| === elsheepo is now known as beatzz | ||
| === elsheepo is now known as beatzz | ||
| lordievader | Good morning | 08:23 |
|---|---|---|
| === Mr_Pan is now known as Mr-Pan | ||
| === Mr-Pan is now known as Mr_Pan_ | ||
| === Mr_Pan_ is now known as Mr_Pan | ||
| === iliv_ is now known as ivan_CMD | ||
| ham5urg | Hi all, I'm trying to get a diskless ubuntu 16.04 client to mount its root via NFS. First the kernel and initrd are transferred via tftp which is working. But after the kernel should mount the NFS I get an error: "read: Connection refused" and "NFS over TCP not available from 172.17.0.1". The NFS-server is up and running but the mount fails. I put all informations into one paste at https://paste.debian.net/1050556/ | 11:07 |
| ahasenack | good morning | 11:10 |
| ahasenack | kantlivelong: good, I'm curious :) | 11:11 |
| ham5urg | $ This works: "sudo mount 172.17.0.1:/srv/nfs/xenial share" | 11:20 |
| ham5urg | This doesn't: "sudo mount localhost:/srv/nfs/xenial share" | 11:20 |
| DenBeiren | lordievader: did you see my last message? | 11:35 |
| rbasak | ahasenack: thank you for the quick review | 11:42 |
| ahasenack | np | 11:43 |
| lordievader | DenBeiren: Err, maybe. Was quite busy yesterday afternoon. What was it? | 11:49 |
| DenBeiren | <lordievader> DenBeiren: Does that .config also exist when transmission is not runnign? | 11:50 |
| DenBeiren | <DenBeiren> lordievader: i believe it is | 11:50 |
| DenBeiren | <DenBeiren> https://pastebin.com/AW168bxj | 11:50 |
| lordievader | DenBeiren: Did you pastebin your config before? | 11:52 |
| ahasenack | kstenerud: what's left for https://code.launchpad.net/~kstenerud/ubuntu/+source/mysql-5.7/+git/mysql-5.7/+merge/358234 ? The salsa MP? | 12:15 |
| kstenerud | For ubuntu side it should be ready. I'm trying to get a merge request in to debian side, but I can't get mysql to build as a package :/ | 12:16 |
| ahasenack | kstenerud: mysql is an ftbfs in debian right now? | 12:16 |
| kstenerud | ftbfs? | 12:16 |
| ahasenack | (fail to build from source) | 12:16 |
| kstenerud | The builder tool just dies | 12:16 |
| kstenerud | sec. Doing one last build attempt | 12:16 |
| kstenerud | pristine buster image this time | 12:17 |
| kstenerud | Oh, on buster it doesn't fail the signature check | 12:18 |
| kstenerud | It does fail with unmet dependencies though: https://pastebin.ubuntu.com/p/nX7hwPk96f/ | 12:19 |
| ahasenack | why don't you install them/ | 12:21 |
| ahasenack | ? | 12:22 |
| kstenerud | I'm trying to figure out how | 12:26 |
| ahasenack | apt install <name>? | 12:26 |
| ahasenack | or, for a helper, use apt-get build-dep ./ while inside the extracted package directory | 12:26 |
| kstenerud | looks like mk-build-deps does everything | 12:27 |
| kstenerud | Ugh I think I'm making a big mess of things. I'm trying out autopkgtest, but it doesn't seem to like running from a git repo of mysql | 12:36 |
| kstenerud | Really, all I want to do is download mysql from debian git repo, build and test | 12:36 |
| kstenerud | run tests | 12:36 |
| ahasenack | your changes they are only in maintainer scripts, right? | 12:37 |
| kstenerud | yes | 12:37 |
| ahasenack | ok, then just submit the mp, it shouldn't affect building the package | 12:37 |
| kstenerud | ok | 12:37 |
| ahasenack | and running tests for debian, that's a big plus, considering the change was just in maintainer scripts | 12:37 |
| ahasenack | you can worry about that later if you want the experience | 12:37 |
| kstenerud | Yeah I mostly wanted to get my own documentation on the subject in order. There seem to be a LOT of different ways, and lots of advice, but no one-true-way document that says "do this" | 12:41 |
| ahasenack | well, there are many ways to isntall a package | 12:41 |
| ahasenack | the core of that error was "sorry, I can't build mysql because I'm missing this package, that package, and this other one" | 12:42 |
| ahasenack | that's why they are called build dependencies | 12:42 |
| kstenerud | I mostly want to get a procedure in place for forking from their repo, making changes, building and installing the new package, running package tests, and submitting a merge request | 12:42 |
| ahasenack | that's an awesome goal | 12:42 |
| ahasenack | there was really nothing different in this mysql git repo from debian than from any other package | 12:43 |
| ahasenack | you install build dependencies, and then build as usual | 12:43 |
| kstenerud | except that no build tool prior to the one in testing works with mysql, which I just discovered :/ | 12:44 |
| ahasenack | I'm building it in sid as we speak | 12:44 |
| ahasenack | just did what I said | 12:44 |
| rbasak | I use sbuild and autopkgtest with the lxd runner. | 12:44 |
| kstenerud | Really? So gbp buildpackage doesn't error out with certificate errors? | 12:45 |
| ahasenack | I just launched a sid container, like "lxc launch images:debian/sid debian-sid" | 12:45 |
| rbasak | However there's a bug in Debian's systemd which stops mysqld from running in a container | 12:45 |
| ahasenack | git cloned the repo | 12:45 |
| ahasenack | installed build dependencies | 12:45 |
| ahasenack | and ran dpkg-buildpackage | 12:45 |
| ahasenack | brb, need to relocate | 12:45 |
| rbasak | So I use an Ubuntu container to run autopkgtest against the Debian deb, which isn't ideal but I consider it good enough if it passes. | 12:45 |
| rbasak | I use dpkg-buildpackage to get a source package and run sbuild against that. | 12:45 |
| kstenerud | hmm don't know any of those tools | 12:46 |
| rbasak | https://wiki.ubuntu.com/SimpleSbuild but I appreciate it's Yet Another Tool so don't feel like you have to use it :) | 12:46 |
| a_ok | I have ubuntu 16.04LTS and need to boot a previous installed kernel. Grub 2 is installed selecting from menu at boot is not an option | 12:46 |
| kstenerud | I was trying to follow https://wiki.debian.org/PackagingWithGit | 12:46 |
| rbasak | And because sbuild is such a pain to set up, that's why we did "git ubuntu build" to try and make it easy. But it's still a little buggy I believe :-/ | 12:47 |
| kstenerud | also https://wikitech.wikimedia.org/wiki/Git-buildpackage | 12:47 |
| kstenerud | but it looks like git-buildpackage is still a bit shaky | 12:47 |
| rbasak | All of these tools are wrappers around what Debian policy states as the interface source packages must provide (debian/rules), FWIW. | 12:49 |
| kstenerud | rbasak: So in your experience, which tools are the best ones to use in terms of stability? | 12:49 |
| rbasak | IMHO sbuild is the most reliable and provides appropriate (but not perfect) isolation. | 12:50 |
| DenBeiren | lordievader: the content of settings.json you mean? | 12:50 |
| rbasak | You can also mess with it for debugging and development with schroot directly, the -p flag, etc. | 12:50 |
| lordievader | DenBeiren: Yes | 12:50 |
| rbasak | However sbuild is not so trivial to set up :-/ | 12:50 |
| DenBeiren | no i didn't | 12:50 |
| DenBeiren | coming up :-) | 12:50 |
| rbasak | OTOH sbuild does always work. | 12:51 |
| kstenerud | Good. I'm sure I can scriptify things for simple use cases like SRUs | 12:52 |
| DenBeiren | https://pastebin.com/31hizU4T lordievader | 12:52 |
| DenBeiren | when i try port 8083 it won't work | 12:53 |
| rbasak | kstenerud: here's a script I have lying around that configures sbuild as I like it on a fresh remote host: https://paste.ubuntu.com/p/xHr6wnd66R/ | 12:53 |
| DenBeiren | 9091 works, but limitations from rpc whitelist | 12:54 |
| DenBeiren | thats why i conclude the configfile isn't read | 12:54 |
| rbasak | kstenerud: I haven't used it in a while though so it might be a little out of date on a few bits | 12:54 |
| kstenerud | rbasak: Thanks! It's a good starting point regardless | 12:54 |
| lordievader | DenBeiren: Is this the full file? This isn't valid json. | 12:55 |
| DenBeiren | yes, there is no more in it | 12:56 |
| lordievader | DenBeiren: That is then probably why it is ignored. This is not valid json. | 12:57 |
| lordievader | Missing at least the closing } | 12:57 |
| a_ok | So its the whole grub submenu stuff that messed me up | 12:57 |
| a_ok | what does this mean? Warning: Please don't use old title `Ubuntu, with Linux 4.4.0-137-generic' for GRUB_DEFAULT, use `Advanced options for Ubuntu>Ubuntu, with Linux 4.4.0-137-generic' (for versions before 2.00) or `gnulinux-advanced-6baa6b94-0498-4047-b41f-72a78923dbe4>gnulinux-4.4.0-137-generic-advanced-6baa6b94-0498-4047-b41f-72a78923dbe4' (for 2.00 or later) | 12:58 |
| DenBeiren | the closing is in there | 12:58 |
| DenBeiren | i missed it in copypaste :-) | 12:59 |
| zzarr | Hello! I have a rented VM with a systemd-journald that's using a lot of CPU, what should I do? | 13:04 |
| ahasenack | back | 13:04 |
| lordievader | DenBeiren: Ah, I see. What happens when you run the deamon in the foreground? | 13:12 |
| ahasenack | zzarr: is "journalctl --follow" spitting out new messages at a very high rate? | 13:15 |
| DenBeiren | lordievader: could you explain a bit more,.. i don't know exactly what you expect me to do | 13:51 |
| lordievader | DenBeiren: What happens when you stop the transmission-daemon service and run `sudo transmission-daemon -f` in a shell. | 13:52 |
| DenBeiren | it's running,.. waiting for more output | 13:55 |
| DenBeiren | maybe i could try with a std settings file? | 13:55 |
| lordievader | DenBeiren: It is not giving errors? | 13:56 |
| DenBeiren | https://pastebin.com/uAmEaF43 | 13:56 |
| lordievader | DenBeiren: You could check which settings file it is loading with `sudo strace -e open transmission-daemon -f` | 13:56 |
| lordievader | Oh, never mind: [2018-11-06 14:53:45.788] Using settings from "/home/denbeiren/.config/transmission-daemon" (daemon.c:528) | 13:57 |
| DenBeiren | ok,.. found that file, and it doesn't have the settings i changed in the other file | 13:58 |
| lordievader | No, it probably never read that file. | 13:59 |
| DenBeiren | stopped TM, edited the port, saved and started TM again,.. no changes :( | 14:02 |
| lordievader | Is it still saying it loads the config from the .config dir? | 14:03 |
| ahasenack | iirc that json config file has to be changed while transmission is stopped | 14:03 |
| ahasenack | but I could be misremembering | 14:03 |
| lordievader | Correct, else it will be overwritten with the 'current' config. | 14:04 |
| DenBeiren | ok we have progress,.. | 14:07 |
| DenBeiren | if i start transmission daemon in the foreground, it seems to work | 14:07 |
| DenBeiren | when i start transmission-daemon with the command sudo service transmission-daemon start it doesn't take the correct settinge | 14:08 |
| lordievader | It might have to do with environment variables. | 14:08 |
| lordievader | My systemd service has an override wich adds `Environment=TRANSMISSION_HOME=/var/lib/transmission/config`. But this is on Gentoo. | 14:09 |
| DenBeiren | question is,.. how do i start/stop/check correctly now | 14:13 |
| DenBeiren | and how can i get the correct version to start at boot? | 14:13 |
| zzarr | ahasenack, The server spits out lines like this over and over: Nov 06 16:18:15 production rc.local[833]: 7f01a4938000-7f01a4b37000 ---p 00007000 ca:01 1845822 | 14:19 |
| ahasenack | looks like it's panicing | 14:20 |
| ahasenack | panic()ing | 14:20 |
| ahasenack | check dmesg | 14:20 |
| zzarr | ahasenack, will do | 14:21 |
| zzarr | ahasenack, I found this [ 7.800343] systemd[1]: Failed to start Load Kernel Modules. | 14:22 |
| lordievader | DenBeiren: Change the `--log-error` to `--log-debug` in the `ExecStart` line of the service file. | 14:36 |
| lordievader | DenBeiren: Then run `sudo systemctl daemon-reload && sudo systemctl restart transmission-daemon && sudo journalctl -u transmission-daemon`. That should give you enough output to see which configuration file it is reading. | 14:37 |
| lordievader | (When you are done you should revert the change to the service file) | 14:37 |
| DenBeiren | where can i find the service file? | 14:37 |
| lordievader | `sudo systemctl cat transmission-daemon` should tell you (or `sudo systemctl status transmission-daemon`) | 14:38 |
| DenBeiren | https://pastebin.com/DgSe24si | 14:46 |
| DenBeiren | permissions issue? | 14:55 |
| lordievader | DenBeiren: I get the feeling the directory doesn't exist. It tries to read the config from `/home/debian-transmission/.config/transmission-daemon`. | 14:58 |
| DenBeiren | shouldn't it have created it with the install? | 14:58 |
| lordievader | I suppose. | 15:03 |
| === JanC_ is now known as JanC | ||
| ham5urg | Has anybody tried with Ubuntu 16.04 to implement a disklessclient with NFS and overlayfs? | 15:38 |
| elfranne | Trying to find if "apt-mark hold <package>" is being respected when doing "apt-get dist-upgrade" ? | 15:43 |
| sdeziel | elfranne: it should otherwise that would be a bug | 15:46 |
| elfranne | trying to find any docs about that... | 15:47 |
| mybalzitch | boy, this new ubuntu 18.04.01 installer sure is nice | 15:47 |
| sdeziel | elfranne: that's easy to test because dist-upgrade will ask your for a confirmation that you can decline if the held package is about to be upgraded | 15:49 |
| elfranne | I ll give it a try with a fresh install without internet on a VM | 15:50 |
| elfranne | thanks | 15:50 |
| sdeziel | elfranne: FYI, I use this technique routinely and it works great | 15:52 |
| kstenerud | rbasak: I'm getting libreadline.so.6: cannot open shared object file no matter which channel I'm on (stable, beta, edge) | 15:54 |
| rbasak | That sounds familiar | 15:58 |
| SJr | I have been going through and wiping dozens of hark disks through an ejectable dock. I have been getting smartctl errors e-mailed to me complaining about failures of a disk I ejected a few days ago. When I look in /dev/ I still see the disk, but hdparm -I just gives garbage for it. Is there a way to fix this without rebooting? | 16:26 |
| sdeziel | SJr: have you tried "eject /dev/sdX" ? | 16:27 |
| SJr | unable to eject, last error: Invalid argument | 16:27 |
| sdeziel | SJr: that needs to be done prior to the real physical ejection/removal | 16:27 |
| SJr | Oh well | 16:28 |
| sdeziel | SJr: echo 1 > /sys/block/sdX/device/delete | 16:28 |
| sdeziel | might work after the fact | 16:28 |
| SJr | yup it did | 16:29 |
| SJr | Thank you | 16:29 |
| sdeziel | you are welcome | 16:29 |
| SJr | Oh hrm, well good news/ bad news. Now I just got an e-mail saying that smartd could not open /dev/sde | 16:30 |
| SJr | Well I guess I'm just going to restart the service | 16:31 |
| sdeziel | it's reassuring that smartd will let you know when a drives suddenly vanishes from existence ;) | 16:32 |
| === trekkie1701c is now known as AndroUser | ||
| === AndroUser is now known as trekkie1701c | ||
| mybalzitch | is releases.ubuntu.com down for anyone else? | 17:03 |
| Glorfindel | mybalzitch: it's up for me | 17:04 |
| mybalzitch | and lol, the 18.04.01 installer errored out trying to probe my ZFS install, because it can't find the zpool command | 17:04 |
| mybalzitch | https://i.imgur.com/4UztzIU.png | 17:05 |
| ahasenack | https://bugs.launchpad.net/subiquity/+bug/1782744 | 17:08 |
| ubottu | Launchpad bug 1782744 in subiquity "Install failed on disk previously used for ZFS" [Medium,Confirmed] | 17:08 |
| ahasenack | you will have to switch to another console (alt-f2) and erase that disk with something like sgdisk -Z | 17:09 |
| ahasenack | iirc (the -Z part) | 17:09 |
| mybalzitch | thanks ahasenack | 17:10 |
| sdeziel | zfsutils-linux should be in the ISO IMHO | 17:11 |
| sdeziel | would make recovery much easier | 17:11 |
| Ussat | I dont think licensing makes that possible | 17:16 |
| sdeziel | Ussat: the zfs.ko (and friends) module is shipped in the ISO | 17:18 |
| sdeziel | so I'd think the userspace tooling would be OK | 17:19 |
| Ussat | OK, I m not a lawyer, was just a guess. RHEL (which I use a lot also) has taken that stance | 17:19 |
| mox- | I'm having some issues building an ISO that boots on the network to reach a Kickstart file to do an unattended installation. It is working great with RHEL/CentOS but with Ubuntu I keep getting "ipconfig: no devices to configure" when the ISO boots. I searched google I can see examples of PXE installations but no bootable ISO installation with a Kickstart. Anyone have an idea what's going on or a reference document for me? Thanks | 19:05 |
| granjero | hi, ubuntu 18.04 server. how to mount window share? cifs mount giving me erro mount error(16): Device or resource busy | 19:08 |
| granjero | 19:08 | |
| ahasenack | will need more info about that error | 19:12 |
| ahasenack | how you are trying to mount it, if dmesg has any other error messages, if you have tried another pointpoint | 19:12 |
| granjero | sudo mount -t cifs -o username=USER,password=MYPASS //SHAREIP/folder /home/myshare | 19:13 |
| granjero | mount error(16): Device or resource busy | 19:13 |
| granjero | Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) | 19:13 |
| ahasenack | /home might be another mountpoint already, can you try mounting it elsewherE? | 19:19 |
| ahasenack | and did you check dmesg? | 19:19 |
| granjero | let me see | 19:19 |
| granjero | was trying to remember how to do it. the dmesg part | 19:20 |
| mox- | just type 'dmesg' | 19:20 |
| granjero | was the SMB prtotocol version | 19:21 |
| ahasenack | yeah, there is a new default | 19:25 |
| granjero | sudo mount -t cifs -o vers=1.0,username=USER,password=MYPASS //SHAREIP/folder /home/myshare | 19:25 |
| ahasenack | "device or resource busy" is not the error message I was expecting for that, though | 19:25 |
| ahasenack | what windows server is that? | 19:25 |
| ahasenack | xp? | 19:26 |
| granjero | also got mount error(13): Permission denied | 19:26 |
| granjero | 2013 | 19:26 |
| ahasenack | that is rejecting higher versions of smb? | 19:26 |
| ahasenack | have you tried vers=3.0? | 19:26 |
| granjero | does vers=3.0 exists? | 19:27 |
| ahasenack | yes | 19:27 |
| ahasenack | check the mount.cifs manpage, search for vers= | 19:27 |
| ahasenack | it has a nice table of protocol versions and windows releases | 19:28 |
| granjero | 3.0 - The SMBv3.0 protocol that was introduced in Microsoft Windows 8 and Windows Server 2012. | 19:28 |
| granjero | vers=3.0 also worked | 19:30 |
| granjero | now i will try to so it on mount with fstab | 19:30 |
| granjero | wish me luck | 19:30 |
| granjero | *do it | 19:31 |
| mox- | Is there a specialised channel for network boot installation help? I have a very specific question that I can't find the answer | 19:32 |
| jayjo | is there consensus on the leading FOSS for highly available shared sstorage? Ceph, glusterfs, just NFS, ... ? I have a cluster of ubuntu servers that run containers of a gitlab instance, when it was all local on a node this worked fine. But now I need to share the data across the cluster. If I were implementing this today (no support restrictions), what's the best solution? | 20:06 |
| blackflow | the industry is betting on ceph | 20:20 |
| chillage | Good time, | 20:37 |
| chillage | Kazaaak, hello, | 20:38 |
| chillage | we with kazaaak have issue, paste kazaaak it | 20:43 |
| Kazaaak | sudo ufw status | 20:44 |
| Kazaaak | WARN: uid is 0 but `/usr` is owned by 4011 | 20:44 |
| Kazaaak | WARN: /usr is group writable! | 20:44 |
| chillage | what's here is wrong? | 20:46 |
| sarnold | namei -l /usr | 20:47 |
| teward | Kazaaak: chillage: what's the output of the command sarnold said? | 20:50 |
| teward | because those errors indicate something went crazy on your system and changed permissions in a non-safe way | 20:51 |
| chillage | we see teward, wait little bit, kazaaak will paste | 20:54 |
| Kazaaak | root@vnc:/home/appbox# namei -l /usr | 20:54 |
| Kazaaak | f: /usr | 20:54 |
| Kazaaak | drwxr-xr-x root root / | 20:54 |
| Kazaaak | drwxrwxr-x 4011 4011 usr | 20:54 |
| Kazaaak | root@vnc:/home/appbox# | 20:54 |
| sarnold | ouch. | 20:55 |
| sarnold | something screwed up *badly* | 20:55 |
| teward | owchies. | 20:55 |
| teward | what sarnold said | 20:55 |
| teward | ***THERE BE DEMONS IN THESE WATERS!*** | 20:56 |
| teward | *shot* | 20:56 |
| chillage | sorry, my friend paste in channel not in pastebin? or whats wrong with system? | 21:00 |
| teward | chillage: no, we're commenting about the permissions | 21:00 |
| sarnold | chillage: something is wrong with your system. *something* running as root did something terribly stupid. | 21:00 |
| teward | let me guess: this is the appbox werver you were asking questions on about a few days or so ago? | 21:00 |
| sarnold | chillage: the usual problem is someone unpacking a tarball into / while root without knowing what the heck they're doing, but it could just be terrible software too. | 21:01 |
| sarnold | chillage: the worst part of this problem is that there's no clear way to know just how bad things are. maybe it's literally this one directory in which case the fix is five seconds of work. or maybe a *huge* number of files and directories have had their permissions set incorrectly.. in which case it's easier to resintall | 21:02 |
| chillage | yes, teward few days ago abuot appbox | 21:03 |
| sarnold | chillage,Kazaaak, pastebin ls -l /usr output, lets see just how bad this thing is.. | 21:08 |
| sarnold | I *really* wish we had something like mtree to notice these problems ;( | 21:08 |
| teward | sarnold: full system auditing :P | 21:09 |
| teward | audit all the permission changes :P | 21:09 |
| chillage | sarnold, and how i can fix it if problem is only /usr ? | 21:10 |
| sarnold | teward: heh, also a good idea. then we'd know what to shoot ;) | 21:10 |
| teward | sarnold: what... or *who*... to shoot :p | 21:10 |
| Kazaaak | https://pastebin.com/n4N9tpa1 | 21:12 |
| sarnold | OUCH | 21:13 |
| sarnold | I've got to run | 21:13 |
| teward | ERROR666 Nuke It From Orbit | 21:13 |
| sarnold | but this one's bad enough that you might want to consider a reinstall | 21:13 |
| teward | Kazaaak: chillage: I'm in agreement with sarnold | 21:14 |
| sarnold | lunch time here :) have fun | 21:14 |
| teward | a reinstall would probably fix things, you're going down a rabbit hole of hell if you try and fix this manually | 21:14 |
| chillage | understand, then thanks for help, guys:) we try reinstall it.. | 21:17 |
| Goop | Hey, I am interested in doing a couple of things, and I am too cheap (and cannot afford) to get G-suite (Google) or Microsoft products for my sort-of company. I have a Linux server I am currently running and wanted to know what you guys recommend for software for creating new users for a company? | 21:39 |
| blackflow | useradd | 21:42 |
| Goop | I mean, like one where I can type in a user into a system, which includes user first name, last name, title, password, phone number, address, etc. etc., which then creates a user account for everything else that the company needs the person to use. | 21:50 |
| blackflow | Goop: that's a very very broad use case. Users are usually application specific. Like, user account "to what". | 21:52 |
| Ussat | you can always script your own solution to do that , but useradd | 22:01 |
| blackflow | otherwise sounds like LDAP, come to think of it | 22:04 |
| cryptodan_mobile | Sounds like single sign on via pki cert | 22:49 |
| rbasak | ahasenack: one for you perhaps: https://www.reddit.com/r/Ubuntu/comments/9usqau/livepatch_isnt_shown_as_enabled_in_motd/ | 22:50 |
| rbasak | Though it's offtopic there really. | 22:50 |
| sarnold | Goop: take a look at /usr/local/sbin/adduser.local in adduser(8) | 22:52 |
| Goop | sarnold, blackflow so what I haven't done too much with LDAP/OAuth/SingleSignOn, but I need something where I can have a graphical (web) interface where I can add new *people* to a company. When I submit adding a new *person* to said piece of software, I want it to setup their users across all the systems we use. | 22:54 |
| === kallesbar_ is now known as kallesbar | ||
| cryptodan_mobile | Goop then you want pki but with that the users will need to request access | 23:48 |
| blackflow | Need help understanding MOTD. on login I see it displayed on a server installed from server ISO. But none of the debootstrap'ed installations (with ubuntu-minimal as base) do that. Both, however, have the login pam module, and both have libpam-systemd. | 23:50 |
| sarnold | how about pam_motd.so ? is that configured in your /etc/pam.d/ ? | 23:51 |
| rbasak | blackflow: compare /etc/update-motd.d/ and use dpkg -S to find out which packages ship things in there. | 23:51 |
| blackflow | I also see differences in what appears to be session management. The server that does display motd, shows logging by logind starting a sessio and user slice being created. not so on the server without motd | 23:52 |
| blackflow | rbasak: I have, but I don't see anything obvious. I think I need to find what creates /run/motd.dynamic | 23:53 |
| rbasak | update-motd | 23:54 |
| blackflow | its' not listed by apt file so it's not somethign installed by packages, and grepping for that in /etc/update-motd.d/ shows nothing | 23:54 |
| sarnold | motd-news.timer | 23:54 |
| blackflow | rbasak: no, update-motd is for manual invocation of /etc/update-motd.d/50-motd-news. at any rate, the server that DOES show motd, does not have update-motd installed | 23:54 |
| blackflow | motd-news.timer only starts the 50-motd-news script, I tried that... | 23:55 |
| blackflow | I tellsya, MYSTERY. | 23:55 |
| sarnold | hmm | 23:55 |
| rbasak | Install ubuntu-server^ and I bet you'll get the same behaviour. | 23:55 |
| blackflow | I did, I did not :) | 23:55 |
| sarnold | I know every time I look into this it's a lot more complicated than I'd expect :) heh | 23:55 |
| blackflow | I installed ubuntu-standard and ubuntu-server metapackages, it pulled in a megaton of stuff, but that didn't change anything. | 23:55 |
| blackflow | sarnold: it'sa black box of black magick, mystery, spiritualism and international politics, all combined into one. | 23:56 |
| sarnold | lol | 23:56 |
| blackflow | I know.... I'll chattr /run/motd.dnymic and see what complains about not being able to update it | 23:57 |
| blackflow | ho-humm.... wait.... can I use tmpfiles to do that? I mean that's /run/ so.... tmpfs... | 23:57 |
| rbasak | update-motd is for all of /etc/update-motd.d/ | 23:57 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
