this is #debianan IRC-Channel at freenode (freenode IRC service closed 2021-06-01)
0[00:00:02] <jhutchins> MESSIAH: SSH?
1[00:00:23] <MESSIAH> nope e-mails
2[00:00:35] <MESSIAH> Probably in future will be ftp and ssh
3[00:01:00] <Vizva> why the gui klvm is dropped?
4[00:01:11] <sney> ,v klvm
5[00:01:12] <judd> No package named 'klvm' was found in amd64.
6[00:01:21] <jhutchins> MESSIAH: You'll need to investigate the documentation for yor software. Most MTAs are able to use a blacklist, but it's not hosts.deny.
7[00:01:43] <sney> ,file bin/klvm
8[00:01:47] <judd> No packages in buster/amd64 were found with that file.
9[00:01:56] <sney> Vizva: any more details?
10[00:01:59] <jhutchins> MESSIAH: hosts.allow and hosts.deny are mostly used by older protocols like telnet, rsh, and ftp.
11[00:02:25] <jhutchins> MESSIAH: I would recommend that you NOT run ftp, but use the sftp feature of ssh.
12[00:03:00] <Vizva> it was a gui to deal with the lvm2
13[00:03:17] <sney> using postfix's domain_access to blackhole entire tlds is also a good way to deny large amounts of spam with minimal overhead.
14[00:03:26] <MESSIAH> jhutchins I prefer block totally IP after 10 failed attempts
57[00:13:53] <Vizva> you must understand thats a little bit a joke from my side
58[00:14:02] <Allain> :D
59[00:14:04] <Vizva> they dont come on my door
60[00:14:08] <Vizva> :P
61[00:14:57] <MESSIAH> How to totally control computer by internet including reboot, shut down or enter to BIOS. I'm wondering if it's possible to remote install system on computer without getting physically contact?
62[00:15:46] <Vizva> look here the hacker needs a new idea for broke into
63[00:15:53] <Vizva> ;)
64[00:15:55] <Allain> system-config-lvm does not exist anymore in Debian's packages after Stratch
66[00:16:40] <sney> MESSIAH: there are "lights out" controllers that you can install in a server to do that. also some VPS providers provide a vnc console that lets you look at grub, etc. but you can't do it with just software on a regular computer
67[00:16:52] <Vizva> if you find a vulnerability you can install cron jobs too
72[00:18:20] <sney> Vizva: this is a debian support channel. "hacker" stuff is offtopic
73[00:19:00] <Vizva> sney i dont, simply
74[00:19:30] <Vizva> <MESSIAH> How to totally control computer by internet including reboot, shut down or enter to BIOS. I'm wondering if it's possible to remote install system on computer without getting physically contact?
75[00:19:54] <sney> Vizva: that was a legitimate question and I answered him. you commented with some offtopic nonsense. please don't do that again.
76[00:20:03] <Vizva> if someone ask such a question - its a little bit on this path and its allowed to still use your brain
77[00:20:34] <sney> especially if you have no idea what you are talking about. it gets in the way and can confuse people
158[01:58:38] <sney> yes, 10.4 upgrades directly to 10.5 with no config changes.
159[01:58:53] <sney> also, your config files will never be wiped without your knowledge
160[01:58:53] <MESSIAH> So I should not worry about nothing?
161[01:59:16] <oxek> If I want to use vi-style keybindings in my terminal, do I have to modify both .bashrc for bash and .inputrc for readline?
162[01:59:28] <sney> MESSIAH: are you sure you aren't on 10.5 already? cat /etc/debian_version
163[01:59:45] <MESSIAH> All apps and server will be work same after upgrade?
164[01:59:50] <sney> yes.
165[02:00:04] <themill> !point release
166[02:00:04] <dpkg> Point releases are updates to <stable> and <oldstable>, fixing security and grave bug fixes. If you track security updates regularly (as you should!) there will often be no updates for you in the point release. You can upgrade to the latest point release with "apt update && apt full-upgrade". Ask me about <9.11>, <10.1>.. replaced-url
167[02:00:05] <MESSIAH> sney nope. I have information
213[02:40:51] <sinack> could someone provide assistance for adding/remove current graphics packages on a new buster xfce install... I have terrible screen tearing/sync issues and the debian wiki suggested removing the current package and installing another. I dont want to get a black screen so do I simply install the new package first?
340[05:54:34] <iateadonut> i have a symlink to create my /etc/hosts, e.g. sudo ln -s ~/etc/hosts /etc/hosts - and i'm getting permissions errors: systemd-resolved[856]: Failed to open /etc/hosts: Permission denied - wondering how i can resolve that.
341[05:55:57] <sney> did you actually use ~ ? that would expand differently for root or a service account than it would for your user.
366[06:13:36] <sney> /etc/hosts shouldn't be a symlink. I doubt anything is tested with that. keep /etc/hosts as the file and put the symlink in /home.
367[06:13:44] <somiaj> iateadonut: systemctl show systemd-resolved shows that User=systemd-resolve, does that user have access to said file in /home/foo/
381[06:19:54] <somiaj> Maybe others might have some other solutions that don't require links. Also maybe just use /home/etc for this instaed of /home/user/etc (here I"m assuming /home is the partition that ist he same across oses)
382[06:20:14] <somiaj> this way you can keep things owned as root, and not have as many layers of permsision to make sure things are readable.
383[06:22:07] <somiaj> I have heard of people using git for the config files in /etc they like to keep a custom version of.
390[06:27:01] <foul_owl_> How do you install flash player in Debian 10?
391[06:27:34] <somiaj> not easily, I personally suggest getting google-chrome from google and using their bundeled flash (though they may loose that soon too)
392[06:28:10] <foul_owl_> Thank you! It's just to play some old flash games
393[06:28:18] <somiaj> ,i gnash
394[06:28:19] <judd> No package named 'gnash' was found in buster/amd64.
402[06:31:05] <somiaj> foul_owl_: adobe still has both the napi and papi plugins, but they are getting harder and harder to actually manually install...flash is dead
403[06:31:21] <somiaj> which is a pain for your use case..:/
404[06:31:30] <foul_owl_> I realize that :( No worries
568[10:05:49] <gvth> Hi; simple question: Does anyone know of a graphical (I mean real GUI, not ncurses or similar) file manager that supports showing only files within a directory that match a given regexp?
614[11:00:13] <wrksx> Hey, got this in the logs, I suspect a disk failure, but not sure, still testing. Had to reboot the server since most usefull commands would return IO errors (readonly disk?). I don't really understand what the log means, is there anything about it that you could tell me?
651[11:42:21] <efloid> hello, trying to configure passwordless sudo for members of sudo group, however the syntax '%sudo ALL=NOPASSWD: ALL' appears to not work with the version of sudo in Buster
652[11:42:55] <efloid> fwiw '%sudo ALL=(ALL) NOPASSWD:ALL' also is not working
664[11:53:15] <eventhorizon5> efloid: I have mine set to a specific user instead of to the sudo group, and it works fine for me. Is there any directive afterwards that is overriding it for you? You could also try it for a single user and see if that works.
701[12:06:40] <ksk> I do not use passwordless. If you say you have identical versions of debian and $packages, and it works on one machine and not the other.. I would somehow suspect you got the syntax right..
702[12:10:01] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
747[13:18:55] <alx^> when using 'sensors -u', I can see that some sensors have an 'alarm' value (e.g. temp5_crit_alarm). It is currently set to '0' but anyway, I don't see any reference to an alarm system in 'sensors'/'lm-sensors'. Is there actually one ?
748[13:19:04] <alx^> My objective is just to send e-mail alerts
749[13:19:39] <alx^> if a given temperature is reached
835[15:20:08] <shtrb> Will debian move to systemd-logind in testing ?
836[15:21:21] <greycat> !debian-next
837[15:21:22] <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
838[15:21:47] <greycat> (Did they move AWAY from it in testing, since stable?! Weird shit.)
839[15:22:15] <greycat> (Are you asking "Will they move back to it, after they moved away from it?" Or are you just confused? The world may never know.)
840[15:22:32] <shtrb> I'm confused
841[15:23:04] <greycat> buster systemd systems are using it.
880[16:16:33] <jelly> etcd, homed, soon: vard? bootd is already taken by OSX innit
881[16:16:53] <Antoine|> Hello, I use LVM. I have three disks (2TB, 2TB and 4TB). Each correspond to a physical volume. All three physical volumes are merged in one volume group. Then I have several logical volume. I would like to convert one logical volume from linear to mirrored. The logical volume is 3.2T and I have 3.2T free space in my volume group. Can I do that?
882[16:17:10] <Antoine|> I tried `sudo lvconvert -m 1 /dev/mvg/Medias` but it said I don't have enough free space
883[16:17:11] <jelly> Antoine|, best ask in #lvm channel
893[16:21:42] <greycat> Go on, amuse me. What is "more sane" about a daemon doing something to user home directories, versus having them be... just... plain... old... directories?
894[16:21:44] <escanor> Hi i installed kali on vm
895[16:21:56] <jelly> I don't care about RedHat's fixed ideas that allow read-only / and read-only /etc functionality
902[16:22:56] <escanor> When i enter credential i get a black screen
903[16:23:10] <escanor> Desktop is not loading
904[16:23:17] <Antoine|> jelly: Oh so if I understood, my issue might be that I'd need to select a pv to mirror my lv but my lv is larger than my 2TB pvs
906[16:23:22] <Antoine|> jelly: I asked in #lvm, thanks :)
907[16:23:25] <greycat> Why not ask in #fedora? Or #macosx? Those are also not Kali. Why pick *us*?
908[16:23:45] <jelly> greycat, ah, there's actually an answer to that, but you won't like it
909[16:24:28] <shtrb> kalie ~ #debian-next
910[16:24:31] <greycat> I can think of a few. "They kicked me from kali's channel" is the main one. "Kali's channel won't talk to me" is the lighter version. "Kali's developers sent me here by their choice of default settings" is probably the one you're thinking of though.
911[16:27:21] <jelly> greycat, wherever there's an issue or a question about a specific component that's imported from upstreamlargerly unchanged (by their opinion), it is likely they might point to the upstream's channel. They consider Debian as upstream.
912[16:27:41] <greycat> But the one that really *bothers* me is "I heard you tell me to go to the kali channel but I am ignoring you. I am going to stay here in the wrong channel and pester everyone."
913[16:28:17] <greycat> Or the lighter variant of that, "But this channel has more people!!!"
914[16:28:20] <jelly> We, OTOH consider almost ANY issue happening on derivative distros, as not topical in here.
922[16:33:46] <jelly> I have no idea what one is supposed to do with it or not; I did manage to figure out the relevant bits of their unwritten policy before getting banned from #kali-linux
923[16:34:42] <NetTerminalGene> why isn't there kiwix software in repo?
930[16:36:16] <jelly> escanor, no problem, you couldn't have known in advance. We can try to help if you get the same issue on an actual Debian installation instead.
962[17:01:29] <sinack> could someone provide assistance for adding/remove current graphics packages on a new buster xfce install... I have terrible screen tearing/sync issues and the debian wiki suggested removing the current package and installing another. I dont want to get a black screen so do I simply install the new package first?
963[17:02:09] <greycat> Try installing firmware first. Check 'dmesg | grep -i firmware' to see if there are files it wants to load and is failing to.
964[17:03:34] <sinack> that gives no results
965[17:03:53] *** Quits: r1nt3c (~r1nt3c@replaced-ip) (Quit: Lost terminal)
969[17:06:15] <sinack> I did uninstall the firmware; is a reboot needed?
970[17:06:44] <greycat> you *what*
971[17:06:57] <MorganSoulless> *grabs popcorn*
972[17:07:03] <dob1> I am mounting /proc with hidepid=1 so every user can see only its own processes: as user I am starting a virtual box vm, if as user I do ps aux or htop i don't see the vm process. if as root I do the same I see the vm process and its user it's my user... why?
974[17:08:26] <sinack> says to use the built-in modesettings driver... is any action needed to activate that?
975[17:08:38] <greycat> sinack: What hardware is it? (lspci -nn) What package did you remove, and why? What wiki page were you reading previously? What package did the wiki page say to reinstall, and why?
998[17:14:12] <greycat> Skylake should work out of the box on Debian after you install the non-free firmware. You *definitely* should have seen some files from "dmesg | grep -i firmware" (as root).
1006[17:15:32] <greycat> The box I'm using now is a Skylake, but lspci -nn says it's got 00:02.0 VGA compatible controller [0300]: Intel Corporation HD Graphics 630 [8086:5912] (rev 04)
1007[17:15:38] <sinack> I haven't manually installed any firmware, so should I do that?
1008[17:15:59] <sinack> lspci -nn says I have HD Graphics 530
1009[17:16:01] <jelly> sinack, if package firmware-misc-nonfree is not installed, try installing it
1010[17:16:09] <jelly> !non-free sources
1011[17:16:09] <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>.
1012[17:16:32] <jelly> that's where /lib/firmware/i915 seems to hide
1020[17:19:19] <jelly> I can't remember the bits needed for hardware accelerated vc9 / h264 / h265 decoding
1021[17:19:54] <jelly> lowers down the temp when playing movies on a laptop by a celsius or two
1022[17:20:53] <jelly> ,i i965-va-driver
1023[17:20:56] <judd> Package i965-va-driver (video, optional) in buster/amd64: VAAPI driver for Intel G45 & HD Graphics family. Version: 2.3.0+dfsg1-1; Size: 304.0k; Installed: 1793k; Homepage: replaced-url
1024[17:21:00] <jelly> that thing.
1025[17:21:52] <jelly> and who knows whether browsers detect that or not. mpv player does in Debian 10, needed some manual tuning previously
1026[17:22:23] <jelly> ,i vainfo
1027[17:22:24] <judd> Package vainfo (utils, optional) in buster/amd64: Video Acceleration (VA) API for Linux -- info program. Version: 2.4.0+ds1-1; Size: 14.7k; Installed: 43k; Homepage: replaced-url
1028[17:22:44] <jelly> ^ cmdline tool to see whether VAAPI codecs are available
1029[17:22:48] *** Quits: ruied (~ruied@replaced-ip) (Remote host closed the connection)
1053[17:41:07] <dacencora> hey guys. Ok still trying to figure this out. I am running debian on a laptop with an i7-9750H and an NVIDIA GTX 1650. On heavy disk I/O, the whole thing freezes. I was looking it up, and based on my searching, I put 'intel_idle.max_cstate=1'
1054[17:41:36] <dacencora> That actually stopped the freezing, but all apps are still not super responsive when under heavy disk I/O
1055[17:41:39] *** ryzokuken_ is now known as ryzokuken
1057[17:42:13] <dacencora> My mouse works, but apps are still more or less broken. So I think this cstate thing is on the right track, but it's not quite the resolution
1090[17:55:06] <jelly> you could blindly try if using a backported kernel makes things any different (also, try it without that cstate tuning if you do)
1091[17:55:11] <jelly> !bdo kernel
1092[17:55:11] <dpkg> Newer kernels for Debian stable releases are available from the <buster-backports> repository. After modifying your sources.list, run «apt update». To install the current backported kernel: «apt -t buster-backports install linux-image-`uname -r|sed 's,[^-]*-[^-]*-,,'`». To list available backported kernel image packages: «aptitude search '?narrow(~nlinux-image,?origin(Debian Backports))'».
1093[17:55:48] <dacencora> I thought the same as you did, and I have tried it with kernels all the way up to 5.6, and even tried different schedulers
1094[17:55:51] <dacencora> I am stumped haha
1095[17:56:01] <jelly> but two things: nvidia driver might break with a newer kernel; newer, matching, firmware from backports might also be needed for the newer kernel to work
1096[17:56:15] <jelly> ah, you've been there already
1097[17:56:28] <dacencora> I have to wonder if maybe it's some hardware flaw of some kind
1098[17:56:44] <gordonfish> dacencora: You might need to adjust swappiness so it swaps less, which can help when you have that much RAM.
1107[17:59:04] <f8e4> hi , how to find out what is on replaced-url
1108[17:59:05] <gordonfish> dacencora: Try setting it to 0 as a test as that should cause it hold off swapping until it's really needed.
1109[17:59:15] <dacencora> OK I will try that
1110[17:59:22] <jelly> you could echo vm.swappiness=1 > /etc/sysctl.d/local.conf to minimize aggressive swapping but does your system actually actively swap or not
1111[17:59:31] <gordonfish> That'll at least give some indication of my bottle neck hypothesis is on the right track.
1112[17:59:38] <jelly> I suspect it doesn't really
1114[18:00:24] <jelly> dacencora, run "vmstat 1" in a second terminal an put it on top of other windows, see if you can see activity in si / so columns when things start to lag
1115[18:00:29] <dacencora> OK I am testing it now. The easiest way for me to reproduce the issue is to start a large download
1116[18:00:30] <gordonfish> jelly: The heavy IO really feels like too much going over the bus at once, so swapping while under leavy IO load already is going to slow things down, I've seen it enough.
1117[18:00:58] <jelly> gordonfish, but with a nvme ssd backend?
1128[18:02:44] <dacencora> At some point the vmstat swpd column changed from 512 to 768
1129[18:03:14] <gordonfish> dacencora: How much RAM is free at this moment?
1130[18:03:39] <jelly> dacencora, that's in kibibytes and a really tiny number, ignorable
1131[18:04:02] <jelly> look at "available" column of "free" command output
1132[18:04:26] <f8e4> got it
1133[18:04:27] <jelly> (or free -h, or free -m, however you like to run it)
1134[18:04:47] <oiaohm> jelly: over all if you are in a deep enough IO hole it really does not matter if you are old school spinning rust or modern SSD neither has enough speed to match ram.
1135[18:05:09] <dacencora> Well I stopped the process
1136[18:05:28] <dacencora> But free -h only shows 806Mi free
1137[18:05:31] <oiaohm> Also once in the hole changing swap values take a long time to take effect.
1138[18:05:45] <dacencora> And 27Gi in buff/cache
1139[18:05:50] <gordonfish> If it's showing a lot of ram being used, I've found `sync && echo 3 > /proc/sys/vm/drop_caches` in the past to be of help (though that was back when I had around 4 or less GB of ram, so I haven't needed that in a long time.)
1147[18:07:46] <dacencora> But now all of that has gone back to fre
1148[18:07:49] <dacencora> *free.
1149[18:07:59] <dacencora> Ok I am trying that command gordonfish
1150[18:08:16] <oiaohm> Please note drop_caches can be double sided.
1151[18:08:30] <oiaohm> Yes you drop the ram caches but then you can need to re populate them.
1152[18:09:00] *** Quits: darunesh (~darunesh@replaced-ip) (Remote host closed the connection)
1153[18:09:03] <gordonfish> oiaohm: It can help when memory is being held onto when it really doesn't need to be
1154[18:09:04] <oiaohm> but if you think the cache system is fragmented/lost dropping is fix.
1155[18:09:17] <gordonfish> Yeah that
1156[18:09:50] <jelly> !free ram
1157[18:09:50] <dpkg> Unlike information, your computer's memory does *not* want to be free. Free RAM is wasted RAM! Linux tries to use free physical memory for caching files from disk which speeds up disk access considerably. Linux releases RAM from these caches if programs need it. If you want to know how much physical memory the free(1) tool says you have left for program use, it's 'free' + 'buffers' + 'cache'. Also ask me about <swapwake>.
1169[18:12:42] <jelly> I'm saying most just because I haven't seen a system without swap deteriorate horribly that way in a while (except if it's still running EL7)
1172[18:12:52] <gordonfish> jelly: Yeah, 'specially compared to the ntkernel (which don't get me wrong, has come a long way since NT 4, but still harbors man bugs from that era...)
1173[18:12:55] <gordonfish> many*
1174[18:12:58] <jelly> (and our context is #debian ...)
1175[18:13:12] <oiaohm> jelly: there are still in the latest release 14 areas using push structure to swap to allow reposition in memory. It is possible to run a system and avoid them all.
1190[18:15:46] <oiaohm> The push to swap turns out to be a safe way to alter structures handling to allow moving without push to swap. If you are needing to pull that structure back from swap while its being moved the handling code is not right is how they do it.
1191[18:16:06] <oiaohm> So new methods for defraging the Linux kernel structures in memory are done by push to swap.
1202[18:18:25] <oiaohm> jhutchins: If that is about swap in ram it does make sense with the Linux kernel.
1203[18:18:25] <longears> Is a way or a tool Detect to detect which drives were part of RAID? I have 7 different HDDs and only two of them were a part of RAID 0. Now just need to find them and reassemble the array.
1204[18:18:41] <jelly> the other half is "zram swap"
1205[18:18:51] <jhutchins> oiaohm: No, it doesn't.
1208[18:19:06] <oiaohm> jhutchins: Linux supports swap in ram.
1209[18:19:18] <oiaohm> Wacky world of Linux.
1210[18:19:27] <jhutchins> oiaohm: If it's in RAM, it's not swapped.
1211[18:19:33] <teclo-> .v musescore
1212[18:19:43] <RoyK> longears: just run mdadm --examine on those drives or partitions
1213[18:19:45] <jhutchins> I think we're using the wrong terms here. Swap is ram-to-disk.
1214[18:19:47] <oiaohm> jhutchins: it is swaped Just because stuff is still in ram
1215[18:20:04] <jelly> jhutchins, Linux supports two different ways to put compressed RAM pages into a different dedicated bit of RAM used as swap backend
1216[18:20:16] <jhutchins> oiaohm: That's something other than the traditional meaning of swap.
1217[18:20:21] <RoyK> longears: btw, there's #linux-raid where people talk a lot about these things
1225[18:22:05] <jelly> jhutchins, zram helps if you're running Chromium and a GUI on 2-4GB RAM, has compression up to 1:3 with 1:2.5 typical on desktop, but unreliable latency. zswap does only up to 1:2 but has reliable latencies.
1226[18:22:12] <dacencora> If you couldn't guess, drop_caches did *not* work unfortunately
1227[18:22:21] <dacencora> Had to hard boot
1228[18:22:45] <jelly> so you probably ARE hitting some edge case
1229[18:23:09] <jelly> or maybe you have nvidia driver doing something crazy, if that's a thing you're using
1230[18:23:20] <dacencora> IIRC, I also had this issue with an 8750h laptop
1231[18:23:35] <dacencora> The 9750h is essentially the same chip
1232[18:23:59] <dacencora> If I were a betting man, I'd wager it's something to do with 6 cores/12 threads on a laptop
1233[18:24:02] <jelly> does it make sense to try disabling nvidia and not loading it, for your use case?
1234[18:24:16] <dacencora> jelly: I can try that for sure
1235[18:25:50] * jelly almost bought a 6c/12thr amd laptop but went with a lighter intel instead
1236[18:26:03] <dacencora> The next step is to try and load Windows on the other hard drive and see if the issue exists there as well
1237[18:26:19] <dacencora> After disabling nvidia, of course
1238[18:27:01] <oiaohm> jhutchins: not really. replaced-url
1239[18:29:09] <oiaohm> jhutchins: big thing is swap in Linux means that its altered in the memory map to no longer be accessible without generating a page fault to pull it back out of where even it been stored.
1240[18:30:25] <oiaohm> Due to the fact Linux kernel can in fact use memory as swap using the swap system to sort of structure defragmentation issues is classed as an acceptable path.
1241[18:33:32] <oiaohm> jhutchins: also fun thing replaced-url
1243[18:34:39] <Zevv> Hi #debian; I have a 11 year old unstable install that I have been updating weekly or so, never any problems. Since last week I have some funny font issues - some missing glyphs in my urxvt, and firefox renders some websites with a strange fat font. I'm not sure how to approach debugging this, any hints?
1244[18:35:10] <greycat> !debian-next
1245[18:35:10] <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
1251[18:38:28] <gordonfish> jelly: It can be, but I sometimes get annoyed when someone asks me "how to add octane" to their computer, since they vaguely heard about it somewhere
1252[18:39:55] <Zevv> Hi all. I have a 11 year old unstable install that I have been updating weekly or so, never any problems. Since last week I have some funny font issues - some missing glyphs in my urxvt, and firefox renders some websites with a strange fat font. I'm not sure how to approach debugging this, any hints?
1263[18:48:10] *** Quits: dacencora (~logan@replaced-ip) (Quit: Lost terminal)
1264[18:48:12] <Copenhagen_Bram> How do I modify the boot config on a Debian live USB?
1265[18:49:07] <Copenhagen_Bram> If I mount it from the USB and try to edit grub.cfg, it's read-only. I tried to edit the ISO with ISO Master, and Etcher says the result has a missing boot partition.
1266[18:49:19] <Copenhagen_Bram> err, missing partition table, that is.
1267[18:51:14] *** Quits: Witch-King (~Witch-Kin@replaced-ip) (Remote host closed the connection)
1287[19:18:47] <somiaj> Copenhagen_Bram: what is the use case, why do you need to modify this (note debian .iso are hibred, so modifing them isn't really possible without rebuilding the whole image)
1288[19:19:01] *** debhelper sets mode: +l 1121
1289[19:19:06] *** Quits: aliasnumber10 (~crix@replaced-ip) (Remote host closed the connection)
1292[19:20:08] <Copenhagen_Bram> The use case is I'm trying to make a Debian Live USB with encrypted persistence.
1293[19:20:10] <somiaj> !hybrid images
1294[19:20:10] <dpkg> Since the 6.0 "Squeeze" release, Debian installation images for x86 systems - e.g. <netinst>, <CD1>, DVD1 - are hybrid images. These can be written directly to CD or HD Media (e.g. USB thumbdrive) without further preparation. See replaced-url
1295[19:20:28] <sney> the debian live manual is here, and has instructions for building your own: replaced-url
1296[19:20:35] <Copenhagen_Bram> Alright
1297[19:20:49] <somiaj> You might have to build your own image for that, unsure if this is something that can just be modified without messing up the image.
1298[19:20:55] <sney> I don't know if it covers persistent live, but there are blogs and howtos around the internet that can be adapted for persistence
1299[19:20:57] <mutante> isn't "live" and "persistence" like the opposite of each other?
1300[19:21:20] <sney> not really, you can have an immutable OS and a writable /home
1301[19:21:31] <mutante> ah, just for /home. ok
1302[19:21:41] <somiaj> Copenhagen_Bram: what method are you using for persistance, maybe adapt the method (some I think manually copy the iso contents and use syslinux)
1303[19:21:43] <Copenhagen_Bram> Well, all I have to do is add some boot options to /boot/grub/grub.cfg and /isolinux/menu.cfg
1305[19:24:12] <Copenhagen_Bram> somiaj: LUKS encrypted partition with an ext4 filesystem labeled "persistence" with a file called persistence.conf containing "/ union". Then edit the boot menus for grub and isolinux to include an option called "Debian Live with Encrypted Persistence" that's identical to the default option but with some options added next to splash and quiet: persistent=cryptsetup persistence-encryption=luks
1306[19:24:14] <Copenhagen_Bram> persistence
1307[19:24:23] *** Quits: tgunr (~davec@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1308[19:24:24] <Copenhagen_Bram> That's the menu I'm trying to use.
1309[19:24:32] <Copenhagen_Bram> err, I mean method
1316[19:33:11] <somiaj> Copenhagen_Bram: maybe there is a way to do this, but my (limited) understanding of the hybrid images is they are very fininky, and you probably can't edit the image directly
1317[19:33:54] *** Quits: black_ant (~antilope@replaced-ip) (Quit: simplicity does not kill)
1318[19:34:03] <somiaj> using live-build and building a custom image might be a better approch, there is also #debian-live on irc.oftc.net (though unsure how much it is used these days)
1319[19:34:28] <mulletman> If i use dd if=/dev/zero of=/dev/sda bs=512 count=1 to nuke my mbr should I still be able to boot via gpt? I am getting errors that I have both and need to get rid of one of them?
1320[19:35:14] <somiaj> gpt partition tables won't have an 'mbr' section on them, so unsure what that is supose to do on a gpt partition table?
1325[19:36:39] <mulletman> well according to the two imaging programs i tried (clonezilla and redorescue) I have both GPT and MBR which is throwing those imaging programs for a loop and both said i need to delete either one or the other (MBR or GPT)
1326[19:37:10] <somiaj> I don't see how you can have both mbr and gpt partititon tables, so this seems odd.
1327[19:37:26] <wwilliam> how to find out the IP address of ehapp1
1339[19:43:33] <mulletman> somiaj, here is a pix of the error replaced-url
1340[19:44:29] <mulletman> so i guess it more correctly says "this disk contains mismatched GPT and MBR partition?
1341[19:44:55] <somiaj> oh this is on /dev/nvme0n1 -- I have heard of some strange things that can happen with nvme, but this is out of my experience.
1342[19:46:11] <mulletman> doh - oh well, I really want to make a backup image and can always use dd itself but not sure if i will be making a crap image - and wouldnt know until I had to write it back to the drive?
1343[19:46:23] <somiaj> what does a tool like (g)parted say about the partition table?
1344[19:48:26] <mulletman> not sure, can look back in a few mins
1347[19:49:23] <somiaj> But it could just be some nvme strangeness. My only single experience with nvme was it did strange things, so I ended up just scraping it and not using it
1354[19:54:58] <mulletman> well I am hoping not to have to change this, it came with the computer I got. Anyway, I booted up gparted and it doesnt seem to tell me much, the first partition's flag is "boot" but thats it, is there some sort of gparted commandline kungfu that might yeild more information?
1356[19:55:45] <somiaj> I am unsure off the top of my head. Are you not able to see what partition table is being used. I almost think fdisk can at least tell you that, what about 'fdisk -l /dev/foo'
1357[19:56:08] <greycat> I'd definitely start with fdisk -l
1364[19:59:57] *** Quits: krzych (krzych@replaced-ip) (Remote host closed the connection)
1365[20:00:32] <somiaj> I do get various google hits about issues with clonezilla and nvme, but nothing that looks exactly like your issue and no real solutions (in my quick scan)
1370[20:03:01] <mulletman> well I tried fdisk -l /dev/nvme0n and it give me information but no mbr or gpt certainly not as the disk label
1371[20:03:39] <greycat> does it have any partitions?
1372[20:03:48] <mulletman> I am thinking it might be parted or something that clonezilla uses because I had almost the same issue with redo resuce and i think it also uses some of the same utils
1400[20:20:54] <mulletman> thats a bit encouraging i think (thank you). So perhaps instead of using clonezilla or redorescue is there another imaging option out there that can verify the image after its written?
1459[21:21:09] <sney> tekcap: if you want to stay on buster with some newer packages, look at buster-backports: replaced-url
1460[21:21:30] *** Quits: forgotmynick (uid24625@replaced-ip) (Quit: Connection closed for inactivity)
1461[21:22:12] <tekcap> @sney It's because I need the latest libc-bin
1462[21:23:00] <sney> tekcap: don't @ people on irc, most clients ignore that kind of highlight.
1463[21:23:16] <sney> tekcap: if you *actually need* newer libc, then your only choice is to upgrade your system to bullseye
1464[21:24:28] <greycat> Another choice would be to set up a virtual machine, and run bullseye inside the VM.
1465[21:24:31] <tekcap> Ok, I'll upgrade to Bullseye then, is it available to download from the debian website, or it's still in testing?
1466[21:24:38] <greycat> Yes, and yes.
1467[21:24:53] <sney> dpkg: stable->testing
1468[21:24:53] <dpkg> In /etc/apt/sources.list, change "buster" to "bullseye", remove lines like buster-backports, debian-multimedia <dmo>, and other 3rd party repos as they are known to cause issues then do: apt update && apt upgrade && apt full-upgrade. Note that testing is a <moving target> and may be buggy, and read the sid FAQ: replaced-url
1491[21:41:46] <oxek> which package is it on ubuntu that suggest installation of a package if the command is not found? I'd like to get that functionality on debian too.
1508[21:50:44] <Hackwar> hi folks, I'm building a Docker image based on debian stretch for a CI system and I need PHP 5.6 in that image. So it seems I have to add an additional repository for that to work, but unfortunately I can't get that to work. How can I debug which repos apt is using?
1509[21:51:30] <greycat> /etc/apt/sources.list and /etc/apt/sources.list.d/*.list
1518[21:53:54] <mutante> so that's not all you are doing?
1519[21:54:41] <Hackwar> When I did the above stuff, I do apt-get install -y git php5.6-cli ... (some more packages) and it then tells me "E: Package 'php5.6-cli' has no installation candidate"
1520[21:55:12] <mutante> Hackwar: you likely need an "apt-get update" in between
1521[21:55:20] <mutante> to make it read the new package lists after adding them
1522[21:55:37] <Hackwar> Sorry, forgot to write that. I already have an apt update inbetween
1523[21:56:00] <mutante> and when you run that you see it connect to sury.org ?
1538[22:00:06] <Hackwar> I'm working on a larger FOSS project and our current version supports PHP 5.3 to 7.4. Our next major version is going to hopefully be released this year, supporting 7.2 and upwards. But until we are EOLing our current major version, we need to also make sure that it works on those ancient PHP versions, as much as I hate it.
1539[22:01:00] <Hackwar> Unfortunately we are running no less than 5 CI systems and I'm trying to consolidate that down to 2, which requires me to build those images, so that I can decommission 2 other CI systems.
1540[22:01:32] <mutante> hmm. see the files starting with php5.6-cli in replaced-url
1541[22:01:54] <mutante> not sure why but there is -dbgsym in the name
1542[22:02:18] <Hackwar> but after those packages, there is also a version without that.
1543[22:02:46] <mutante> right, the other ones are the debug symbols.. hmm
1544[22:02:51] *** Quits: platvoeten (~platvoete@replaced-ip) (Remote host closed the connection)
1545[22:03:09] <Hackwar> I'm a bit at a loss here...
1554[22:06:21] <Hackwar> Ok, progress. I had used the php:5.6-apache image and tried adding my stuff onto that. But it seems as if there is something wrong there. Using debian:stretch seems to work.
1587[22:26:18] <sney> unless you have security updates disabled, you didn't upgrade from 10.4 to 10.5 all at once, so it should be relatively easy to narrow down
1588[22:26:29] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
1615[22:54:33] <allan_wind> Hello friends, minor issue, when I reboot my computer one of mdadm arrays are created using /dev/md/$name whuile the others are just /dev/md1, /dev/md2. If I unmount, stop and reassmble it works as expected, but reboot creates the different name. Any ideas?
1624[22:57:30] *** Quits: otisolsen70 (~otisolsen@replaced-ip) (Remote host closed the connection)
1625[22:57:50] <allan_wind> sney: when I do -D --scan it does give me path that is created during boot so I am guessing that comes from the super block opposed to my conf
1626[22:57:50] <sney> yeah, I haven't used md recently but afaik both types of dev node are fully supported
1660[23:16:55] <|aaron_> Is it possible to load a preseed.cfg from a floppy, on buster? Doesn't look like the floppy kernel module gets loaded initially?
1663[23:23:26] <|aaron_> Nevermind, I found the answer, the floppy module isnt loaded by the installer since jessie, I would need to create my own initrd
1682[23:42:06] <greycat> Be more specific and helpful in your diagnostics. Instead of saying "doesnt work", say what *happened*. Did you get an error? What did it say? Check "systemctl status apache2". Check the logs in /var/log/apache*/.