-randall: but the kernel does :)
860[11:49:35] *** Quits: T-zef (~tyzef@replaced-ip) (Remote host closed the connection)
862[11:50:05] <petn-randall> ratrace: writeback cache is not the same as dirty pages. The former gives you a huge performance boost on things like fdatasync().
865[11:51:15] <Mr_Queue> I don't think hw raid is really needed these days. You should have more than one array to read and write from and disk cache in memory should be the fastest anyhow.
866[11:51:33] <ratrace> petn-randall: so is ZFS SLOG ;)
867[11:52:18] <petn-randall> Mr_Queue: You can't have write cache in memory.
871[11:54:08] <ratrace> anyhoo, those few battery backed (or else bye, mr. data) MB of write burst dampeners is cute. insufficient "discount" on the otherwise high "price" of hw raids ;)
872[11:54:34] <Mr_Queue> cache-cade or what megaraid calls it. Expensive for what you get.
874[11:55:23] <petn-randall> Well, if you have a dedicated device for writeback cache you're basically just reimplementing what a hw raid does.
875[11:55:44] <ratrace> have y'all seen the performance of a nvme SLOG? that's a beast... multi-GB beast. eats the BBU multi-MB hw-raid cache for breakfast.
887[12:00:48] <Mr_Queue> bifurcation across the lanes really.. I think today's PCI is a little better but you still need to cut it up a bit..
888[12:01:08] <Mr_Queue> I didn't care, the performance I needed was there.. There were a total of 3 arrays configured for HA so it wasn't a big deal to lose one, or even two.
930[12:33:09] <dragestil> thanks. but is a nonfree wireless firmware less secure than a free wireless firmware, in that the former can log your packets etc?
931[12:33:24] <noisemaker> ratrace: I will take a look, ty
932[12:33:47] <freem> dragestil: both FOSS and non-FOSS can log stuff. The only security is when there is an actual audit.
933[12:34:53] <freem> I can imagine highly secure yet non-FOSS code, and FOSS code leaking like a sieve. Without much efforts.
934[12:34:54] <dragestil> does being in debian firmware-nonfree make it less likely to log stuff?
935[12:35:03] <shtrb> dragestil, I would say that audited coded is more secure than non audited
936[12:35:35] <shtrb> dragestil, no, your running firmware can do lot's of bad stuff
937[12:35:38] <freem> and to be clear, I can also imagine highly safe FOSS code and non-FOSS code leaking like a sieve. Easily, too.
938[12:35:45] <dragestil> ...less likely compared to nonfree firmawares not in debian firmaware-nonfree?
939[12:36:18] <dragestil> alright thanks
940[12:36:58] <dragestil> but is stuff in debian firmware-nonfree audited in any way?
951[12:39:19] <dragestil> petn-randall: argh, that sucks
952[12:39:20] *** cdown_ is now known as cdown
953[12:40:11] <tuxmania> Hi all. Trying to recover an old application installed on Debian Squeeze (6.0). It seems that I can no more access squeeze repositories. When doing an apg-get update, it gets stucked on "Waiting for headers".
954[12:40:13] <freem> dragestil: you may use CVE databases to evaluate how much you can trust something I guess. Or try to break it yourself, but that's... well, *not* easy
955[12:40:21] <tuxmania> Should I change the repositories?
957[12:41:14] <freem> also, there are people which sell hardware they claim with as little as possible close firmwares. But that's usually *not cheap* if you compare performances
958[12:41:21] <tarzeau> tuxmania: what is the applications package name?
961[12:41:53] <dpkg> Volunteer-provided security support for Debian 6.0 "Squeeze" ended on 2016-02-29 for a limited set of packages for i386 and amd64. An appropriate line for your sources.list is «deb replaced-url
976[12:44:57] <freem> for example, you might use puri.sm 's products. From what I've read, they try really hard to make sure 1) hardware is as FOSS as possible and 2) can't phone home if you don't want it (physical switches)
977[12:45:16] <freem> there are probably lot of other solutions around, but that's a niche market
978[12:45:21] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
979[12:45:47] <tuxmania> tarzeau: I think I am confused with all. Not in the debian repo, you re right. But since I was not able to do an apt-get update, I thought this might be linked
980[12:46:05] <dragestil> freem: what is "phone home"?
982[12:46:40] <quadrathoch2> dragestil transfering data to the company back for example
983[12:46:55] <freem> you run some hardware produced by society A. It does it's job, but also logs stuff. Logs by themselves are useless, so the hardware "phones home" (to A) to send the logs
996[12:51:19] <freem> I'm certain of it, but I think they did a lot to remove as much as possible for an intel-arch, and have hardware switches to cut power/signal to camera, microphone and wifi stuff. As sad as it is, it's the best I've heard about :/
997[12:51:49] <freem> shtrb: ^ so if you know better in terms of computers for actual work, please share :)
1087[14:25:35] <deeplogic> is kernel 5.10 going to break my machine ?
1088[14:25:54] <petn-randall> !raspbian
1089[14:25:54] <dpkg> Raspberry Pi OS (previously called Raspbian) is a distribution <based on Debian> made specifically for the <Raspberry Pi>. Raspbian is not Debian and it is not supported in #debian. Please use #raspbian (or #raspberrypi) on irc.freenode.net for support. replaced-url
1090[14:26:03] <petn-randall> Woops, stuck in backlog.
1091[14:26:30] <petn-randall> deeplogic: What makes you assume that?
1092[14:26:57] <deeplogic> seems like a tremendous jump from 5.0.9
1113[14:31:02] <deeplogic> 5.9.0-4 is the only recent kernel that doesn't crash intermittently for me
1114[14:31:19] <shtrb> you mean an oops ?
1115[14:31:37] <deeplogic> nope, this laptop locks up
1116[14:32:16] <deeplogic> the magic key doesn't help
1117[14:32:52] <petn-randall> !tias
1118[14:32:52] <dpkg> TIAS is "Try It And See".
1119[14:32:54] <petn-randall> deeplogic: ^
1120[14:33:09] <petn-randall> deeplogic: Are you running buster with backports, or testing/sid?
1121[14:33:22] <shtrb> deeplogic, a lockup might indicate non kernel problems
1122[14:34:11] <deeplogic> testing
1123[14:34:14] <shtrb> deeplogic, when you are hit with a lockup check if pushing on capslock / numlock change the led state for example , if not , the lookup is on firmware level
1144[14:55:15] <noisemaker> Does any one has errors at boot related with "ACPI BIOS Error (bug) Could not resolve ..." and error related with "Volume group '-some-name-vg-' not found - Cannot process volume group '-some-name-bg'" ?
1145[14:55:31] <noisemaker> (this is a fresh install by the way)
1146[14:56:15] <shtrb> First one can happen if you have a firmware (board) that requrie an ugrade , or use of nuevo with some nvidia cards
1183[15:12:40] <noisemaker> shtrb: i will ignore it. Do you have any idea about the other about volume group? I see in another machine the same error, both ar using encryption.
1184[15:13:33] <shtrb> no sorry
1185[15:13:41] <noisemaker> np
1186[15:13:41] <shtrb> wait for ratrace to arrive he know everything
1207[15:30:17] <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
1214[15:32:08] *** Quits: hyiltiz (~quassel@replaced-ip) (Remote host closed the connection)
1215[15:33:02] <jhutchins> Kliment: You need to include details. We aren't aware of any failures in network configuration, and since there are at least three ways to configure it, we'd be surprised if they were all to fail at once. Usually things break one at a time in testing.
1216[15:33:11] <jhutchins> Kliment: What did you try to do? How did you try to do it? What did you expect to happen? What happened instead?
1217[15:33:45] <jhutchins> Kliment: We can answer general questions here, but things specific to testing are best asked in the other channel.
1218[15:34:10] <Kliment> jhutchins: Oh, sorry, I mean there is no UI for network configuration, network manager or similar
1219[15:34:18] <Kliment> I guess just missing package
1220[15:34:23] <Kliment> I'll ask on -next
1221[15:34:50] <jhutchins> Kliment: Yes, CLI and GUI intefaces for Network Manager. It's often an applet that needs to be added to your task bar, but usually installed by default.
1222[15:35:04] <jhutchins> Kliment: You can also configure it manually by editing config files.
1226[15:36:01] <dpkg> If you are looking for manuals, guides and procedures focusing on Debian - from a user perspective, the administrator's view and others - see replaced-url
1227[15:36:32] <Kliment> jhutchins: it appears networkmanager is not installed at all on that image (unlike, say, the cinnamon image I also tried)
1228[15:36:49] <jelly> that seems unusual for a Plasma desktop
1229[15:36:58] <Kliment> indeed, which is why I'm bringing it up
1251[15:50:27] <gnat_x> iterati: the gtk theme you use probably has the most affect. but that's limited, since the software doesn't follow the gtk3 paradigms that are different from gtk2.
1262[15:57:47] <Rob_Jones> can anyone help me with why im getting my ftp error? Status: Connection attempt failed with "ECONNREFUSED - Connection refused by server".
1263[15:58:19] <Rob_Jones> I have port 21 enabled by the way in my firewall (21/tcp ALLOW Anywhere)
1264[15:59:12] <iterati> gnat_x: thanks. I see there is a package gnome-themes-extra "This is the version of Adwaita, the standard GNOME theme, for the
1265[15:59:12] <iterati> GTK+ 2.x toolkit". Probably that's it, I guess.
1266[16:00:07] <gnat_x> Rob_Jones: i'm going to assume you've been warned that ftp is a bad idea, and need it for something specific... are you seeing anything in the ftp logs on the server?
1267[16:00:07] <ratrace> Rob_Jones: do you have ip_conntrack_ftp kernel module loaded?
1273[16:01:14] <ratrace> it's automatic punching of holes in the firewall for PASV connections -- otherwise you must manually open the ports in your PASV ports range
1282[16:03:26] <ksk> Rob_Jones: for that to work you would however not need any ftp server running..
1283[16:03:28] <greycat> ... please tell me this is a joke or a misunderstanding.
1284[16:03:31] <ratrace> Rob_Jones: but WP is then a FTP client, not a server..... which would affect your _outbound_ ports, not inbound
1285[16:03:51] <greycat> And indeed, if you only wanted to download updates you would need an FTP client, not an FTP server.
1286[16:03:52] <ratrace> and you should not block outbound connections unless you know what you're doing.
1287[16:04:22] <ratrace> I'm calling it....
1288[16:04:23] <ratrace> !xy
1289[16:04:24] <dpkg> Slow down for a bit! Are you sure that you need to jump through that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective... We know that people often falsely diagnose problems because they are too close to them -- it's easy to miss that there is a better way to proceed. See replaced-url
1291[16:04:45] <Rob_Jones> so i dont need an ftp account for my wordpress?
1292[16:04:48] <ratrace> primarily because WP would/should/must use https these days.
1293[16:05:01] <ksk> Rob_Jones: to the best of my knowledge even in the wordpress update there is no ftp included. wordpress had an integrated updater for quite a while now, you can use it via your admin panel, or even enable automatic updates..
1294[16:05:33] <ksk> Rob_Jones: maybe refer to the wordpress documentation before you continue implementing anything.
1295[16:05:45] <ratrace> personally I'd recommend WP not do any kind of self-modifying things, and you can use wp-cli from teh command line to upgrade it.... but that's maybe too advanced suggestion at this stage.
1302[16:11:08] <xoyi-> I've installed a fresh Debian (buster) image on a newly acquired Carbon X1 8th gen. However I believe it doesn't recognize the VGA card yet
1303[16:11:13] <shtrb> Rob_Jones, AFAIK you don't need ftp for wp, some people do provide it but that is absolutly not needed
1304[16:11:23] <shtrb> VGA card ?!
1305[16:11:56] *** Quits: Souler (~Souler@replaced-ip) (Remote host closed the connection)
1313[16:14:35] <zege> Rob_Jones: Than you don't need ftp at all. I was confused because Wp documentation mentions uploading the wp-files to the server via ftp.,
1314[16:14:52] <b1ack0p> zup
1315[16:15:28] <zege> Rob_Jones: what plugin? And why would one need that?
1316[16:15:33] <greycat> I think the WP documentation is written for people who have extremely bizarre setups, e.g. they can ONLY send files to their "web space" via FTP or something.
1317[16:15:42] <ksk> this.
1318[16:16:12] <greycat> I wonder if they have a set of documentation for people who *don't* have bizarre needs or intellectual handicaps.
1319[16:16:22] <Kliment> the kde live image network manager problem has been resolved in today's weekly
1320[16:16:23] <xoyi-> @greycat any idea how can I find out the chipset?
1329[16:18:21] <shtrb> there is only one true 8086 !
1330[16:18:51] <greycat> !non-free sources
1331[16:18:51] <dpkg> Edit /etc/apt/sources.list, ensure that the two main Debian mirror lines end with "main contrib non-free" rather than just "main", then «apt-get update». But bear in mind that you'll be installing <non-free> software. These may have onerous terms; check the licenses. See also <sources.list>.
1332[16:18:54] <ratrace> shtrb: AMD called. They want a word with you.
1333[16:19:12] <Kliment> ratrace: they can have 8088 :P
1334[16:19:29] <shtrb> that's intel too :P
1335[16:19:34] *** Quits: asymptotically (~asymptoti@replaced-ip) (Remote host closed the connection)
1336[16:19:38] <Kliment> Yeah I know
1337[16:19:50] <zege> Thats Intel UHD graphics. Looks like you need kernel > 4.19. replaced-url
1348[16:23:07] <zege> xoyi-: in your case you need a newer kernel from buster-backports
1349[16:23:10] <greycat> xoyi-: if zege is correct, you will need a kernel from buster-backports, and probably firmware from there as well.
1350[16:23:13] <shtrb> xoyi-, look at what greycat had said to you
1351[16:23:35] <greycat> !buster-backports
1352[16:23:35] <dpkg> Some packages intended for Bullseye (Debian 11) but recompiled for use with Buster (Debian 10) can be found in the buster-backports repository. See replaced-url
1371[16:37:41] <Kliment> I'm getting "mdadm: no arrays found in config file or automatically" for tens of seconds on startup after install
1372[16:38:04] <ratrace> Kliment: what exactly is your setup like?
1373[16:38:05] <xoyi-> there's just 1 glitch now related to X11/i3: when I run `startx` (which automatically runs `exec i3`), the screen goes black. It's only fixed after I switch to a new TTY and then back to the previous one. It seems like the resolution is off(?)
1374[16:38:17] <shtrb> Kliment, do you have LVM setup ?
1375[16:38:31] <Kliment> ratrace: default install with encryption
1376[16:38:42] <Kliment> on an nvme drive
1377[16:39:43] <ratrace> Kliment: do you even have mdadm raid there?
1378[16:40:17] <Kliment> not that I know of? I literally just installed this
1550[18:19:22] <switchnode> i'm trying to build some software with autotools. it depends on a third-party library, libzstd, which i have installed under ~/.local (this is a multi-user box on which i do not have root access).
1551[18:19:27] <switchnode> i have set PKG_CONFIG_PATH to ~/.local/lib/pkgconfig; autoconf successfully finds libzstd.pc there. however, the build fails with "error while loading shared libraries: libzstd.so.1: cannot open shared object file: No such file or directory".
1552[18:19:36] <switchnode> `pkg-config libzstd --libs` outputs "-L<absolute path to ~>/.local//lib -lzstd"; the generated Makefile.in has the line "ZSTD_LIBS = @ZSTD_LIBS@"; the generated Makefile has the line "ZSTD_LIBS = -L<absolute path to ~>/.local//lib -lzstd", as expected.
1553[18:19:43] <switchnode> the compilation string printed by automake includes those flags, and there is definitely a libzstd.so.1 under ~/.local/lib (it's a symlink to libzstd.so.1.4.4 in the same directory). so why can't the linker find it?
1554[18:19:49] <switchnode> (i've also tried setting LDFLAGS directly to ~/.local/lib, but that didn't help.)
1599[18:44:18] <azidhaka> my question is if i am correct to look into libc for that types of problems
1600[18:44:22] <xoyi-> any ideas why network doesn't work (fresh Debian Buster installation) but only when I connect via WiFi? (other devices in the same WiFi work, and this doesn't seem a DNS issue because I can't either ping via IP)
1601[18:45:41] <karlpinc> xoyi-: ? Either the network/nic works or does not. What does "doesn't work" mean?
1602[18:45:54] <jhutchins> azidhaka: I'd start with the kaspersky forums. Is this a two year old package? What debian release are you on?
1605[18:47:06] <karlpinc> xoyi-: Did you use the installer with firmware included?
1606[18:47:11] <karlpinc> !firmware images
1607[18:47:11] <dpkg> There are <live> system and <netinst> and DVD images containing non-free Debian <firmware> packages available from replaced-url
1608[18:47:17] <jhutchins> xoyi-: "doesn't work" is pretty vague. Have you checked for messages in dmesg?
1609[18:47:28] <azidhaka> jhutchins: buster, i've send a support ticket, but they are really slow in answering, and a package update caused the breakage, that's why i am asking here
1610[18:47:35] <xoyi-> karlpinc: I can connect to the network (via nmtui) but it seems like the network is down(?). For example, `ping 8.8.8.8` returns "Destination Host Unreachable"
1611[18:47:58] <azidhaka> someone might have experienced this
1612[18:48:22] <jhutchins> azidhaka: See if they have a support forum.
1615[18:49:26] <azidhaka> jhutchins: i will, thanks! its an obscure problem
1616[18:50:05] <jhutchins> azidhaka: Make sure you're running their latest version.
1617[18:50:22] <azidhaka> jhutchins: it's the latest version. Will it be possible to see your nsswitch.conf, what does it have as resolution order for group: ?
1618[18:50:41] <jhutchins> azidhaka: Commercial software often doesn't get updates on all releases and distros.
1619[18:50:44] <xoyi-> I think the NIC works, since I can connect to the WiFi network via nmtui
1620[18:51:46] <jhutchins> Mine only has "compat".
1621[18:52:19] <azidhaka> jhutchins: it is supported, i've got paid support, but thought it will be faster to see which linux system update broke it
1622[18:52:44] <azidhaka> jhutchins: our mail server is running without av/spam protection right now
1623[18:53:42] <jhutchins> xoyi-: So what's the actual problem? How are you trying to connect that it fails? We don't know what you've done with your system.
1631[18:54:54] <jhutchins> azidhaka: Most of us don't run paid software, we're pretty much all open source, but if you're handling mail for Windows you probably need it.
1632[18:55:05] <jhutchins> azidhaka: avg used to have a free package.
1634[18:55:41] <azidhaka> jhutchins: that's the only paid software we use, everything else is open-source. And yes, that's handling mail for Windows workstations
1635[18:56:56] <jhutchins> xoyi-: So you're saying that this system connects to wifi, but you can't ping google. Can you ping other sites? Do you get dig/host replies? Can you ping devices on your local network? Is the correct gateway being assigned?
1636[18:56:56] <azidhaka> jhutchins: couldn't assemble open-source system which filters viruses/spam as well as this
1637[18:57:33] <karlpinc> xoyi-: xoyi- You need to try to ping your local gateway.
1650[19:08:08] <jhutchins> Makes me suspect that maybe it was a permissions/user issue. Ah well, it works. "Fixed" costs a lot less than "why".
1651[19:08:37] <azidhaka> jhutchins: sudo-ing as the user, i am able to read /etc/group
1652[19:08:55] <greycat> azidhaka: if the service in question is started by systemd, it could have been started with the same modifications that buster's login uses, that breaks NIS until you install nscd or equivalent. Or it could be an AppArmor thing.
1661[19:11:32] <greycat> I very strongly doubt the same issue could arise from an init.d script. The config options in question are highly systemd specific.
1676[19:25:49] <unixbsd> is eternal lands to be soon or later, free into the debian repository?
1677[19:27:29] <sney> unixbsd: you can find this out yourself, for any piece of software. a) is the license DFSG-free? b) is there a wnpp bug?
1678[19:28:25] <unixbsd> I saw. ... what an idea. this is not thinking about opensource community at all.
1679[19:28:41] <greycat> What?
1680[19:29:03] <unixbsd> EL with GNU would be ideal for all.
1681[19:30:02] <greycat> So, let's pretend for a brief moment that anyone in this entire channel other than you knows what "eternal lands" is. How does this make it OUR responsibility to discover its status within the Debian development framework for you? Why can't it be YOUR responsibility to find the information that you want?
1682[19:32:08] <unixbsd> after some time, people give their project to gnu licensing.
1683[19:32:22] <sney> wouldn't that be wonderful if it were true
1691[19:37:30] <Kliment> anyone know how to enable fingerprint authentication in testing? I have libpam-fprintd installed
1692[19:37:54] <karlpinc> !debian-next
1693[19:37:54] <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
1694[19:38:01] *** debhelper sets mode: +l 1232
1695[19:38:09] <karlpinc> Kliment: See above if no one answers.
1696[19:38:29] <Kliment> karlpinc: did that. thanks
1703[19:41:42] <Filystyn> I will try not to explode
1704[19:42:19] <ratrace> zutat: there's a verb missing in that sentence. but I'm gonna hazarad a guess about your question, and say that any partitioning tool is also a recovery, since gpt uses backup sectors. take a look at sgdisk for example.
1725[19:48:25] *** jotaxpe__ is now known as jotaxpe
1726[19:50:54] <frazzydee> I'm running debian buster, and I notice I have the following entry in my /etc/apt/sources.list.d, is this OK? "deb replaced-url
1728[19:50:59] <frazzydee> not sure where that entry came from.
1729[19:52:05] <ratrace> well either you, or an impostor root, added it there
1730[19:52:26] <frazzydee> ratrace: Yes, could have been me. I don't remember how or why I added that in. Shouldn't it say buster not sarge?
1731[19:52:34] <frazzydee> Maybe some old install script.
1732[19:53:02] <ratrace> whether it's okay or not depends on whether you want to even attempt installing who knows what level of quality software from notoriously poor-security webmin's _sarge_ repo.
1733[19:53:36] <greycat> if it's truly a repository of sarge pkgs, it will probably not work very well
1734[19:53:56] <greycat> !webmin
1735[19:53:56] <dpkg> Webmin is a lame web-based interface for unsafe system administration for Unix. Check it out at replaced-url
1736[19:54:03] <frazzydee> honestly I don't even remember installing webmin.
1737[19:54:21] <Filystyn> Ok qos module did it
1738[19:54:23] <frazzydee> It must be used by some other program I have installed, perhaps znc or my rtorrent/rutorrent?
1739[19:54:26] <Filystyn> sorry for fuse
1740[19:54:41] <greycat> just because you have a repository in sources.list does not mean it was ever *used*
1741[19:54:57] <ratrace> frazzydee: if YOU don't remember why you added it...... I can do it for you. I have a crystall bal. That'll be $49999 payable in advance, tho
1742[19:55:19] <frazzydee> ratrace: lol
1743[19:55:31] <frazzydee> Don't need a crystal ball, I'm just expressing how weird I find it.
1745[19:55:51] <greycat> If you don't use webmin, I would comment out the epository first thing.
1746[19:55:55] <ratrace> frazzydee: webmin is a (terrible) web admin panel like cpanel or plesk, but infinitely more bad. probably nothing to do with znc or rtorrent
1747[19:56:06] <greycat> Second thing, find out whether webmin is actually installed.
1766[20:08:28] <jhutchins> frazzydee: Check top, check ps -ax, identify as many processes as you can. Disconnect from the net and see if anything changes. Look for odd directories.
1769[20:09:58] <wobelinger> einen wunder schönen guten morgen
1770[20:09:59] <jhutchins> frazzydee: debsums can be useful, but it has to be installed _before_ exposure. Tripwire can be useful. rkhunter and chkrootkit may return false negatives and should be initially run from separate boot media.
1774[20:12:47] <ratrace> no tool can be trusted if there's suspicion of pwnage.
1775[20:12:50] <jhutchins> wobelinger: Guten tag.
1776[20:12:52] <jhutchins> !de
1777[20:12:53] <dpkg> Deutschsprachige Hilfe bekommt ihr in #debian.de (auf irc.oftc.net, irc.freenode.net oder irc.belwue.de) - debian-user-german@lists.debian.org
1779[20:13:05] <unixbsd> I just compiled eternal lands on amd64, it is very easy actually to compile. this project just deserve to be GNU, and free. I will make a little fork with little installer.
1824[20:27:05] <jhutchins> So non-free is stuff that may be under an non-dfsg license, right?
1825[20:27:30] <ratrace> yup
1826[20:27:33] <unixbsd> Anyhow I upload on gitlab ;)
1827[20:27:47] <ratrace> requires at least permission to redist
1828[20:27:51] <greycat> You still need to be legally *allowed* to distribute the packages.
1829[20:27:57] *** Quits: alexeightsix (~alexeight@replaced-ip) (Remote host closed the connection)
1830[20:28:20] <ratrace> unixbsd: does the license permit you to?
1831[20:28:21] <greycat> One of the primary reasons a license may be considered non-free is because it doesn't allow free redistribution of the package, or distribution of patched versions.
1835[20:28:54] <epsilonKNOT> I wanted to submit a package for debian
1836[20:28:57] <epsilonKNOT> Yo ratratce
1837[20:28:58] <greycat> I don't know QPL. So you'd have to take it up with debian-legal or the copyright holder.
1838[20:29:14] <epsilonKNOT> Pardon my mobile typos
1839[20:29:38] <epsilonKNOT> So, like how do I submit a package to Debian?
1840[20:29:47] <ratrace> I'd definitely go with copyright holder's written permission, or maybe they're even willing to relicense. it's a very old game
1841[20:29:48] <epsilonKNOT> I don't see any ports tree or something
1842[20:31:46] <Mister00X> ratrace: What sudo CVE do you mean?
1843[20:32:22] <ratrace> epsilonKNOT: only DDs can upload methinks... you'd need a sponsor or something . replaced-url
1854[20:36:19] <ratrace> epsilonKNOT: more resources here. I've been actually going through this from cover through cover and playing with building some packages for our company's in house software :: replaced-url
1855[20:36:56] <greycat> !debian-mentors
1856[20:36:56] <dpkg> Find packaging help for the Debian project in #debian-mentors on irc.oftc.net. If the channel is invite-only it's because you missed it being on OFTC.
1863[20:40:51] <jhutchins> Man, learning to build RPMs on CentOS was a total pain. They had a scripted chroot build that basically built a new server for each build.
1864[20:41:22] <jhutchins> There was no documentation, and trying to ask questions about the process would get you kick-banned.
1865[20:41:28] <ratrace> not a terrible idea. FreeBSD ports system "Poudriere" does the same with jails. clean build from scratch for each package
1866[20:41:52] <ratrace> makes sure no stray package or file or leftover or something pollutes the building
1868[20:42:57] <jhutchins> ratrace: It's not a horrible idea, unless you're in the build-tweak-build part of development. This was a little more than ten years ago, when kernel compiles were still used as benchmarks. Not as much horsepower as we have today.
1869[20:43:15] <jhutchins> RPM is also a heavier system to build for.
1870[20:43:39] <jhutchins> ratrace: Yeah, or that development environment you were using on the same box.
1871[20:43:53] <ratrace> right.
1872[20:45:35] <ratrace> build-tweak-build is okay if the framework is supporting it. don't remember if the RPM framework did .... whatwasit .... koji? last tiem I wrote a spec file and built RPMs was ... many orbits ago
1876[20:46:46] <ratrace> from that standpoint building debs is easier. but the format is so complex compared to a single spec, .ebuild, or ports makefile
1942[22:00:29] <sney> (also, a lot of consumer ISPs will just block TCP/25 due to spambot activity, so if you're trying to host mail from your home computer that is likely the problem)
1943[22:02:09] <epsilonKNOT> jhutchins: just tried to report bug and it failed. Gimme sec will get error
1944[22:02:41] <sney> don't use reportbug for tech support issues, package maintainers are busy as it is
1949[22:03:50] <aisha> so i'm epsilonKNOT (just switched to webchat for easier typing)
1950[22:03:50] <greycat> sney: I'm guessing he hasn't got a working MTA set up...
1951[22:04:10] <aisha> error on reportbug - SMTP send failure: [Errno 101] Network is unreachable. You can retry, or save the report and exit. Do you want to retry [Y|n|q|?]? ?
1952[22:04:41] <jhutchins> epsilonKNOT: It has an off-line mode that doesn't require a mailer.
1962[22:07:49] <sney> though there are ways that it can get skipped, or environments where the default wouldn't be able to send mail to internet addresses
1963[22:08:13] <aisha> this sounds crazy
1964[22:08:22] <aisha> cant it just use an external email server?
1975[22:11:50] <greycat> In past versions of Debian, cron and exim/exim4 were installed by default, and the user was prompted for basic MTA configuration during the installation. At some point, Debian stopped doing it that way. So now there's a shit-ton of Debian systems out there without a working MTA installed.
1976[22:12:09] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1982[22:13:29] <Jen_smiles> i have an application that i installed ( awscliv2 ) and it works under my user ( if i do a "which aws" it give me /user/local/bin/aws )
1983[22:13:44] <Jen_smiles> but if i run it as root it won't work, and "which aws" can't locate it
1987[22:14:16] <sney> no, it's likely a PATH issue
1988[22:14:33] <sney> did you mean /usr/local/bin? was 'user' a typo?
1989[22:14:53] <Jen_smiles> sorry! yes that was a typo
1990[22:15:13] <sney> ok, so become root and check the output of 'echo $PATH'
1991[22:15:16] <Jen_smiles> from my regular user 'which aws' = /usr/local/bin/aws
1992[22:15:37] <Jen_smiles> as soon as you said $path i checked that, i'm just going over it now
1993[22:16:07] <greycat> Some of the ways that you can become root will change your PATH variable, and some of them will not.
1994[22:16:35] <greycat> But even if your PATH is changed, it's odd that /usr/local/bin isn't in the new one.
1995[22:16:39] <sney> indeed
1996[22:16:47] <sney> it could also be that it's not set as +x for everyone
1997[22:17:07] <sney> though anything in /usr/local should still be owned by root unless it was installed strangely
1998[22:18:42] <greycat> The three obvious ways to become root with a root-specific PATH are direct console login, sudo -s (or su if configured with ALWAYS_SET_PATH yes), or sudo -i (or "su -"). I think all 3 of those should give you a PATH with /usr/local/bin in it, though.
2003[22:21:27] <greycat> Direct console login, sudo -i, su -, or ssh root@ all give you a login shell as root, so it picks up PATH from /etc/profile and/or /root/.profile. The one in /etc/profile definitely has PATH with /usr/local/bin, and at least *my* /root/.profile doesn't touch PATH, although I'm not 100% sure it's a default .profile.
2004[22:21:46] <Jen_smiles> i'm basically logging in as a non-root user, then doing a "sudo su"
2005[22:21:54] <sney> haha,
2006[22:21:56] <sney> !sudo su
2007[22:21:56] <dpkg> Typing "sudo su" is like typing "cat file | cat". If you want a non-login shell as root, just use "sudo -s". If you want a login shell, "sudo -i". There's no need to run two different superuser-elevation programs chained back to back. See also <buster su>
2008[22:22:00] <greycat> !sudo su
2009[22:22:02] <sney> but still this probably isn't the problem
2012[22:23:02] <greycat> In that case, sudo is setting the PATH as it sees fit, and then invoking a non-login su, which will either leave PATH untouched (default behavior), or overwrite PATH if you configured ALWAYS_SET_PATH in /etc/default/su.
2013[22:23:28] <sney> indeed, on a debian system with no PATH tweaks: replaced-url
2035[22:27:45] <sney> ok, this is amazon's fault. update /etc/profile and send amazon an email about not changing defaults that they don't understand
2036[22:27:51] <greycat> That's the part that intrigues me. Can you figure out why that happened? Did someone put PATH=... in your /root/.profile or some other file?
2037[22:27:52] <sney> next
2038[22:28:04] <Jen_smiles> after doing "sudo -s" my $PATH is ==> /sbin:/bin:/usr/sbin:/usr/bin
2039[22:28:23] <greycat> That is not normal for Debian.
2040[22:28:41] *** Quits: dreamon (~dreamon@replaced-ip) (Remote host closed the connection)
2041[22:29:00] <greycat> sudo -s on my system (buster) gives /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
2042[22:29:03] <Jen_smiles> our AWS account is so large we have 11 TAMs assigned to us, so I can get an answer on that in an hour, and sure I can ask
2048[22:30:49] <sney> this is still a dumb way to accomplish that, but it may have an ostensibly security-related justification, so you may want to check before doing it anyway
2049[22:31:29] <Jen_smiles> is "sudo -s" over "sudo su" a standard that should be followed on just debian or all linux distros ?
2061[22:35:33] <jhutchins> Jen_smiles: sudo -s or sudo -i
2062[22:35:46] <sney> !sudo
2063[22:35:46] <dpkg> Sudo, aka Substitute User DO, is a common way to give limited super user privileges to specific users. Not the magic "do what I want" button. Good in scripts with "username ALL = NOPASSWD: /some/program". See <sudo path>, <visudo>, <sudoers>. replaced-url
2064[22:36:25] <jhutchins> I think sudo is pretty consistent across distros. Su wasn't, but I think most conform to the RHEL standard these days (su -).
2065[22:37:18] <ratrace> too bad sudo insults have to be compiled in, and not just a config option
2066[22:37:33] *** Quits: mezzo (~mezzo@replaced-ip) (Quit: leaving)
2088[22:55:36] *** Quits: emptylica (~pi@replaced-ip) (Remote host closed the connection)
2089[22:56:06] *** Quits: Jad (~Nashmi@replaced-ip) (Quit: Benefits I derive from freedom are largely the result of the uses of freedom by others, and mostly of those uses of freedom that I could never avail myself of.)
2092[22:57:17] <aisha> user not found in sudoers file, proceeding to throw banana grenades and wipe them from the face of the earth
2093[22:57:41] <noisemaker> What's the name of the package to tweak the mouse pad of a laptop ?
2094[22:58:01] *** debhelper sets mode: +l 1225
2095[22:58:12] <ratrace> aisha: moar raciss with correct flags :)
2096[22:58:31] <aisha> ooof
2097[22:59:06] <sney> noisemaker: there used to be a 'synclient' but I don't think it works anymore since the switch to libinput. you should some controls in your DE settings widget at least
2099[23:00:41] <noisemaker> At ubuntu old version I didn't need to press the pad to make the click, I just needed to tap for it, so what I need to config ?
2108[23:04:03] <noisemaker> ratrace: well, yeah, I did, I can't find the setting :)
2109[23:04:06] <sney> Kliment: it's not clear, there doesn't seem to be a bug about it. common reasons can be maintainer mia, missing new dependencies, upstream license changes
2110[23:04:26] <aisha> how long does it take for package to appear in mentors package listing?
2133[23:13:25] <Kliment> I think they're the worst idea ever
2134[23:13:32] <ratrace> then no blender for you.
2135[23:13:48] <Kliment> I used ubuntu for the past 15 years
2136[23:14:02] <ratrace> now... it's one thing when it's forced and defaulted to like ubuntu, and quite another where you have the _option_ to use them, and no better alternatives.
2137[23:14:11] <ratrace> that said, maybe there's flatpak, if you dislike snapd
2138[23:14:16] <Kliment> Well, I'm uncomfortable with the snap ecosystem
2139[23:15:21] <ratrace> or maybe you can just download and unpack the tarball: replaced-url
2140[23:15:38] <Kliment> on ubuntu I use a number of ppas for applications I need to be up to date on, and it updates in a sane way with the rest of the system
2141[23:15:39] <sney> you could probably use uupdate to update the debian package too
2142[23:15:40] <ratrace> ,v glibc
2143[23:15:41] <judd> No package named 'glibc' was found in amd64.
2152[23:17:39] <Kliment> ratrace: the forcing of it is not what bothers me - I mean apt is effectively forced on debian too. What bothers me is the technical design of snaps - single source, severe limitations on functionaluty, and really terrible performance
2159[23:18:39] *** Quits: Deknos (~deknos@replaced-ip) (Remote host closed the connection)
2160[23:18:43] <Kliment> what is this uupdate thing?
2161[23:18:46] <Kliment> !uupdate
2162[23:18:46] <dpkg> uupdate (in the <devscripts> package) upgrades a source code package from an upstream revision, or e.g. if the newest foo in debian is 1.2, and upstream is at 1.4: apt-get source foo; wget -nd replaced-url
2163[23:19:05] *** Quits: T-zef (~tyzef@replaced-ip) (Remote host closed the connection)
2164[23:19:11] <Kliment> interesting
2165[23:19:29] <sney> careful, it's the gateway drug to actually being a debian maintainer :P
2210[23:57:55] <jhutchins> I'd love to see Arthur get more involved in this than facebook posts, but at his age he's entitled to do as little as he likes.
2211[23:59:02] *** Quits: gonzo (~gonzo@replaced-ip) (Remote host closed the connection)