IRC Logs for #crux Wednesday, 2017-10-11

john_cephalopodaIt could also be done manually. Actually, I think that's how 6c37 does it (manually).00:00
john_cephalopodaNight00:13
*** DaViruz has quit IRC02:33
*** DaViruz has joined #crux02:33
*** jaeger has quit IRC02:43
*** jaeger has joined #crux02:45
*** _07IABE62D has joined #crux02:52
*** SiFuh_ has quit IRC04:23
*** SiFuh has joined #crux04:24
*** SiFuh has quit IRC05:50
*** SiFuh has joined #crux05:51
*** Workster has quit IRC05:52
*** vqmyesvurnqnlxua has joined #crux05:52
*** Workster has joined #crux05:52
*** parlos has joined #crux06:21
*** SiFuh has quit IRC06:30
*** SiFuh has joined #crux06:31
prologicjaeger: thanks!06:54
prologicthat's (I believe) why I was confused on all the SPF suggestions06:54
prologicit should have on effect on inbound mail06:55
*** timcowchip has joined #crux06:56
*** JanC_ has joined #crux07:02
*** parlos has quit IRC07:32
*** igbmkenhvsmyxaxk has joined #crux08:03
*** Workster has quit IRC08:03
*** Workster has joined #crux08:03
*** parlos has joined #crux08:19
timcowchipafter updateing qt5, qtwebengine and qtwebkit along with all my lxqt ports, lxqt-panel is broken08:26
timcowchipthis kind isn't worth to me anymore08:26
timcowchipI can spend the time to fix08:26
timcowchipit08:26
timcowchipbut, for how long...until some other updated dependency breaks something else08:27
timcowchipI don't even use crux and as far as I know I'm maintaing unused ports for lxqt08:28
timcowchipI'm getting a life now.....bye08:29
joacimi only really maintain ports that i use myself08:30
timcowchipI hear that08:31
joacimto me it seems like a better way to keep yourself motivated to maintain your ports08:31
timcowchipmaybe I'll go back to dual-booting someday08:32
timcowchipbut not this day08:32
timcowchipbye for now :(08:32
*** timcowchip has quit IRC08:34
*** leetspete1 has quit IRC08:36
*** leetspete1 has joined #crux08:41
*** abenz_ has joined #crux08:57
*** john_cephalopoda has joined #crux09:42
john_cephalopodaHi09:42
abenz_Hi john_cephalopoda10:30
pedjawell, 'Internal authentication error' is as generic error code as it gets10:54
pedja'may be caused by multiple environmental and configuration issue'. Gee, Oracle, thanks for clarifying that10:55
*** john_cephalopoda has quit IRC11:09
*** john_cephalopoda has joined #crux11:24
*** onodera has joined #crux11:26
*** Lukc has quit IRC11:27
*** Lukc has joined #crux11:27
*** s-mutin has quit IRC11:28
*** s-mutin has joined #crux11:29
*** parlos has quit IRC12:34
pekka_10well, Im also compiling the same apps right now12:49
pekka_10Qt5, Qtwebengine, ffmpg, etc12:49
pekka_10and I'm enjoying the thinkering that goes with it12:50
pedjadebugging why the build fails just sometimes is fun (for weird values of 'fun')12:59
pedjaon the bright side, it's been a while since I came across a heisenbug13:01
pekka_10well, its just as the petrol-heads, they like he smell of gasoline, and enjoy tweaking whats under the hood13:01
*** Workster has quit IRC13:01
pekka_10heisenbug?13:01
pekka_10yea13:02
pedjaa bug that seems to appear only when I am looking :)13:03
pekka_10I installed IceWM just to get rid of the many dependices13:04
pekka_10the only thing I need is a decent browser, and QT5 seems to be the way to go13:04
pedjawhat's wrong with FF?13:05
pedjait's a decent browser13:05
pekka_10well, I just cant find an easy way to force FF to use a tmpfs for caching13:06
pekka_10until recently /home has been on a SSD, and I moved it to a mirrored raid13:07
pekka_10and now it sounds like popcorn every time I open FF13:07
pekka_10every page, and popcorn is bouncing out of my PC13:08
pekka_10this was actually a nobrainer with Qupzilla13:08
pekka_10and, since Im using esoteric HW like ARM socs, there is only a 16 GB emmc13:09
pekka_10so this uncontrolled caching to emmc will shorten the time of this gadget13:10
pekka_10and13:10
pekka_10Qupzilla is Wayland ready, and so is Qt513:11
pekka_10and, I will stay away from GTK3 applications13:13
pedjawouldn't setting 'browser.cache.disk.enable' to false do that?13:14
pekka_10so, well, apart form the hassle with configuring FF, nothing is wrong with it13:14
pekka_10I've just made some decisions, and I'm intended to stay on the lines13:14
pekka_10I've tried it, but still my HD's are popping13:15
*** druid_droid has quit IRC13:17
*** druid_droid has joined #crux13:19
pedjaspeaking of wayland, this is interesting https://drewdevault.com/2017/10/09/Future-of-sway.html13:25
pekka_10yea, I've been looking at Sway13:26
pedjawlroots is the interesting part :)13:27
pekka_10for me, it was an revilation when installing Ice, it didnt pull in any GTK or QT dependices13:27
pedjaand what wayland developers should've done13:28
pekka_10and I hope Sway behave the same in Wayland13:28
jaegerprologic: yeah, it's just a sort of "good citizen" thing to setup for others. On the other side of the coin, setting up something like postfix-policyd-spf on your server could reduce incoming spam a bit13:44
*** parlos has joined #crux13:50
frinnstjaeger: saw this and thought of you: https://www.reddit.com/r/sysadmin/comments/75fnmv/15_hour_vmware_call_and_5_different_vmware_techs/14:10
frinnstreminds me of microsoft. Seems they can't figure out what our networking issue is14:14
*** blueness has quit IRC14:51
*** blueness has joined #crux14:51
jaegersheesh15:21
jaegerI'm glad to say I've never had an experience that bad with vmware15:21
frinnstOur hyperv plans pretty much died today15:21
frinnstso its vsphere 6.5 instead15:22
frinnstmicrosoft thought our "workaround" was a solution - a static arp entry for EVERYTHING15:22
frinnstits pretty much our only micorsoft support ticket we ever had - and that set the tone. so no thanks15:22
jaegerew, that sucks15:24
ryu0frinnst: wait, they screwed up something as basic as ARP?15:39
ryu0o.o15:40
joacimim told fairly regularly that microsoft won, that ill see nothing but hyperv15:41
joacimmy teachers might be ms fanboys tho15:41
ryu0hyperv? that hypervisor?15:42
joacimyes15:42
ryu0does hyperv even work with KVM?15:42
joacimsame about azure15:42
ryu0haha. good one. most cloud providers aren't running Windows hosts.15:43
ryu0everywhere i go i see some Linux specific technology at work it seems like.15:43
ryu0OpenVZ, LXD, KVM, ...15:44
joacimabout kvm. think i figured out my performance issues with gpu passthrough15:44
ryu0containers are pretty nice. they don't require you to permanently allocate memory to them.15:45
joacimsomething about npt. performance graphics performance should improve by disabling npt. but that would bring its own issues i guess15:45
frinnstthe biggest problem we face wrt. kvm is the management. unless you have a good devel team you are screwed15:49
joacimi guess libvirt isnt that great for such things15:49
joacimwhen compared against hyperv and vmware15:49
frinnstno you cant compare those15:50
frinnsthttps://gfycat.com/EachImpressiveDrongo i laughed out loud15:58
SiFuhfrinnst: that is pretty cool. I could never sleep that deep. If a mosquito landed on my arm, I would wake up.16:01
SiFuhhttps://gfycat.com/gifs/detail/EnviousCarefulAmericanblackvulture16:05
SiFuhThat's real fisshing16:05
frinnstbrutal16:11
pedjaiirc, there are 3 major players as management frontends for kvm, each with its own unique set of issues :)16:44
pedjaoVirt (upstream for RedHat EV), Proxmox and OpenStack.16:46
pedjaa friend of mine deployed HPE Helion OpenStack at work16:58
pedjayear ago, or something like that16:59
*** tsaop has joined #crux16:59
pedjahe had a few more grey hairs last time I saw him, but he said that's mostly from dealing with local HPE support17:00
tsaopHello17:08
tsaopIt seems runc must be updated following the docker update17:08
tsaopI get the following error message when I try to start a container: json: cannot unmarshal object into Go struct field Process.capabilities of type []string17:09
*** parlos has quit IRC17:36
pedjatsaop, paste the output of 'docker info'17:37
pedjadocker will complain if there is a difference between runc/containerd version found and expected17:41
tsaoppedja: http://paste.lisp.org/display/35832817:41
pedjarunc/containerd are the wrong version17:42
tsaopmmh17:43
tsaopI am pretty sure I have the latest version17:43
tsaoplet's force-update17:43
pedjarebuild them, and restart docker17:43
tsaopuhh, it seems you're right17:44
tsaophow come din't prt-get catch the update?17:44
tsaopmaybe because of the version?17:44
pedjayeah, commits are weird like that17:44
tsaopit works now17:44
tsaopthanks17:44
pedjanp17:45
tsaopnow I only have to build qt5 and qtwebengine17:45
pedjagood luck17:46
j_vi wonder if prt-get not catching that update is related to the 'preferhigher' prt-get.conf option?17:46
tsaopj_v: I have set it to yes17:47
tsaopmaybe that's the cause?17:47
j_vseems possible, since those ports are using commit versions17:47
j_vas port versions17:48
j_va newer git commit can easily sort lower than an older one17:50
pedjathat's why I bump version for them, so they get picked up17:55
pedjas/version/release/17:56
*** p4cman has joined #crux18:34
*** fnords has quit IRC18:38
*** fnords has joined #crux18:38
cruxbot[xorg.git/3.3]: [notify] xorg-libxfont2: updated to 2.0.2. Fix for CVE-2017-13720, CVE-2017-1372218:49
cruxbot[xorg.git/3.3]: xorg-libxres; updated to 1.2.018:49
jaegerI recommend NOT using "preferhigher" in prt-get.conf for this very reason18:49
jaegerbut I could also consider using version names like YYYYMMDD-<githash> or something if that's better18:50
tsaopjaeger: nice to know18:50
*** tsaop has quit IRC21:18
*** tsaop has joined #crux21:29
pedja'bridge' from iproute2 shows just one bridge, 'brctl' from bridge-utils shows all 3. hm21:43
*** tsaop has quit IRC21:46
*** onodera has quit IRC22:01
*** john_cephalopoda has quit IRC23:13
*** Workster has joined #crux23:15
*** p4cman has quit IRC23:33

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!