[06:37] <tjaalton> apw: drm-intel-nightly builds on amd64 seem to be failing
[06:38] <tjaalton> Cannot use CONFIG_CC_STACKPROTECTOR_STRONG: -fstack-protector-strong available but compiler is broken
[12:50] <WeiJunLi> i have a vm with 10.5gb available is it enough space to build 4.8.0 kernel?
[15:24] <neoark> anyone have a fix for this issue? https://askubuntu.com/questions/844583/state-check-failed-when-running-canonical-livepatch#new-answer
[15:25] <neoark> It seem like I am getting lot of 403 errors
[15:38] <cyphermox> bjf: on the system where you get the grub upgrade errors, could you please run /usr/lib/grub/powerpc-ieee1275/prep-bootdev ?
[15:39] <cyphermox> I've been working on a fix for the issue how I think I managed to reproduce it here in a VM, but it would help to make sure I'm working on the right issue :)
[15:39] <bjf> jsalisbury, ^
[15:40] <WeiJunLi> the maximium of mem for vmare fusion is around 8gb, everytime i create a new vm it only has 10gb available and I want to compile 4.8.0 kernel? Any idea to have big space on a vm, my host has around 60gb free
[15:42] <bjf> PHLin, ^
[15:42] <bjf> jsalisbury, sorry, i want PHLin to look at that problem with ppc64el and not you
[15:42] <PHLin> bjf, ack
[15:43]  * PHLin redeploy the system
[15:43] <cyphermox> I expect it returns something like /dev/sda or something, and I'll hand you some more code to run in a bit
[15:44] <bjf> PHLin, be careful that if kernels land in -proposed the automatic testing is going to grab that system
[15:45] <cyphermox> PHLin: and this is my proposed fix: http://paste.ubuntu.com/23480943/, should return some /dev/mapper/mpathsomething
[15:46] <PHLin> bjf, ok, maas is now doing its job
[15:46] <PHLin> cyphermox, got it
[15:48] <cyphermox> PHLin: if you want to easily compare the "old" behavior and the new one, you'll want to drop "&& !multipath_path" from line 34 in the file (line 38 in the pastebin), and run it as prep-bootdev -l
[15:48] <cyphermox> ie. that would likely list /dev/sda1 (or something equivalent), being the real disk, and /dev/mapper/mpath0-part1 (the multipath layer)
[15:49] <PHLin> okok
[15:49] <bjf> cyphermox, doesn't foundations have a ppc64el system for testing?
[15:49] <cyphermox> if I'm looking at the right thing, the issue is that prep-bootdev, when run by the grub installer, lists disks in order but it tends to always get them with the "real" disks first, whereas they're claimed by multipath
[15:49] <cyphermox> bjf: I'm working on a ppc64el in VMs
[15:50] <cyphermox> it's usually good enough, and easier to redeploy
[15:50] <cyphermox> (and I can keep more than one deployed to compare things)
[15:50] <bjf> cyphermox, right and that's not good enough, hence asking us to test (which we are happy to do)
[15:51] <cyphermox> bjf: only to make sure I really was looking at the right thing
[15:52] <cyphermox> there's clearly a bug in the way grub-install gets run for multipath on ppc64el, and it's the most likely issue on your system, but it helps to make sure.
[15:52] <jsalisbury> bjf, ack
[15:58] <PHLin> cyphermox, http://pastebin.ubuntu.com/23480984/ first result on 3.13.0-101
[16:01] <PHLin> cyphermox, result with your code: http://pastebin.ubuntu.com/23481000/
[16:05] <PHLin> cyphermox, looks like it's working as expected, the /dev/mapper/mpath0p1
[16:07] <cyphermox> yeah, that looks about right
[16:07] <cyphermox> PHLin: so I'll get you next a new grub package with that change in, so we'll be able to make sure the upgrade goes clean
[16:08] <PHLin> cyphermox, ok cool
[16:26] <WeiJunLi> the maximium of mem for vmware fusion is around 8gb, everytime i create a new vm it only has 10gb available and I want to compile 4.8.0 kernel (want to enable some more debug features). Any idea to have big space on a vm, my host has around 60gb free
[16:33] <PHLin> cyphermox, got to go, just send me a mail and I can work on this tomorrow