[04:04] good morning to all [05:59] https://extensions.gnome.org/extension/669/wobbly-windows/ [05:59] lol [06:26] good morning [07:19] !info ubuntu-unity-desktop [07:19] ubuntu-unity-desktop (source: ubuntu-unity-meta): The Ubuntu Unity desktop system. In component universe, is optional. Version 0.1 (bionic), package size 2 kB, installed size 8 kB (Only available for amd64; arm64; armhf; i386; ppc64el) [07:19] ducasse: ^ [10:16] Howdy folks [10:49] hi BluesKaj - all well today? [10:50] hey ducasse , yeah fine here, and you? [10:51] i'm ok, just come home from shopping groceries [10:52] raining a bit, although not much [10:55] it's dry as a bone here, no rain for almost 3 weeks [10:55] or very little [10:58] damn uefi, users ar still having trouble , even with auto installs on clean hdd-disks [10:58] we've had a little bit, but not enough to make a real impact [10:59] uefi has never given me any problems, but i think i'm lucky enough to have a fairly good implementation [11:00] yeah, had no trouble on my laptop, but I used legacy on the desktop ..still boots very fast less then 10secs from post page to desktop [11:01] counting the login [11:02] to me uefi is useless since I don't use windows anywhere [11:03] it's nice if you boot more than one distro on one machine [11:04] my laptop had problems without uefi so i finally succumbed and followed a very well written tutorial to install Kubuntu using gpt/eufi [11:05] what's the advantage over legacy with 2 distros on one pc? [11:09] i just think it's handled in a nicer way, each distro gets a subdir on the efi partition [11:17] And other OSes don't stamp on the bootloader, so it plays nicely with Windows etc [11:22] no more windows on my machines [11:22] !info roxterm cosmic [11:22] Package roxterm does not exist in cosmic [11:23] BluesKaj: me neither, ubuntu everywhere here [11:24] !info roxterm xenial [11:24] roxterm (source: roxterm): Multi-tabbed GTK+/VTE terminal emulator - binaries. In component universe, is optional. Version 3.3.2-1 (xenial), package size 123 kB, installed size 409 kB [11:25] aha [11:25] I have 3 Kubuntu machines and one RPI3 with Raspbian [11:25] little gap [11:25] lotuspsychje: ^^ that was when it was last in the repos [11:25] cool ducasse [11:25] BluesKaj: what purpose has your pi? [11:28] basically a toy at first , then I thought of it as an alternative pc in case my other/older pc died, but then my son gave me a new MSIZ370 board with 15cpu and 16GB Ram m so now the RPI3 is basically idle [11:29] kk [11:29] so it's just a toy again [11:33] guess I could use it as a Kodi box, but I have Kodi on 2 other machines already [11:34] looks like the community wiki's CSS got broken? https://help.ubuntu.com/community/InternetRelayChat/Registration [11:34] main page now https://help.ubuntu.com/community/CommunityHelpWiki [11:35] !register [11:35] For information on registering your IRC nick, see https://help.ubuntu.com/community/InternetRelayChat/Registration - For any further help, ask in #freenode. [11:35] tomreyn: forward it to the ops plz [11:35] main page on jan 2 http://web.archive.org/web/20180103071122/https://help.ubuntu.com/community/CommunityHelpWiki [11:35] lotuspsychje: which ones? [11:36] ircops seems like the wrong place to report it [11:36] tomreyn: thanks for the bug updates; the reproducer is a pain. I was up until 4am working on it last night. Feeling very frazzled right now :) [11:37] guess i'll report the broken css to #canonical-sysadmin although community stuff isn't really their domain. [11:37] or perhaps wiki team [11:37] TJ-: i read over it, was worrying about your sanity ;) [11:38] LOL yeah, it's a weird one...they named the function well though ..._racy_ ... I'm thinking there's a timing issue somewhere [11:38] not really, but you seem to do a good (maybe too good?) job on focussing on issues [11:39] i can be the same if i really want to work something out. [11:39] But I still am annoyed there's no explanation why Ubuntu is still using a 2012 release of polkit and backporting rather than keeping with upstream! Doesn't make sense, and I bet that is where the issue is [11:39] yes this strikes me as wrong, too [11:39] I read through all the commits on upstream and there's been some major changes. [11:40] so you'd have to know about all of those, and their implications, in backporting. Canonical has never been great at that kind of thing either [11:41] For those wondering what we're talking about: possible regression due to a CVE patch backport in Policykit. Bug #1784964 [11:41] bug 1784964 in policykit-1 (Ubuntu) "Regression due to CVE-2018-1116 (processes not inheriting user ID or groups )" [Undecided,Confirmed] https://launchpad.net/bugs/1784964 [11:42] debian also got 0.105 everywhere excpet in experimental [11:42] right, and Ubuntu syncs from debian then backports! What a mess [11:42] i guess this can be why, see its all their fault!!11 ;-) [11:42] I wonder if there's anything in the mailing lists as to why [11:43] you'll be all frazzled tomorrow morning. [11:43] lol [11:44] I'm trying to have a day off today! Been hard at programming since last Saturday. Crash-course in Ruby/Vagrant to hack vagrant-mutate to fix bugs before I can move on with another project! [11:51] Don't get me wrong, I'm glad you're spending this much effort on it, i guess it can and will affect many users. [11:53] I don't like bugs that can pop up like this at inconvenient moments. That usually then happens at the worst possible time. Like me yesterday - only fired up this PC to access some devices only it has the credentials for... should take 5 minutes... 3 hours later still trying to figure out why I can't read logs :) [11:53] that bot attack just resumed on #ubuntu [12:01] its all over BluesKaj freenode all channels, other networks [12:07] couldn't they make the channel +r (at least for a few days), it worked well on the few channels I 'live' in [12:07] most of the Ubuntu channels were +r yesterday [12:07] where's hggdh when needed!? :) [12:08] what a bad timing for .1 [12:08] It's a great way to collect lists of compromised IP addresses that should be blacklisted though :) [12:09] yeah, was looking back & was done to this channel ... unit193 did it on some channels... [12:09] this botnet is surely an insider, one of his bots name in #gnome this morning was 'bazhang7' [12:09] nah, that's just the way it is generating nicknames from those it sees [12:10] what gets me is the text is generally the same, the servers themselves could k-line any client that sends that text without sending the text onto the network [12:11] TJ-, i agree; that's the way I'd stop it (until I worked out a smarter way..); though maybe some coder is testing a better fix (just hasn't got it working yet.... [12:12] they must be slower than me learning Ruby then :p [12:12] I've deployed complex lambdas but got no idea how they achieve what they do :D [12:13] all because Ruby doesn't have the concept of static local variables in methods [12:14] :) [12:30] tomreyn: so, we have another victim... using KDE [13:01] most of the namespace is under +r; some channels decide for themselves, some other we enforce [14:05] hi everyone [14:06] Hi pauljw [14:06] morning BluesKaj [14:07] praying for rain here [14:07] been a while since the last? [14:08] :( sorry about that, we've been luck here and have had plenty. [14:08] right [14:08] almost 3 weeks since [14:08] ouch [14:10] bush is tinder dry ....forest fires all over our region ..not real close to us, but the potential is there [14:10] that's pretty scary [14:11] 20 min t-storms don't do much except start new fires if the lightning strikes [14:12] that's not good. [14:13] i joke about hating mowing the yard, but would much rather have to cut twice a week than to have worry about it catching fire... [14:14] a major hwy might be closed if the fire nearby gets much closer than 3 miles....the smoke is already causing visibility problems [14:19] hope my daughter can get through tomorrow on that hwy without having to do a detour [16:32] With our IRC ad service you can reach a global audience of entrepreneurs and fentanyl addicts with extraordinary engagement rates! https://williampitcock.com/ [16:32] I thought you guys might be interested in this blog by freenode staff member Bryan 'kloeri' Ostergaard https://bryanostergaard.com/ [16:52] TJ-: i see, so it's definitely not gnome / gtk related, good to have this ruled out. [16:55] it's a weird one for sure. obviously related to login procedures. I think it's something to do with systemd session management - something like a lightdm session doesn't associate UID correctly with the user-slice, but console login does. That'd explain why the order of console terminal vs GUI terminal makes a difference [16:57] i also noticed libpam-kwallet* were updated, and I see lightdm/kwallet error messages in syslog. So far I've not thought of a way that would cause the symptom though [16:58] TJ-: does screen locking work for you on the affected system, though? [16:58] I believe so, but not tried it. I'll test. [16:58] my other issues are chromium-browser taking forever to load and the system not shutting down properly becuse session manager fails to shut down my users' session [16:59] tomreyn: Yes [16:59] okay so i guess i do have another issue on 16.04 [16:59] an additional issue [16:59] hggdh: im idling in #gnome they just have +cnt, spambots all day [17:01] do you see "lightdm: pam_kwallet5(lightdm:session): pam_kwallet5: Impossible to write walletKey to walletPipe" in auth.log ? [17:05] TJ-: plenty, but it's a couple hours ago now: https://paste.ubuntu.com/p/Rz5PMY4648/ [17:07] tomreyn: right, so just like 16.04 where it used to complain that pam_kwallet.so not found all the time :) [17:08] TJ-: this IS 16.04 [17:08] tomreyn: is that a liveISO instance? I notice mine shows "pam_kwallet5: final socket path: /run/user/1000/kwallet5.socket" <--- note the path diff [17:09] i dont have an affected 18.04 system at this time. [17:09] tomreyn: ahhh, 16.04 possibly used /tmp instead. Can't recall seeing that on 16.04 before upgrade though [17:10] ~/.xsession-errors contains this line: dbus-update-activation-environment: setting PAM_KWALLET_LOGIN=/tmp/kwallet_user1.socket [17:10] and this docket exists. [17:10] *socket [17:21] oh, thanks for making me look! I found something suspicious but launchpad won't accept comments right now. [17:21] Just noticed in $HOME/.xsession-errors the following: [17:21] (polkit-gnome-authentication-agent-1:4029): polkit-gnome-1-WARNING **: 15:04:54.498: Failed to register client: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.gnome.SessionManager was not provided by any .service files [17:23] lotuspsychje: yes indeed, but it seems the bots are learning, and (for the moment) not getting in channels where they found +r [17:24] lotuspsychje: since +r is rather restrictive, we would like to take it out as soon as possible, We can set it on again if needed [17:26] Suddenly the support question about Hiri makes total sense where it came from [17:27] posted yesterday: https://itsfoss.com/hiri-email-review/ [17:27] i got those in syslog, TJ [17:28] but i think those have always been all over the place [17:35] 16.04.5 released [17:41] yay! [17:59] pragmaticenigma: oh goodness, good luck [18:01] nacc: It's a ... if it's more than a keyboard mapping issue... I'm going to tell them to backup what they can, wipe the machine and start over [18:05] pragmaticenigma: i'm super confused by their symptoms. Are they logged in now? How do they know their password doesn't work, etc? [18:06] I think they're local system is set to autologin to GUI, which allows them access right now. I honestly have a feeling they just forgot they changed their password [18:12] right [18:12] could easily be [18:19] I wish them the best [18:46] https://marc.info/?l=linux-kernel&m=153323015508428&w=2 [21:39] * hggdh wonders how bad Red Hat is doing with a 3.6 kernel officially supported on RH 6.9 [21:39] heh [21:39] they have some good kernel engineers ;) [21:40] oh, but it is outdated! [21:40] i have no idea how they deal with spectre/meltdown, if they do [21:40] :-) [21:40] heh [21:41] they probably backported somehow. Given the size of the kernel teams there, I would not be really surprised [21:41] yeah, I'm just not even sure how you'd backport some of those changes [21:41] they were hard enough to move back a few ubuntu versions [21:42] I know. But there must have been a LOT of pressure on the kernel team there to Do Something. [21:42] yeah, absolutely [21:43] their customers would have happily paid for it [21:44] oh, I am sorry, just checked at one of the servers. RH 6.9 running with kernel 2.6 [21:45] even better! :) [21:45] :-) [22:04] my my, making old android phones look current ;D [22:07] daftykins: arch is the best thing for serious devops [22:07] you know it! :) [22:07] Lol [22:08] second to a gentoo stage 1 install of course [22:08] if that was their term for the compile-everything one [22:08] But you don't understand all the flaws [22:09] that's true, i must only use something where i know the true extent of broken-ness [22:09] Lol [22:09] for the other guys - we had someone arguing that arch should be used on production servers because it's the only distro you can install 'minimal' - and then it's the most stable because they know all if its' flaws [22:10] xD [22:14] mwsb: i'm imagining zipper running servers as being a dude spinning plates on sticks now :D [22:15] "it is ok, because we know all the problems, 1) the plate could crack 2) gravity" [22:16] Lol [22:24] \o daftykins - picked up your bike? [22:24] sure did - and am very disappointed [22:47] hi [22:48] lo [22:49] i have just finish ubuntu 18 installation on my computer [22:49] and it works perfectly [22:49] so happy [22:49] ;-) [22:52] !yay | toto_ [22:52] toto_: Glad you made it! :-) [22:54] i'm discovering it and i like everything for the moment [22:54] because usually i install debian or fedora [22:56] i'm french and you? [22:57] toto_: Arkansas ridge runner . [23:00] nick toto_fr === toto_ is now known as toto_fr [23:01] here's a current flash utility for insyde bioses on (amongst other) linux (/Linux/x64/x64/H2OFFT-Lx64): http://download.udoo.org/files/UDOO_X86/UEFI_update/UDOOX86_B02-UEFI_Update_rel104.zip [23:01] most acer's require this. [23:02] tomreyn: Noteing :) .. acer. what a pain . [23:02] source: https://www.udoo.org/forum/threads/updating-fw.7729/ https://github.com/malaire/misc/tree/master/udoo-x86 [23:02] it's for udoo's new x86 gear [23:03] but the firmware writer is generic [23:04] https://www.insyde.com/products/developertools [23:04] also worth a bookmark https://firmwaresecurity.com/tag/insyde-software/ [23:05] exit [23:06] tomreyn: Yup - 'nother good one :) [23:06] apparently the proper pronounciation of UEFI and SCPI are the hot topics in insyde firmware development these days [23:06] and yes, acer is the pain [23:07] surely acer do windows utilities :) [23:08] tomreyn: Wifey runs her graphic station on an Acer machine ( Win10) .. I dread if and when she wants ubuntu back . [23:08] :) [23:09] * tomreyn crosses fingers [23:10] 10 wfm ^_^ [23:51] tomreyn: in case it's relevant, IntelCore is in my /ignore [23:51] :) i guess they should be on mine, too [23:51] thanks [23:52] indeed all Core and Core 2 systems should be ignored now too ;) [23:54] daftykins: that was my initial thought months ago when they first started asking non-ubuntu questions in #ubuntu, as well :) [23:54] daftykins: also, no one who is actually associated with Intel would put it in their nick [23:54] maybe intel pentium N4000, too [23:58] just trying to give them a last reasonable chance before i decide to /ignore