[02:40] nicomachus: document for next time ;) [02:41] forget to move the printer to your next' place, problem solved [02:41] XD [02:41] morning team! [02:41] creme de la creme [02:42] My dog woke me up for a cookie. [02:43] must learn him how to browse to oerheks.nl [02:44] hehehe [02:44] perhaps time for an electronics dispenser machine project! [02:57] daftykins: it wasn't easy... WPS didn't really work out. had to connect directly to the wireless that the printer puts out, find the LAN IP, get onto the EWP that way, and then enter the login info for the printer. [02:57] s/printer/network/ [02:59] I'm sure there was an easier way... but... oh well. [03:00] mmm a real pig when you're out in the wild and don't know the IP of a device [03:01] if I had a better handle on nmap or traceroute or something I think it coulda been easier... but idk. [03:08] i think i'd sooner factory reset a device [05:53] good morning to all [05:53] morning lotuspsychje [05:53] hey there EriC^^ [05:53] greetings o/ [05:53] hi daftykins :p [05:53] hey daftykins [05:54] lets c whats new in ubuntu land [05:54] this "The Night Manager" series seems pretty good if you folks can stand British accents :) [05:55] didnt hear of it yet [05:56] http://www.omgubuntu.co.uk/2016/05/ubuntu-image-size-increase-2gb [05:56] uh-oh [05:59] have you seen 'the website is down' [05:59] came across it yesterday [06:00] haha yeah, the guy playing games whilst getting bad support calls [06:01] lotuspsychje: concerning news, whilst i don't see any problem in the OS growing... it has to be for good reason, hopefully not just inefficiency creeping in [06:02] yeah avoiding the bloatware [06:02] iirc the last big jump was language support [06:03] seems to be what this article mentions yeah, i've seen more recent versions ask to download language support post install [06:03] dax: more countries? [06:04] after install clicking the language always asks me to update aswell [06:04] * dax shrugs [06:04] ubiquity confuses me a lot, i'm more the debian-installer type [06:05] i had a language ubiquity bug on xenial alpha also [06:06] final got that solved [06:06] heh my islands got set to US keyboard for about 5 years :| bug went ignored for the entire time [06:06] ouch [06:07] some hardware we got the other day had the keyboard set to UK English [06:07] kinda useful being a UK expat, I guess. rest of the office was floundering trying to put in an email address :) [06:08] hehe [06:10] dax: one for you: https://www.howtoforge.com/tutorial/perfect-server-debian-8-4-jessie-apache-bind-dovecot-ispconfig-3-1/ [06:10] eww control panels :P [06:11] i do need to set up bind some day though, it and samba are pretty much the only common daemons i haven't [06:11] :p [06:11] article came across lxer [06:12] need to pop to a clients and prep a test VPN server later! [06:32] kk [07:47] bbl movietime [07:47] :D [10:47] cfhowlett: howdy [10:48] ? yowza, m 'lord [10:48] cfhowlett: do you know an online service that can make wallpaper for desktop/mobile/logo for a company? [10:48] im looking for something universal [10:48] you mean graphically design it and all? [10:48] i found many company for just a company logo [10:48] but none of them make a wallpaper for pc/mobile [10:49] yes [10:49] ah. don't recall seeing same anywhere. sounds like a possible startup though, doesn't it. [10:49] yes yes, this year ill start :p [10:50] let me know! I can do remote work! [10:50] well i have already thinked of that [10:51] to propose you guys $$ for remote work [10:51] heck yes!!! [10:52] cfhowlett: would be neat to ssh into a new customers machine, and let you guys tweak it for its needs [10:52] i plugin cable and let you do some work [10:52] gotta be a way, yes? [10:53] escpacially when business grows [10:53] but also for install issues ill hang here [10:54] cfhowlett: is it hard to make a mobile wallpaper for example? [10:54] you can do a reverse shell [10:55] http://www.mobileswall.com/ [10:55] like he types bash ......... your ip [10:55] and you get a shell at his pc [10:55] EriC^^: nice idea [10:55] lotuspsychje, no. get the size specs, design and export to a supported file format. [10:56] ok [10:57] TJ-: wb [10:59] EriC^^: doesnt matter that the machine would have dutch language to do some remote work? [10:59] lotuspsychje: yeah np you can use export LANG=C [10:59] most customers will be dutch & french [10:59] aha [11:00] thanks... laptop had suspended whilst I was out running :) [11:00] thats why those clevo machines would be nice EriC^^ as keyboard is .be and .usa choosable [11:00] aha, yeah [11:01] TJ-: was saying this morning, how i already thought of it what an impact unity8 is for ubuntu's future, they want it on every machine like MS right? [11:01] like that lumia comercial, with windows on all devices [11:02] and MS want patent on the sort of convergence [11:02] EriC^^: is that a shell setting or package install? [11:03] lang=c [11:03] lotuspsychje: let me get it for you 1 sec [11:03] oh [11:03] yeah it's a environment variable [11:03] LANG=C [11:03] EriC^^: can you explain how it would work [11:04] the problem to me is that Canonical is driving this idea for a very small number of devices, at the expense of the larger desktop community, in the quest to have some control over those 'billions' of handheld devices [11:04] TJ-: but isnt their aim the desktop also with unity8? [11:04] And they're sacrificed the power desktop user's abilities to control their desktop, and dumbed things down in the name of 'convergence' [11:04] yeah i understand [11:05] lotuspsychje: it is, but the *design* is driven by small (screen) devices [11:05] lotuspsychje: this is the reverse shell bash -i > /dev/tcp/127.0.0.1/8080 0<&1 2>&1 [11:05] EriC^^: cool [11:06] lotuspsychje: on your pc you type netcat -l -p 8080 [11:06] and as I said earlier, this weeks' UDS schedule which I went looking at for things to contribute to is overwhelmingly (I'd guess 90%) about convergence, snap, and 'apps' [11:06] and on the customer's machine you type bash -i > /dev/tcp//8080 0<&1 2>&1 [11:06] EriC^^: and where does the lang=c come in? [11:06] you get a shell prompt on the customer's [11:06] after the shell type export LANG=C [11:07] then install stuff as usual [11:07] EriC^^: cool [11:08] EriC^^: first ill try to do most work myself [11:08] EriC^^: but on crowdy days i might need some remote help [11:08] yeah maybe maintenance stuff [11:09] would come in handy [11:09] yeah [11:09] EriC^^: well the way i want it, is to deliver tweaked ubuntu boxes to the customer with all the handiest packages preinstalled [11:10] EriC^^: so after i plugged ssd in and installed, packages,updates,tweaks needed [11:10] lotuspsychje, OEM installation --- [11:11] cfhowlett: gonna try to avoid that, as packages changes [11:11] OK [11:11] i want to be able to adapt at changes [11:11] new drivers, new hardware etc [11:11] seeded auto-install using PXE boot [11:12] ill do the physical vanilla install, and plugin cable [11:12] configure your seed file and let it all be done automatically [11:13] TJ-: i installed xenial in 10min on the 850 pro [11:13] most work is the tweaking and packages [11:14] and every customer needs another layout [11:14] we dont need gimp or krita for a gamer machine [11:14] or no steam on granma's box :p [11:15] i want to deliver tweaked boxes with specific needs [11:15] with auto security updates n such [11:15] lotuspsychje: right, seed files are the way to automate it. Have a core file with the common requirements, plus specialised seeds on top for each purpose. [11:16] hi BluesKaj [11:16] Hey folks [11:17] Hi lotuspsychje [11:20] for those who havent readed it yet: http://www.omgubuntu.co.uk/2016/05/ubuntu-image-size-increase-2gb [11:21] does anybody know a quick fix for the 16.04 suspend problem? [11:21] which suspend problem? [11:22] after suspending i get a black screen and it's unresponsive [11:22] even alt+sysrq doesn't work [11:22] EFI? [11:22] yes [11:22] does the PC have an EFI GOP module? [11:23] gop? [11:23] (in the firmware) ... dmesg should indicate if there is one the kernel can talk to [11:23] dmesg | grep -i gop shows nothing [11:24] Graphic Output Protocol - an UEFI driver to do modesetting etc. [11:25] doesn't look like it no idea [11:25] what's that fw.efi file in 16.04 btw? [11:25] fwupx64.efi [11:26] they also made /boot/efi read only for root now [11:26] umask=0077 in my fstab [11:26] in dmesg you'll have something like "fb0: EFI VGA frame buffer device" [11:27] i added that for a new user while giving him a sample line for fstab and grub complained, he wasn't on 16.04 i guess [11:27] that's the firmware update executable; with capsule files linux can now do firmware updates natively [11:27] nice [11:28] no efi stuff just one [ 2.159749] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device [11:29] that was something we worked in last UDS for 15.10, in collab with the redhat devs [11:29] the umask=0077 ? [11:29] grub.cfg isn't read for all anymore too [11:29] Zesty Zany Zebra [11:29] so when the PC resumes from S3 it calls ACPI methods some of which should re-enable the GPU; it's possible simply putting in a correct acpi_osi= would fix it [11:30] TJ-: why does the alt+sysrq not work though? curious [11:30] isn't that a kernel thing? [11:30] this pc had a weirdness where when charging at power-up the USB touchscreen digitiser wasn't powered up! [11:31] EriC^^: yes, it's possible the kernel panics, or ACPI resume methods get stuck [11:31] ok [11:31] using acpi_osi fixed it [11:31] cfhowlett: lol [11:31] I'm seeing that most PCs now *need* acpi_osi to function optimally, else the ACPI DSDT doesn't enable all power functionality [11:32] btw yesterday i was watching a movie with the hdmi plugged in, and contrary to usual the player started lagging a lot and the whole pc lagged and i found a bunch of irq1 irq2 etc in top taking up the cpu [11:32] wonder if that's related? [11:32] it used to work flawlessly with the hdmi [11:33] could be yes. check what OSIs the ACPI DSDT supports and add the latest to the kernel command-line [11:33] what was the command for that if you dont mind? [11:34] "sudo strings /sys/firmware/acpi/tables/DSDT | grep -i windows" and identify the lastest version. usually it'll be something like "Windows 2013" or similar [11:34] thank you [11:35] then add to the kernel command line: "apci_osi=! acpi_osi="Windows 2013" " [11:35] looks like Windows 2012 [11:35] the ! deletes all existing OSIs then you add in just the one [11:35] oh ok [11:35] trying [11:36] brb [11:37] cfhowlett: zesty zebra will have unity8 as default also :o [11:38] TJ-: i wonder where this all will take us [11:38] yes. that will make it ... zestier? [11:38] lol [11:38] for the non-lovers there will be still our flavors right [11:39] cfhowlett: OerHeks is seeding iso flavors, and told us studio is most downloaded of all? [11:39] Oh, wait! I was wrong. [11:39] Zaftig Zebra! [11:39] lol [11:40] !!?! could be. [11:40] I haven't seen stats [11:40] !codenames [11:40] Ubuntu has awesome release codenames. See http://wiki.ubuntu.com/DevelopmentCodeNames for more [11:41] cfhowlett: http://i.imgur.com/7lnhB2B.png seeded by $ OerHeks $ [11:42] nice [11:43] TJ-: no luck [11:43] i was helping a chap the other night with a dell inspiron with a laggy touchpad, he was about to RMA it to Dell. We added the latest acpi_osi and it fixed that issue and doubled battery life and generally made the PC feel 'zippier' [11:43] i noticed it didn't even suspend, it just turned the screen black, the fan started turning faster, then hung, alt+sysrq didnt work [11:44] EriC^^: OK, so then it may be a failure to suspend at all [11:44] nice [11:44] oh, it's nothing acpi related? [11:44] i tried acpi_osi="Linux" on this boot [11:44] btw how can we see the older dmesg now? [11:44] "Linux" won't work [11:44] /var/log/dmesg.0 is empty [11:44] kern.log kern.log.1 etc [11:45] even /var/log/dmesg is [11:45] ah thanks [11:45] dmesg has always been directed to kern.log in Debian/Ubuntu [11:45] it had a dmesg too though i think [11:45] EriC^^: the only OSI strings that will work are those present in the ACPI DSDT, revealed by that 'strings' command [11:46] Yes, usually the boot-time dmesg was saved to a file [11:46] ok i uploaded kern.log.1 , let us see [11:46] http://paste.ubuntu.com/16200730/ [11:46] it's huge though [11:47] oh crap [11:47] EriC^^: to debug the suspend issue, reboot with "no_console_suspend initcall_debug debug" on the kernel command line... then *use a console* to initiate the suspend "sudo systemctl suspend& dmesg --follow" and you *may* see some useful info [11:47] it's old, just til april 30 [11:48] notice how the systemctl command is put into the background so you can watch the dmesg output (--follow is like tail -f ) [11:48] ok [11:50] ok brb [11:53] all these same returning issues on linux is a pain [11:53] the symptom may be similar but the causes often are different [11:53] yeah but same suspend problems are suffered by so many users over and over [11:54] newer hardware brings fresh challenges for newer driver code and so on [11:54] true [11:55] but on windows working by default? [11:55] well, much of it is down to Windows begin the target of the manufacturers, and they fix bugs on there with their own drivers that do workarounds, but don't publish that info so Linux devs can do the same workarounds, so we end up having to reverse-engineer the problems to understand them, to create solutions [11:55] right [11:55] i hope i got it right [11:55] yeah [11:55] initcall_debug debug [11:55] you put a vanilla windows install on many PCs they'll have many devices not working... you've got to pile on the manufacturer's drivers for all the devices too [11:56] TJ-: also true [11:56] EriC^^: the most important is "no_console_suspend" so you can continue seeing messages right up to the power going off [11:56] bah! he missed that! [11:57] fast as lucky luke [11:57] we have a saying... up and down more than a whore's drawers :) [11:57] lol [11:58] I had a weird suspend/resume issue a few days ago that seems to have gone away. the pc was waking up every 30 minutes then going back to sleep overnight [11:59] last 3 nights its been OK... maybe it was just dying for the toilet :D [11:59] EriC^^: the most important is "no_console_suspend" so you can continue seeing messages right up to the power going off [11:59] TJ-: it looked normal [11:59] it put a bunch of text really quickly [11:59] then reached systemd streamed .... something a bunch of messages [12:00] it paused there for a bit then the screen turned off and the fan started spinning quicker [12:00] EriC^^: I usually video it and play it back slowly so I can analyse the messages properly [12:00] it was intensely quick [12:00] the fan speeding up does sound like an ACPI event [12:00] can a tail -f do any help? [12:00] Yes, it should be [12:00] lotuspsychje: "dmesg --follow" [12:01] i tried acpi=off now [12:01] let me give it a shot [12:01] cross fingers [12:01] :D [12:02] interesting systemctl suspend isn't working from the gui [12:03] acpi=off is a good way to fry the system these days [12:03] lol i just noticed [12:03] temp at 88 [12:03] brb [12:03] somthing smells here.. [12:04] oh its me! i need a shower :p [12:04] could be me, had been burritos last night... [12:04] bean even [12:04] haha [12:05] didn't realize ubuntu had implemented 'smellivision' [12:05] In the interests of 'convergence' Canonical is introducing the unity-scope-smellovision in the 16.10 cycle [12:05] lol [12:05] wahaha [12:05] great minds :p [12:05] phew [12:06] EriC^^: did you put it in the freezer? :D [12:06] ok back with no_console_suspend [12:06] lol, no i only do that to my iphone [12:06] :D [12:06] EriC^^: does the PC have a serial port? [12:06] (no joke i had to fix the wifi that way) [12:07] this is going to sound dumb but what does it look like [12:07] serial is the vga ? [12:07] or the long printer one you mean? [12:08] I used to set the kernel to using a serial port to output its messages so I could capture them on another PC. netconsole might work but I'd guess the network will be down far too early to help. Since serial ports went away I now have a custom-made USB>serial<>serialserial converter chips back-to-back so I can do the same thing over USB [12:08] oh [12:09] i'd use the iphone to ssh to it [12:09] RS232 serial interface is a 9-pin D-type connector with 2 rows of pins. VGA is a 15-pin D-type with 3 rows of sockets [12:09] but it's making a disconnect sound after the command [12:09] ah [12:09] i dont think it has a serial port [12:09] yes, you need to direct connection to capture the suspend messages if you can't video-capture the screen messages [12:09] i will try to video record them [12:10] brb [12:10] EriC^^: you can force the kernel to wait after each message to stop the speed scroll too, hang on! [12:10] wait! [12:10] speedy bloosy gonzales! [12:11] lol [12:11] how? [12:11] I'm not sure if this will take effect during suspend but... [12:11] each message will be too much i think though [12:12] ... "boot_delay=" will put a delay after each kernel message at boot-time. It may also work at suspend/resume, but I've never tested it [12:12] ok i'll try it [12:12] let me test now as it is booted [12:12] e.g. "boot_delay=100" is 1/10 second [12:12] ok [12:12] surely needed on ssd's [12:12] :D [12:13] brb [12:13] * lotuspsychje off to shower now for real [12:13] * TJ- thinks Eric^^ needs a 2nd PC [12:14] TJ-: not everyone has 6 matrix-style screens in fron of them lol [12:15] mr. anderson, working in IT company overdays, hacker overnight :p [12:16] I don't right now whilst working on this T300; it's hard work having to flip between windows that obscure each other... a very primitive way of working! [12:19] it wouldn't suspend, just gave some messages then said watchdog at the end [12:19] when i press enter it just gives the shell back [12:20] wb cfhowlett [12:20] OK, that's useful [12:20] it is? [12:21] of course... can you show us the messages? [12:21] lol i deleted the video [12:21] presumably some device failed to suspend correctly so the kernel aborted the operation [12:22] pm-suspend did work btw in case it helps [12:22] i'll reboot with the cmdline parameters and video again [12:22] any final thoughts? [12:23] ill brb [12:41] TJ-: this is the last message it displays http://imgur.com/dfPgR9R [12:43] there's no suspend related messages there at all [12:43] the very last thing is displays at the bottom is watchdog=1 [12:44] i'm uploading the video of the dmesg output, half way there === JanC is now known as Guest52941 === JanC_ is now known as JanC [12:45] there's a message in the scroll that says can't disable aspm, os doesn't have aspm control [12:46] CRAT: table not found [12:46] btw boot_delay didn't boot the pc [12:46] i think it's seconds instead of milliseconds [12:47] or for the whole time it waits that much [12:47] i hope it didn't butcher the quality too much [12:47] https://sendvid.com/ofy9uw04 [12:48] hmm it's not downloadable i think :/ [12:49] let me paste from my working suspend/resume kernel log [12:52] hmm i think none of the scroll is related to the suspend [12:52] cause the time is 44 then 87 [12:52] http://paste.ubuntu.com/16201163/ [12:53] the point at which it had just started the Resume is where you find "ACPI: Low-level resume complete" [12:54] the timestamps of the messages before that, when the system was suspending, are the time the system resumed. The reason is they stayed in the kernel's log buffer and were only flushed to log-file when the resume had happened, and they get timestamped when added to the log, not when generated [13:04] EriC^^: another option would be booting newer kernel [13:05] 4.4.8 or so [13:07] seems like they made a page for xenial too: https://help.ubuntu.com/16.04/ubuntu-help/power-suspendfail.html [13:07] TJ-: they mention indeed might be the hardware not supported [13:08] lotuspsychje: good idea [13:08] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1566302 [13:08] Ubuntu bug 1566302 in linux (Ubuntu) "Ubuntu 16.04: Suspend freezes the system after upgrade to linux image 4.4.0-16" [Medium,Confirmed] [13:09] this guy fiddles with kernels also [13:10] !mainline [13:10] The kernel team supply continuous mainline kernel builds which can be useful for tracking down issues or testing recent changes in the Linux kernel. More information is available at https://wiki.ubuntu.com/Kernel/MainlineBuilds [13:12] also interesting https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1574120 [13:12] Ubuntu bug 1574120 in systemd (Ubuntu) "Notebook doesn't suspend when lid is closed after update to 16.04" [Medium,Incomplete] [13:13] seems easy enough to install a new kernel [13:13] just d/l then dpkg -i [13:14] yep yep :p [13:14] hi pauljw :) [13:14] im curious [13:14] hi EriC^^ [13:14] EriC^^: im reading 4.6 or 4.8 [13:14] how's it going? [13:14] I wonder if those systems all have hybrid GPUs? [13:14] going well, EriC^^ tks. you? [13:14] lotuspsychje: i'm trying the 4.6rc2 [13:14] EriC^^: ok! [13:14] 4.6rc6* [13:15] pauljw: good, thanks [13:16] TJ-: i talked to the @ from gamingonlinux this morning saying that all our GTX 9xx series can be solved with: disable intel cpu graphics, even when not optimus [13:16] TJ-: grabbing latest nvidia driver from ppa [13:17] or disabling fastboot [13:17] for users who cant get in to installing ubuntu/black screens [13:17] headers package has a problem, linux-headers-4.6.0-040600rc6-generic depends on linux-headers-4.6.0-040600rc6 [13:18] you need the linux-headerse*all.deb as well as the architecture-specfic debs [13:18] do i need them just to test though? [13:18] it's just for dkms virtualbox right? [13:18] aha [13:19] thank god for apt, makes life real easy [13:20] can't imagine what it'd be like without it [13:20] :p [13:21] the _all.deb worked [13:21] ok, testing :D [13:21] brb [13:24] works! [13:24] !yay [13:24] Glad you made it! :-) [13:25] i guess it's a kernel bug then? [13:25] EriC^^: perhaps you should recreate a bug and solve it right away? [13:25] how? [13:25] !bug [13:25] If you find a bug in Ubuntu or any of its derivatives, please file a bug using the command « ubuntu-bug » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs. [13:26] TJ-: is the new dmesg useful? [13:26] and then attach yourself how you fixxed it [13:28]