| === gerald is now known as Guest78324 | ||
| sbeattie | bjf: Hi. I just looked at the apparmor failure log in http://kernel.ubuntu.com/testing/onibi__3.13.0-53.88__2015-05-18_17-33-00/console_output.txt.html ; it looks like it pulled in the apparmor packages in trusty-proposed, which means you shouldn't see the apparmor apport hook failure anymore (yay!), but it it looks like the trusty-proposed deb-src line wasn't enabled in apt's sources.list. | 06:07 |
|---|---|---|
| sbeattie | bjf: that may be the cause of the different failures seen this time, as part of the apparmor qrt test script downloads the source package and runs make check in it. | 06:09 |
| bjf | sbeattie, ack, i'll look at it tomorrow. thanks | 06:11 |
| sbeattie | bjf: no worries, thanks. | 06:14 |
| sbeattie | ah, yep, I've reproduced the the failures with the sources from apparmor 2.8.95~2430-0ubuntu5.1 while binary packages from 2.8.95~2430-0ubuntu5.2 are installed. | 06:22 |
| FourDollars | sforshee: Hi, did you remember the patch you made for Dell New XPS 13? https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/hid/hid-multitouch.c?id=2c6e0277e1eab3df5db81c59e408b7b1c14b1b72 | 06:41 |
| FourDollars | sforshee: Does it fix any problem? | 06:41 |
| === gerald is now known as Guest33142 | ||
| pkern | The new 3.13 contains "tcp: make connect() mem charging friendly" but doesn't seem to contain "tcp: Fix crash in TCP Fast Open", is that intentional? | 09:16 |
| pkern | I didn't check if it crashes in the same way as 3.16 did, but it's somewhat concerning that the fixup wasn't picked up if it's needed. |: | 09:17 |
| pkern | Yeah, it's dead on arrival for us. | 09:21 |
| apw | pkern, new 3.13 stable or new 3.13 in -updates in trusty ? | 09:22 |
| pkern | 3.13 in -updates in trusty. | 09:22 |
| apw | henrix, ^ | 09:22 |
| pkern | Of course it relies on things to use fast open. Which probably just Google does? I don't know. | 09:22 |
| apw | yes, you are likely the only people in the world advertising that feature sadly | 09:23 |
| pkern | And using it in clients, I guess. | 09:23 |
| * henrix goes have a look | 09:23 | |
| apw | that patch shows applied across the board (as it was a cve) but it was behind the other releases in trusty | 09:24 |
| apw | but i had assumed that meant the introducing commit was also not released either ... hmmm | 09:24 |
| henrix | ugh | 09:25 |
| henrix | yeah, that's correct | 09:25 |
| henrix | pkern: thanks for spotting that! i'll go work on fixing that ;) | 09:28 |
| infinity | henrix: Hrm. Should we do a single-commit respin to pick that up? | 09:28 |
| pkern | henrix: ta! | 09:28 |
| infinity | According to Ben's commit message, it should only affect 3.13 and 3.16. | 09:29 |
| infinity | And it was fixed in 3.16 already, so just 3.13. | 09:29 |
| apw | yeah somehow we paired it up in 3.16 and not in 3.13 | 09:30 |
| henrix | infinity: yeah, i believe a quick respin will be needed. i go finish the V respin and start with fixing T | 09:30 |
| infinity | henrix: +1 from me for a single-commit resping of T and lts-T. | 09:30 |
| infinity | respin, too. | 09:30 |
| infinity | Typing hard. | 09:30 |
| infinity | pkern: Since you seem to be a pro at reproduction, can I get you to test henrix's kernel when he does the deed, so we can turn it around ASAP? | 09:31 |
| pkern | infinity: Sure. | 09:32 |
| genkgo | infinity: last week you helped me out with getting cloud tools for kernel 3.19. we switched to this kernel for our problems with creating vss snapshots on our hyperv machine. | 09:36 |
| genkgo | infinity: unfortunately, it did not solve the issue. our vps goes read-only sometimes while creating these vss snapshots. now we have switched to a new kernel, the error is different than before. | 09:37 |
| infinity | genkgo: You want apw, not me. | 09:37 |
| genkgo | blk_update_request i/o error, dev sa, sector xxx | 09:37 |
| infinity | genkgo: I was just helping with the packaging mysteries, not the hyperv madness. | 09:37 |
| genkgo | infinity: ok :) | 09:37 |
| genkgo | apw: i still have some hyperv madness :) | 09:38 |
| apw | genkgo, ok add that info to the bug, and the new dmesg, and remind me of the bug number, and i'll see if we can get someone to reproduce it | 09:38 |
| pkern | henrix: Just drop me a mail at $nick@google.com if you have something to test for me. I take it that a respin would likely be released earlier than next Sunday? :) | 09:39 |
| henrix | pkern: yeah, for sure before Sunday :) | 09:40 |
| henrix | pkern: i hope to have it uploaded and building later today, i'll email you once we have a test kernel | 09:40 |
| pkern | henrix: Thank you :) | 09:41 |
| genkgo | apw: i did not report a bug before, because we thought it was related to the kernel version. if i run "ubuntu-bug linux" i get "This is not an official Ubuntu package. Please remove any third party package and try again." (*** Problem in linux-image-3.19.0-17-generic) | 09:41 |
| infinity | pkern: If all goes well, I'd expect to release it in < 24h. | 09:43 |
| === dupondje_ is now known as dupondje | ||
| genkgo | apw: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1456985 | 10:23 |
| ubot5 | Launchpad bug 1456985 in linux (Ubuntu) "Filesystem goes read-only while creating VSS snapshot on HyperV" [Undecided,New] | 10:23 |
| genkgo | apw: While browsing through other bugs related to Hyper-V in the Ubuntu kernel, I found this one https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1445195. These patches will solve my problem. The reported problem here https://social.technet.microsoft.com/Forums/windowsserver/en-US/8807f61c-565e-45bc-abc4-af09abf59de2/ubuntu-14042-lts-generation-2-scsi-errors-on-vss-based-backups is an exact copy of the problem we are having. | 10:34 |
| ubot5 | Launchpad bug 1445195 in linux (Ubuntu Wily) "[Hyper-V] Kernel patches for storvsc" [High,Triaged] | 10:34 |
| infinity | genkgo: Ahh, lovely, thanks. | 10:52 |
| infinity | apw: Bonus points if you can get those on to all the master-nexts before the next SRU cycle happens. :) | 10:53 |
| genkgo | infinity: for me this also great news. i had the feeling i was on my own with this problem. now fortunately i see there are many others that have the problem. and there is already a patch it. | 10:54 |
| infinity | genkgo: Always nice to see that it's ultimately not our fault, too. :P | 10:54 |
| genkgo | hehe, i totally understand that! | 10:55 |
| apw | sigh, i believe we are waiting on some testing for those from microsoft, will chace that up | 11:19 |
| genkgo | apw: if there is any testing we can do: please let me know. i'd love to help. | 11:30 |
| apw | genkgo, yep, will make sure you get some test kerenls | 11:31 |
| genkgo | apw: wonderful, thanks for the good work | 11:31 |
| jsalisbury | genkgo, Should have test kernels for each release in about 15 to 20 minutes | 12:08 |
| genkgo | jsalisbury: wonderful! | 12:08 |
| genkgo | jsalisbury: should build a kernel myself then or is it apt-get install? i am not extremely familiar with building a kernel.. :) | 12:09 |
| jsalisbury | genkgo, I'll have the kernels as a .deb file, then you can dpkg -i them | 12:10 |
| genkgo | jsalisbury: aight, that would be no problem :) | 12:10 |
| jsalisbury | genkgo, there is already a vivid kernel here: http://kernel.ubuntu.com/~jsalisbury/lp1445195/vivid/ Building the other releases now | 12:11 |
| genkgo | jsalisbury: we are on trusty with vivid kernel, so i can use those already | 12:12 |
| genkgo | jsalisbury: i am still working on something else now, but will make sure to install today. last time it took 36 hours before we had a machine crashed on backup, and then we were creating backups every hour. | 12:13 |
| jsalisbury | genkgo, great. You just need to install the linux-image and linux-image-extra .deb packages. | 12:13 |
| jsalisbury | genkgo, ok, thanks | 12:14 |
| genkgo | jsalisbury: great, thanks for the work | 12:14 |
| jsalisbury | genkgo, np | 12:14 |
| jsalisbury | genkgo, thanks for offering to test! | 12:15 |
| genkgo | jsalisbury: sure, we would like to get rid of the problem asap. i am going to run the tests on a production machine. on machines with not enough activity (test machine) we were not able to reproduce the bug. or we did not wait long enough (4 days, backup every hour). | 12:18 |
| sforshee | FourDollars: that patch gets INPUT_PROP_BUTTONPAD set for the xps 13 touchpad in i2c mode, though 015fdaa9f8edd89a456b3331088e1b77ebdad9d0 does as well | 12:36 |
| genkgo | jsalisbury: could you say which patches are applied to that build? | 12:45 |
| jsalisbury | genkgo, yes, they are the following: | 12:47 |
| jsalisbury | scsi: storvsc: Increase the ring buffer size | 12:47 |
| jsalisbury | scsi: storvsc: Size the queue depth based on the ringbuffer size | 12:47 |
| jsalisbury | scsi: storvsc: Always send on the selected outgoing channel | 12:47 |
| jsalisbury | scsi: storvsc: Retrieve information about the capability of the target | 12:47 |
| jsalisbury | scsi: storvsc: Fix a bug in copy_from_bounce_buffer() | 12:47 |
| jsalisbury | scsi: storvsc: Don't assume that the scatterlist is not chained | 12:47 |
| jsalisbury | scsi: storvsc: Set the tablesize based on the information given by the | 12:47 |
| genkgo | jsalisbury: thanks, how about this one: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1454758? | 12:47 |
| ubot5 | Launchpad bug 1454758 in linux (Ubuntu Wily) "[Hyper-V] storvsc: Set the SRB flags correctly when no data transfer is needed" [Critical,Triaged] | 12:47 |
| jsalisbury | genkgo, that particular commit is not in that test kernel. It was cc'd to upstream stable and should come down through the normal stable process. | 12:50 |
| jsalisbury | genkgo, I can build a test kernel the that commit and the previous 7 I posted | 12:51 |
| genkgo | jsalisbury: i am not able to qualify whether the particular bug we are having (backups with hyperv) is related to that one, or to the storvsc onw. can you? | 12:52 |
| jsalisbury | genkgo, I don't believe it is related to eh backups bug, which is why it was requested in a seperate bug report | 12:53 |
| genkgo | jsalisbury: just to be sure: the build you created fixes a.o. the backup bug, right? | 12:54 |
| jsalisbury | genkgo, correct | 12:54 |
| genkgo | jsalisbury: lovely :) | 12:54 |
| jsalisbury | genkgo, I'll be building Utopic and Trusty test kernels as well, the commits are requiring a little backporting, so taking a little longer than expected. | 12:55 |
| genkgo | jsalisbury: ok, but i need the one you already have builded, so that is no problem | 12:56 |
| jsalisbury | genkgo, cool | 12:56 |
| genkgo | jsalisbury: at least, i do not see a reason why i should go back to 3.13 kernel | 12:56 |
| FourDollars | sforshee: So what's the difference between yours and the original one? I mean if there is no your commit, what's going to happen? | 13:08 |
| sforshee | FourDollars: basically they arrived on the list at about the same time, and they went ahead and took both. For the xps 13 the results are identical; possibly for some other touchpads one or the other might not work. | 13:11 |
| FourDollars | sforshee: I see | 13:13 |
| FourDollars | sforshee: Thanks for your explanation. | 13:13 |
| sforshee | FourDollars: obviously I like my approach better ;-) | 13:13 |
| FourDollars | sforshee: haha | 13:13 |
| genkgo | jsalisbury: i am having dependency problems | 16:38 |
| jsalisbury | genkgo, I now have test kernels for Trusty and Utopic if your willing to test? All the test kernels are available from: http://kernel.ubuntu.com/~jsalisbury/lp1445195/ | 16:39 |
| jsalisbury | genkgo, Can you post any errors you are seeing | 16:39 |
| genkgo | jsalisbury: http://paste.ubuntu.com/11247594/ | 16:40 |
| genkgo | jsalisbury: while runing sudo dpkg -i *.deb | 16:40 |
| jsalisbury | genkgo, try only installing one package at a time. First the linux-image package then the linux-image-extra package | 16:41 |
| genkgo | jsalisbury: no difference | 16:42 |
| genkgo | jsalisbury: could it be the same dependency errors that i was having before? | 16:43 |
| genkgo | apw: could this be the same dependency problem? | 16:43 |
| genkgo | jsalisbury: there is only a problem with tools and cloud tools | 16:44 |
| genkgo | image headers and extra are doing fine | 16:44 |
| jsalisbury | genkgo, you won't need linux-tools or linux-cloud for this testing, only linux-image and linux-image-extra | 16:45 |
| jsalisbury | genkgo, So these two .debs: | 16:45 |
| jsalisbury | linux-image-3.19.0-17-generic_3.19.0-17.17~lp1445195_amd64.deb | 16:45 |
| apw | jsalisbury, yes if you are installing a "standard" kernel, not lts-foo, then you need to switch to the default linux-tools-common things | 16:46 |
| jsalisbury | linux-image-extra-3.19.0-17-generic_3.19.0-17.17~lp1445195_amd64.deb | 16:46 |
| genkgo | jsalisbury: those are installed | 16:46 |
| jsalisbury | apw, is linux-tools needed for testing these hyper-v commits? I didn't think so | 16:47 |
| genkgo | jsalisbury: microsoft is saying they are required | 16:48 |
| genkgo | jsalisbury: they are required, in which sense (backuping or otherwise) is unknown | 16:49 |
| genkgo | jsalisbury: http://paste.ubuntu.com/11247689/ | 16:51 |
| genkgo | jsalisbury: so i still have the linux-lts-vivid-cloud-tools-common package | 16:52 |
| jsalisbury | genkgo, Hmm, I wonder if we need to remove that first. Is this on a non-production system, so we can remove it first. Then run: | 16:53 |
| jsalisbury | sudo apt-get install -f | 16:53 |
| jsalisbury | sudo apt-get clean | 16:53 |
| jsalisbury | sudo dpkg -i linux-tools-3.19.0-17-generic_3.19.0-17.17~lp1445195_amd64.deb | 16:53 |
| genkgo | jsalisbury: this is production :) | 16:55 |
| genkgo | jsalisbury: not working, same error: linux-tools-3.19.0-17-generic depends on linux-tools-3.19.0-17; however: linux-tools-3.19.0-17-generic depends on linux-tools-3.19.0-17; however: | 16:56 |
| jsalisbury | genkgo, let me see if I can install it locally | 16:59 |
| apw | i think you need to remove the lts-foo versions of the common tools packages, which in fixed packages wont exist | 17:00 |
| apw | and then you ought to get the appropriate ones installed instead, the ones that got removed last time when installing lts-vivid | 17:00 |
| jsalisbury | genkgo, so remove this one: linux-lts-vivid-cloud-tools-common | 17:00 |
| genkgo | jsalisbury: already did | 17:01 |
| genkgo | no differece | 17:01 |
| jsalisbury | genkgo, can you run the dpkg -l command again, to see what else is there | 17:02 |
| genkgo | jsalisbury: | 17:04 |
| genkgo | jsalisbury: http://paste.ubuntu.com/11247908/ | 17:04 |
| jsalisbury | genkgo, can you purge all of the linux-tools pacakges then reinstall just linux-tools-3.19.0-17-generic_3.19.0-17.17~lp1445195_amd64.deb | 17:06 |
| jsalisbury | sudo dpkg --purge PACKAGENAME | 17:06 |
| genkgo | jsalisbury: sorry, but same issue | 17:07 |
| genkgo | http://paste.ubuntu.com/11247941/ | 17:08 |
| genkgo | you can see, all purged, won't install | 17:08 |
| genkgo | hmm | 17:08 |
| jsalisbury | genkgo, It looks like it did install, see line 586 | 17:08 |
| jsalisbury | iU linux-tools-3.19.0-17-generic 3.19.0-17.17~lp1445195 amd64 Linux kernel version specific tools for version 3.19.0-17 | 17:08 |
| genkgo | no, it got errors while installing | 17:09 |
| jsalisbury | genkgo, same dependency error: | 17:10 |
| jsalisbury | ? | 17:10 |
| jsalisbury | genkgo, hmm, I wound if it's because I added ~lp1445195 onto the package name. It doesn't cause issues with the kernel, but maybe the tools package doesn't like it | 17:11 |
| jsalisbury | genkgo, let me build one more kernel without that extra text | 17:11 |
| genkgo | jsalisbury: ok, i will wait for that, my system is pretty clean now: http://paste.ubuntu.com/11248013/. | 17:15 |
| jsalisbury | genkgo, ok, building now, it should only take 15minutes or so | 17:16 |
| genkgo | jsalisbury: i just got a reply from microsoft on my bug report | 17:36 |
| genkgo | https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1456985 | 17:36 |
| ubot5 | Launchpad bug 1445195 in linux (Ubuntu Wily) "duplicate for #1456985 [Hyper-V] Kernel patches for storvsc" [High,Triaged] | 17:36 |
| genkgo | https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1456985 | 17:37 |
| genkgo | jsalisbury: he is telling we also need https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1454758, which is not included in your build | 17:37 |
| ubot5 | Launchpad bug 1454758 in linux (Ubuntu Wily) "[Hyper-V] storvsc: Set the SRB flags correctly when no data transfer is needed" [Critical,Triaged] | 17:37 |
| jsalisbury | genkgo, Ok, I'll add it to my next build. | 17:38 |
| genkgo | jsalisbury: ok, thanks a lot | 17:38 |
| genkgo | i will wait for that before testing, thanks for the stuff you have already done | 17:39 |
| jsalisbury | genkgo, np | 17:41 |
| jsalisbury | genkgo, I have an updated kernel available at: http://kernel.ubuntu.com/~jsalisbury/lp1445195/vivid/ | 18:30 |
| infinity | pkern: The trusty kernel with that fix should be hitting -proposed on mirrors over the next 15-30 minutes (it just hit disk on ftpmaster). | 18:54 |
| genkgo- | jsalisbury: still have dependency problems: Package linux-cloud-tools-3.19.0-18 is not installed. | 20:14 |
| genkgo- | Same for the tools package | 20:15 |
| jsalisbury | genkgo, I'll have to investigate further. We may be able to just use the --ignore-depends with dpkg since this is only a test kernel. | 20:18 |
| genkgo- | apw: can this be related to dependency problems we had before? | 20:18 |
| infinity | No, it's not related to the previous issue, it's that jsalisbury builds his test kernels a bit differently from how they're built in the archive, I'd guess. | 20:19 |
| infinity | Oh, and it's a vivid kernel, not an lts-vivid kernel, which changes things somewhat. | 20:19 |
| infinity | Hrm. That tools package is indeed useless. :P | 20:21 |
| infinity | Just a bunch of symlinks into the missing package that you don't have built. | 20:22 |
| jsalisbury | genkgo, we might get away without installing tools just to test this bug, but I'll test locally to see if I can get it to install. | 20:24 |
| infinity | jsalisbury: AFAIK, he needs the tools for the backup thingee to work. | 20:25 |
| infinity | jsalisbury: And your build didn't produce all the right packages. | 20:25 |
| genkgo- | infinity: that is what i believe too | 20:25 |
| infinity | jsalisbury: Did you just do a binary-generic build instead of a full build? | 20:25 |
| genkgo- | i dont know how ignore depends works | 20:26 |
| infinity | genkgo-: Don't bother, ignoring deps will just leave you a bunch of dangling symlinks. | 20:26 |
| jsalisbury | infinity, I used this: | 20:26 |
| infinity | No actual binaries. | 20:26 |
| jsalisbury | fakeroot debian/rules clean && fakeroot debian/rules binary-headers && fakeroot debian/rules binary-generic | 20:26 |
| jsalisbury | so thats probably it | 20:26 |
| infinity | jsalisbury: Yeah. Do a full build instead. | 20:26 |
| infinity | jsalisbury: THen you'll get the common packages, etc. | 20:26 |
| jsalisbury | infinity, ack, thanks. First time someone needed the tools package from me :-) | 20:27 |
| jsalisbury | infinity, do you happen to know the option for a full build off hand? If not, I'll dig it up | 20:29 |
| apw | jsalisbury, to make the tools you need binary-tools as well | 20:30 |
| jsalisbury | apw, got it, Thanks! | 20:31 |
| jsalisbury | genkgo, I should have an updated tools package for you shortly | 20:33 |
| genkgo- | jsalisbury: aight, i will make sure to go ahead with it right away, thanks again! | 20:35 |
| jsalisbury | apw, doing a 'fakeroot debian/rules binary-tools' produced an error: | 20:36 |
| jsalisbury | dpkg-gencontrol: error: package linux-image-3.19.0-18-tools not in control info | 20:36 |
| jsalisbury | dh_gencontrol: dpkg-gencontrol -plinux-image-3.19.0-18-tools -ldebian/changelog -Tdebian/linux-image-3.19.0-18-tools.substvars -Pdebian/linux-image-3.19.0-18-tools returned exit code 255 | 20:36 |
| jsalisbury | debian/rules.d/2-binary-arch.mk:407: recipe for target 'binary-tools' failed | 20:36 |
| jsalisbury | make: *** [binary-tools] Error 2 | 20:36 |
| flexiondotorg | infinity, I've had a few sponsor requests turned down. | 20:40 |
| flexiondotorg | infinity, Please take a peek at this - https://bugs.launchpad.net/ubuntu-mate/+bug/1456591 | 20:40 |
| ubot5 | Launchpad bug 1456591 in ubuntu-mate-artwork (Ubuntu) " ubuntu-mate-artwork 0.4.8 bug fix release [debdiff attached]" [Wishlist,Incomplete] | 20:40 |
| flexiondotorg | infinity, I'm just wanting to know if this is standard practice? | 20:41 |
| flexiondotorg | infinity, https://bugs.launchpad.net/ubuntu-mate/+bug/1456597 | 20:41 |
| ubot5 | Launchpad bug 1456597 in ubuntu-mate-settings (Ubuntu) " ubuntu-mate-settings 0.4.5 release [debdiff attached]" [Wishlist,Incomplete] | 20:41 |
| flexiondotorg | sorry. Wrong channel. | 20:43 |
| jsalisbury | apw, ah ha. This what is needed to build the tools: fakeroot debian/rules binary-perarch | 21:25 |
| jsalisbury | genkgo, There is a new tools package called linux-tools-3.19.0-18_3.19.0-18.18_amd64.deb at the URL: | 21:32 |
| jsalisbury | http://kernel.ubuntu.com/~jsalisbury/lp1445195/vivid/ | 21:32 |
| garrettr | I'm working on building Ubuntu kernels w/ grsecurity | 21:45 |
| garrettr | The command I use to build the kernel is make-kpkg --initrd --overlay-dir=../ubuntu-package kernel_image kernel_headers | 21:46 |
| garrettr | Where ubuntu-package is a copy of /usr/share/ubuntu-package from an up-to-date version of Ubuntu trusty (currently running 3.13.0-53) | 21:46 |
| garrettr | I have seen other guides recommend download the ubuntu-trusty git repo from kernel.ubuntu.com, and replace some of the Debian control scripts from /usr/share/kernel-package with the ones in the git repo | 21:48 |
| garrettr | My questions are: 1. why do this? 2. why don't the control scripts in ubuntu-trusty/debian/control-scripts match those in /usr/share/kernel-package/{image,headers}? | 21:49 |
| bjf | sbeattie, i'm following up on the apparmor tests | 22:37 |
| bjf | sbeattie, i'm looking at: http://kernel.ubuntu.com/testing/dagmar__3.13.0-53.89__2015-05-20_21-43-00/ubuntu_qrt_apparmor/results/ubuntu_qrt_apparmor.test-apparmor.py/debug/ubuntu_qrt_apparmor.test-apparmor.py.DEBUG.html | 22:37 |
| cyking | jsalisbury, let me know when/if you have time to assist with that bug report. thanks :) | 23:23 |
| jsalisbury | cyking, sorry, I got side tracked today, I'll have the next kernel built shortly. | 23:26 |
| cyking | jsalisbury, all good! thanks. | 23:27 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!