/srv/irclogs.linaro.org/2012/01/03/#linaro.txt

rustyjk-: hi!00:41
jk-hey rusty!00:44
jk-rusty: got bored of powerpc? :D00:44
rustyjk-: Naah, just looking to port lguest to ARM.00:46
jk-rusty: oh nice00:46
rustyjk-: Then I can put it my flying car....00:46
rustyjk-: I think it unlikely that lguest ARM is ever going to rise high enough on my prio list to become reality.  However, others seem interested, so you never know...00:50
=== 18WAALUCN is now known as jkridner_
beaglebonerSorry03:10
=== Guest66638 is now known as lag
amitkmorning08:00
hrwhi08:42
benjiGplars: hello, do I need special rights to submit a job on validation.linaro.org ? 08:51
zygahi08:52
zygabenjiG, yes08:52
benjiGzyga: who is able to give me the rights ? 08:53
zygabenjiG, many people can, including me08:53
benjiGzyga: that is a good news :-) can you upgrade my rights ? 08:54
zygabenjiG, sure, would you mind sending a message to validation@linaro.org for bookkeeping?08:57
zygabenjiG, and please tell me your launchpad user name08:57
benjiGzyga: mail send 09:00
zygabenjiG, is it benjamin-gaignard?09:00
benjiGyes09:00
zygaah09:00
zygaexcellent09:00
zygadone09:00
benjiGlets try to submit a job 09:01
zygabenjiG, ping me if you have any problems with that09:01
* zyga gets back to making THAT part more user friendly :-)09:01
benjiGzyga: "submitted as job id: 7589" :-)09:01
zyganice :)09:01
mabacrsalveti, pong09:07
hrwzyga: have a moment to talk about lava jobs and ssh mode?09:20
zygahrw, sure09:20
zygahrw, here or in private/pl?09:20
hrwzyga: query will remove layers of translations ;)09:21
hrwbug 91114209:54
ubot2`Launchpad bug 911142 in lava-project "Using custom tests in the LAVA farm is not documented" [Critical,Confirmed] https://launchpad.net/bugs/91114209:54
benjiGis git.linaro.org working fine for you ? I can't see the lastest commit I have done in my repo ...   10:18
asacbenjiG: did you push?:10:22
asac:)10:22
benjiGasac: yes :-) a "cat refs/heads/master" give me the same commit ID than I have on local 10:23
benjiGasac: but if I clone (or look at git.linaro.org) the repo this latest commit isn't fetched 10:25
pfalconHi! Happy New Year!10:36
hrwpfalcon: HNY!10:38
hrwpfalcon: will you celebrate HNY with Ortodox calendar too?10:39
pfalconhrw: yeah, "old" new year on 13th ;-)10:40
asacpfalcon: welcome back!10:43
hrwdoko: can you document things like 'renamed binutils-2.22.tar.xz to binutils-2.22.0.tar.xz' in changelogs?10:45
dokohrw: ohh, it should not ... will revert10:50
zygafabo, hi10:50
zygafabo, can we IRC today? I have no mumble gear with me10:50
hrwdoko: thx10:50
hrwdoko: I just had to upload cross toolchain update when pbuilder hit me with build error10:50
hrwdoko: as side effect I dropped code duplication in my debian/rules ;D10:51
fabozyga: hi, sorry I have troubles with my provider... flaky internet connection11:13
zygafabo, sure, no problems11:13
zygafabo, would you mind if we did our daily on irc?11:14
fabozyga: no problem, I hope to stay connected long enough11:14
=== leio_ is now known as leio
rsalvetimabac: hey, are you the one taking care of status.linaro now?11:57
rsalvetimabac: I want to understand why one card we have is not showing there, as we also have a few bps linking into it 11:58
hrw libglib2.0-dev-dbgsym : Wymaga: libglib2.0-dev (= 2.30.0-0ubuntu4) ale 2.30.0-0ubuntu4linaro9 ma zostańá zainstalowany12:00
hrwsorry for pl_PL12:00
hrwrsalveti: can you bump glib2.0 to 2.30.0-0ubuntu4linaro10 in linaro overlay12:02
hrw?12:06
mabacrsalveti, I'm part of that effort at least :) which card is it and can you point me to one or a couple of the bps?12:10
rsalvetihrw: sure, wonder why this package is depending on the old version12:19
hrwrsalveti: because it is present in ubuntu and not present in linaro12:19
rsalvetihrw: well, in theory linaro9 would have all the dbgsym packages for it12:20
rsalvetiat least they are generated the same way by launchpad12:20
hrwI have to check ddebs - moment12:21
rsalvetimabac: https://blueprints.launchpad.net/linaro-ubuntu/+spec/ci-cross-linaro-gcc-test-definition and https://blueprints.launchpad.net/linaro-ubuntu/+spec/ci-native-linaro-gcc-lava-test-enablement12:21
rsalvetimabac: Roadmap id: LINUX2011-TOOLCHAIN-CI12:21
rsalvetihttps://linaro-public.papyrs.com/public/4120/LINUX2011-TOOLCHAIN-CI12:21
hrwlibglib2.0-dev-dbgsym exists there12:22
rsalvetihrw: maybe this was changed once suihkulokki converted the package to multiarch12:22
rsalvetisuihkulokki: around?12:23
mabacrsalveti, I think it looks ok? http://status.linaro.org/card/LINUX2011-TOOLCHAIN-CI.html12:25
rsalvetimabac: oh, ok, someone moved it to 2012Q112:30
rsalvetieven if we started it before, it's just tracked at 2012Q112:30
mabacrsalveti, yup that's where the roadmap (Kanbantool) decides it is scheduled12:31
rsalvetihm, don't know if that's the behavior we want 12:31
rsalvetifor example, we have it scheduled to be done at 2012Q1, but we start it at 2011Q412:31
rsalvetithe work done at 2011Q4 will not be part of that view12:32
rsalvetibut well, maybe this was what was planned anyway12:32
mabacrsalveti, that's right. but a Card can't span Lanes so according to the roadmap process that's correct12:33
mabacrsalveti, I'm open for all changes as long as you all agree to how s.l.o should behave :O12:33
mabac:)12:33
rsalvetimabac: sure, will discuss this with some other folks then, thanks for the help :-)12:34
mabacrsalveti, no problem. kiko and joey would be good to talk to about that I guess12:34
* mabac will fight the flu by lying down a bit. bbl12:35
rsalvetiyup12:35
rsalvetimabac: get better soon12:35
hrwlinaro-nano was 414MB, now is 1.1GB with dbgsym added (and lot are missing)12:39
rsalvetihrw: yup, that issue happened when suihkulokki converted the package to multiarch, moving arch stuff to -dev-bin*12:42
rsalvetihrw: who is depending on this package?12:42
hrwrsalveti: nothing - I just noticed that issue12:46
rsalvetihrw: oh, ok12:46
hrwubuntu-desktop sysroot with dbgsym is now 4.9GB instead of 1.1GB12:47
hrwand we lack many dbgsym packages12:47
dzindeepti1 fabo gesha jamestunnicliffe mabac pfalcon salgado: infra meeting in 2 min.12:59
kikomabac, sure, let's talk about it13:01
kikorsalveti, I don't really know how to handle that. ideally the card should have been split in two, I think.13:01
dzinmabac: ping13:05
rsalvetikiko: yeah, guess splitting it into two would make more sense13:08
hrwrsalveti: generation of sysroot takes less time then creating archive with it. not counting upload times...13:12
rsalvetihrw: cool, and what about the missing dbgsym packages?13:26
rsalvetiwhat is the issue?13:26
rsalvetior just that you didn't request to install all packages yet?13:27
hrwthey do not exist 13:27
hrwrsalveti: if package is not using debhelper then it has to call pkg-create-dbgsym (or dh_strip) by hand or will not have dbgsym package.13:29
rsalvetihrw: yeah, true, it's just a hook using debhelper 13:29
hrwrsalveti: look at bug 91118813:30
ubot2`Launchpad bug 911188 in pkg-create-dbgsym "Lack of *-dbgsym package" [Undecided,New] https://launchpad.net/bugs/91118813:30
rsalvetiprobably most of the old style packages13:30
hrwyep13:31
hrwor scripted ones13:31
rsalvetiyeah, that's a pain13:32
hrwuploading alip and nano now13:40
suihkulokkirsalveti: the multiarch work shouldn't affect any dbgsym packages13:41
rsalvetisuihkulokki: it was just that once the bins were moved from the -dev package, the dbgsym for it was not created anymore13:43
hrwand we use multiarch package in overlay 13:43
rsalvetiwhich is the correct behavior 13:43
hrwcreating of sysroot + archiving it takes less time then uploading13:46
suihkulokkirsalveti: doesn't that just mean that instead of libglib2.0-dev-dbgsym, a libglib2.0-dev-bin-dbgsym should be generated?13:50
rsalvetisuihkulokki: that was what happened 13:50
suihkulokkiso what is the issue? :)13:51
rsalvetiit's just that once you fixed the package, the -dev-dbgsym package doesn't exist anymore, but it's is there at the ubuntu archive13:51
rsalvetiso if people try to install it, as probably what hrw did, it'll complain about broken dependencies 13:51
=== nytowl_away is now known as nytowl
hrwwe have just 20GB free on people.linaro.org13:54
hrwmy sysroots upload will eat 3GB13:54
hrw300KB/s upload suxx13:56
plarszyga: hey, how's it going?13:57
zygaplars, hi13:57
zygaplars, well, adjusting to new workplace here13:57
plarszyga: up for re-attempting push to staging/production today?13:58
plarszyga: ah, yes :)13:58
zygaevening? sure 13:58
plarszyga: you are still in poland?13:58
zygaI'll have updates to l-d-t by then13:58
zygaplars, yes 13:58
zygaplars, we don't have the documents yet13:58
zygaplars, and I decided to stay rather than go to .es on my own13:58
plarszyga: yeah :(13:58
zygaplars, plus the tickets are very expensive now13:58
zygaplars, likely we'll return in 2-3 weeks if we manage to get a good deal on the plane 13:59
zygaplars, documents should be ready in a few days13:59
zygaplars, how was your holiday?14:02
plarszyga: busy, ready to get back to work on LAVA :)14:02
zygaplars, same here :)14:03
dzinrsalveti: ping14:04
rsalvetidzin: hey, do we have a meeting now?14:04
rsalvetidzin: just saw it at my calendar but don't remember which day was our sync :-)14:04
dzinrsalveti: yup :-) mumble14:04
rsalvetilet me grab my headset14:04
anmarmansson_: Hello Tony. Happy New Year!14:31
anmarmansson_: Are you guys planning on fixing any bugs related to GingerBread or just focusing on ICS moving forward?14:32
mansson_Hello. Same to you!14:32
mansson_Not sure we have any decision on that, but it's a resource issue, and I'm sure we put priority on ICS.14:33
mansson_Any particular bug?14:33
faboanmar: mansson_: a lots of bugs have been set to wontfix14:55
faboMarking Gingerbread open bugs to "Won't fix" state (on Linaro-Android14:55
fabohelpdesk) as fixes are done only on ICS track now. Please raise this bug14:55
faboagain against latest Linaro-ICS release if it still exist.14:55
anmarfabo: makes sense. thanks for the clarification and Happy New Year :-)14:56
faboyou too :) happy new year14:57
asacfabo: how do we determine if something is gingerbread specific? or is that usually obvious?15:01
faboasac: I hope people tries to reproduce the initial bug on ICS and set wontfix accordingly15:04
faboasac: though I'm not sure it's done that way. amit has done most of this work, he should be able to tell us more15:04
faboamit - > pundiramit15:05
hrwI am starting to like tmux ;)15:17
hrwok, who has working mx53loco?15:22
hrw3.1.0-1002-linaro-lt-mx5 does not boot on mine15:23
fabohrw: tgall_foo 15:28
Dr_Whohrw: problems ?15:29
hrwDr_Who: [    2.407589] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)15:29
hrwDr_Who: thats last message15:29
Dr_Whohmm can't say that I've seen that one before15:30
hrwDr_Who: thats with hdd connected15:30
hrwunplugged and rebooted15:30
hrwI want to hurt a bit designer of mx53loco15:31
hrw[    2.407559] ata1: SATA link down (SStatus 0 SControl 300) 15:31
hrwand halt15:31
Dr_WhoI've found the vga connector is flimsy, I'm managed to break mine on my first one ...15:32
hrwnever used15:32
Dr_Whowell I'll see if I can replicate here in a bit15:32
hrwin worst case I will regenerate rootfs15:33
hrw~curse panda for having one sd slot by defaul15:33
hrwt15:33
hrwand then curse flash-kernel15:37
hrwKernel /boot/vmlinuz-3.1.0-4-linaro-omap does not match your subarchitecture omap4, therefore not writing it to flash.15:37
hrwanyway will try linux-image-3.2.0-1403-omap4 on one panda15:39
plarsmansson_: hi15:41
plarsmansson_: can you tell me more about the test suite at http://git.linaro.org/gitweb?p=people/mansson/snowball/ltp.git;a=summary ?15:41
mansson_plars: Not much15:41
mansson_They were basically dumped there.15:42
mansson_Its a set of LTP tests used at ST-Ericsson.15:42
plarsmansson_: do you know what kind of shape they are in? is someone at STE actively running or maintaining them?15:43
mansson_I was told later that the proper thing to do was to wield out the ST-Ericsson specific stuff but that never got done.15:43
mansson_I think they are maintained internally. 15:44
mansson_They should go upstream but there are no resources.15:44
plarsmansson_: cool, we'd like to take a look at getting them to run under lava at least15:46
hrwlava...15:47
hrwbug 91114215:47
ubot2`Launchpad bug 911142 in lava-project "Using custom tests in the LAVA farm is not documented" [Critical,Confirmed] https://launchpad.net/bugs/91114215:47
mansson_plars: Good, that was the intention.15:47
tgall_fookrtaylor, ping15:47
hrwplars: can I suggest solving of that bug by someone from lava team?15:47
plarshrw: of course, you did that by filing it this morning :)15:49
hrwplars: and I plan to remind it at least once per day15:49
plarshrw: just out of curiosity, what's the test you are wanting to run?15:49
hrwplars: you have lot of hw there and I want to make use of it15:49
ibirisplars: fabo: ping - there is some need for help to run benchmarks for armhf - using a panda + an ubuntu latest image from http://cdimage.ubuntu.com/daily-preinstalled/current/15:49
hrwplars: apt-get install gcc-linaro + build some packages with it15:50
ibirisplars: fabo: is there someone at validation who could help markos_ with the armhf benchmarks?15:50
ibirisplars: fabo: btw markos_ will give assistance, even though the task is not complicated (just run the benchmarks and wait for the tests to finish)15:51
plarshrw: package build? is this actually testing something, or are you trying to augment the available builders?15:51
plarsibiris: are these automated tests that we could make available as a test in lava-test so that they can be run automatically?15:52
hrwplars: testing toolchainwg compiler (nearly) without ubuntu changes15:52
ibirismarkos_: ^^^ see question from plars15:52
markos_plars, not package build, running tests like povray, cray, etc, most from phoronix-test-suite package15:52
hrwplars: and my bug is mostly about 'there is no way to get info how to run own test on lava farm'15:53
hrwplars: and 'pester random lava developer' is not a way15:54
plarsmarkos_: cool, we have some things from phoronix integrated, but if there are others we'd like to know15:54
asac16:46 < asac> tgall_foo: do you know what the status of kurts end2end audio testing effort is? was he able to finish all the test payload code etc. on x86 while panda audio is still broken?15:54
plarshrw: agreed, how far have you gotten so far? Istr you were working on this before the holidays right?15:55
markos_plars, almost all the benchmarks comparing debian armhf+ubuntu armel were based on phoronix, we'd like to add ubuntu armhf in the picture now15:55
hrwplars: I have test which runs fine with lava-test15:55
plarshrw: is this something you want to merge into lava-test?15:56
tgall_fooasac, my understanding is it was not completely finished however krtaylor can speak up and correct me15:56
hrwplars: it is simple job description which I want to be able to run on lava farm15:57
hrwplars: http://paste.ubuntu.com/791426/15:57
ibirisasac: about e2e - the proto for x86 was finished afaik - the code was placed in http://git.linaro.org/gitweb?p=people/kurt-r-taylor/e2eaudiotest.git for 11.12 15:57
plarshrw: the documentation is one issue, but I'm also interested in making sure you have what you need to get things going right away15:57
ibirisasac: and https://blueprints.launchpad.net/linaro-multimedia-project/+spec/linaro-mmwg-e2eaudiotesting-basic was marked as implemented15:57
hrwplars: according to examples this http://paste.ubuntu.com/791777/ should be more or less proper job description.15:59
asacibiris: thanks. is there a lava instrumentation blueprint that will ensure that all boards will get audio to audio cables?15:59
plarshrw: you've tried that then?15:59
asaci think thats important to prepare everything ... and then the stuff needs to be added to the set of tests used by our daily LEB lava-test runs16:00
dzinamitk, rsalveti, tgall_foo, dsaxena, plars, mansson_, JesseBarker,  ramana: sync for WG -> plat in #linaro-meeting16:00
JesseBarkerdzin, i have another meeting at this time.16:00
ibirisasac: https://blueprints.launchpad.net/linaro-multimedia-project/+spec/linaro-mmwg-e2eaudiotesting-lava16:00
hrwplars: I do not want to setup whole lava infrastructure on my systems just to test is it ok. I tested my test locally with lava-test (big thanks to zyga for help) and it works.16:00
hrwplars: from my understanding job description needs running scheduler/dispatcher and I want to avoid running it locally16:01
dzinJesseBarker: ok, is there a better time for you. michaelh1|away can not attend either16:01
dzinJesseBarker: is there someone who can represent Graphics?16:01
JesseBarkerdzin, today or just in general?16:01
ibirisdzin: apologies I will not be able to attend either (even if optional)16:02
krtaylorasac, tgall_foo - yes it is completed - needs to be tested on panda and then ported to Android16:02
plarshrw: I'm confused then, it sounds like this works for you on lava, but why are you thinking you need to run a full lava stack locally?16:02
dzinJesseBarker: both16:02
ramanadzin: What's this sync-up about ?16:03
JesseBarkerdzin, current time is generally fine, just not on tuesdays.16:03
dzinramana: WG -> platform issues16:03
tgall_fookrtaylor, so by completed, completed on intel at least :-)16:03
hrwplars: lava-test is running locally on my machines. I want to be able to just set 'run this set of tests on any arm machine' and be able to look at results16:03
ramanadzin: ok. if this is going to be a regular meeting then I have a conflict every alternate Tuesday16:04
hrwplars: and for me lava-test (tool running locally) != lava (dispatcher/scheduler/hwfarm)16:04
krtaylortgall_foo, yes, that was the question re: x86 right?16:04
dzinramana: just talking about rescheduling.16:04
plarshrw: but you said that you've tried submitting http://paste.ubuntu.com/791777/ to lava and it worked for you right?16:05
hrwplars: no, I never submitted that job16:06
tgall_fookrtaylor, it's cool just being clear16:06
tgall_fookrtaylor, do you have an option to use audio over HDMI ?16:06
plarshrw: ah, ok, I think I understand now then16:06
krtaylortgall_foo, yes, device is selectable - needs to be tested as a part of the panda blueprint16:07
krtaylortgall_foo, I just fixed the priorities on these BPs, we need to talk about what happens this month16:08
joeydoanac: that was some awesome jerky.  The ingredients were wild16:10
plarshrw: so http://lava-scheduler.readthedocs.org/en/latest/usage.html#submitting-jobs talks about how to submit a job like that16:11
doanacjoey: glad you liked it.16:12
dzinmounirb: are you around?16:16
hrwplars: sure, but submitting job which uses not existing test is rather useless...16:17
mounirbdzin: yes16:19
dzinmounirb: want to join the plat->wg meeting in #linaro-meeting?16:19
mounirbdzin:yes - just did16:20
salgadodzin, do we have IRC logs for today's meeting already?16:25
salgadofabo, about your work on getting lmc running on Fedora, I guess you've seen https://wiki.linaro.org/Platform/DevPlatform/Ubuntu/ImageInstallation/NonUbuntu ?16:26
dzinsalgado: yes, I've not entered them yet16:26
dzinsalgado: http://irclogs.linaro.org/meeting-logs/linaro-meeting/2012/linaro-meeting.2012-01-03-13.01.wiki.txt16:26
salgadodzin, cool, thanks!16:27
=== ramana is now known as ramana`
=== ramana_ is now known as ramana
salgadowow, that was a really short one. :)16:27
fabosalgado: yes16:33
zygaplars, busy morning16:40
plarszyga: indeed16:40
zygaplars, the bug hrw is talking abut is about not having a documented way (on lava.rtfd.org) that describes how to run custom tests in our farm16:40
zygaplars, I was thinking that we should have a few task-oriented chapters there16:40
plarszyga: yeah, but arguably, lava.rtfd.org may be the wrong place for information specific to validation.linaro.org.  Wasn't that the point of the FAQ stuff that we started a while back? put the generic documentation on rtfd, and the site-specifics on the faq wiki?16:42
zygaplars, true, but this is not that specific, it's about running external tests, we can omit "ping plars for access" from the docs ;-)16:43
plarszyga: s/plars/validation@linaro.org16:45
plarsbut yeah, I agree16:45
zygaplars, true, fortunately most people go here to get this sorted out so they ping you ;-)16:45
hrwand s/hrw/author of new benchmark/ and you get other usecase16:50
joeyakgraner: nice format on the activity email, thanks16:54
akgranerjoey, thanks!16:55
hrwok. all 3 sysroots uploaded to people.linaro.org17:01
=== salgado is now known as salgado-lunch
* zyga finds WRAP interesting17:09
zygahttp://hueniverse.com/2009/11/wrap-and-the-demise-of-the-oauth-community/17:09
=== vila is now known as bzr-RM
=== bzr-RM is now known as vila
hrwhave a nice rest of day17:15
=== ramana is now known as ramana_away
=== salgado-lunch is now known as salgado
zmaHow do you guys work with closed source drivers? For example some SGX drivers are closed source, are those closed source drivers compatible with other graphics subsystems than X.org?18:26
rsalvetizma: we generally use the drivers provided by the vendor 18:31
rsalvetizma: and they usually work with just one architecture, like for SGX (on normal linux) it only works with X1118:31
rsalvetiand with an specific ABI18:32
zmazma: but is the driver interface specificly for X only? For example SGX540?18:32
zmarsalveti: I see18:32
rsalvetizma: it's not, as you usually just use egl/gl, but the rendering part uses the X11 driver18:32
rsalvetithe windowing system needs to be compatible with the driver18:33
rsalvetiin theory the vendor is able to build the driver using just the framebuffer or other windowing system18:33
rsalvetibut it's a build time option 18:33
rsalvetiand we only get the final binaries 18:33
zmaI wonder, would there be any sense to build Qt to use DirectFB using X11 driver?18:34
zmahmm, yeah.. I'm getting the full picture now. Thanks.18:34
zmaI wish more open source drivers would exist18:35
rsalvetizma: it would make sense, and a lot of people requests a driver that's also compatible with fb18:35
rsalvetito be used with Qt and directfb 18:35
rsalvetibut at least with SGX, we only have the one compatible with X1118:36
rsalvetiand with android 18:36
asacrsalveti: we have no pure framebuffer egl for sgx? thought we had that at some point18:36
zygaasac, I think there is something like that for imx18:36
* zyga gets back to crypto18:37
rsalvetiasac: not from the current driver we're using from TI18:37
rsalvetibut I know there's another tarball that TI releases, together with the OMAP3 driver 18:37
asacrsalveti: ok, but roughly 6 month back they used to ship framebuffer + X11 in one package, no?18:37
rsalvetibut I'm not yet sure if that supports only fb 18:37
rsalvetiasac: nops, only fb never worked 18:37
asacrsalveti: oh. could be that i have seen it on the omap3 front though.18:37
zygamwhudson, hi18:41
mwhudsonhi18:41
zygamwhudson, how was your holiday?18:55
joeytgall_foo: I'm on mumble waiting for you :-) 19:02
=== michaelh1|away is now known as michaelh1
zyga-nc10re19:26
zyga-nc10mwhudson, plars: here19:26
* mwhudson is more "here" now too19:27
mwhudsonbut drowning in email, obviously19:27
zyga-nc10mwhudson: if it's important they'll re-send ;-)19:28
zyga-nc10mwhudson: may I post a challange for you?19:28
mwhudsonzyga-nc10: sure, i reserve the right to not think about it until tomorrow :)19:29
zyga-nc10mwhudson: imagine system.createToken(token, user), is there any point to ading crypto layers if (once approved on the web page) the user will simply send it in plain text again?19:31
mwhudsonzyga-nc10: i assume everything is happening over ssl here?19:32
zyga-nc10it either is or is not19:32
mwhudsonalso i guess i don't quite understand what you mean19:32
zyga-nc10regardless of ssl the answer should be the same19:33
mwhudsonno19:33
mwhudsoni mean i don't understand the original question19:33
zyga-nc10mwhudson: rephrased, is there any point in making createToken more "secure" if we use basic auth19:34
plarszyga-nc10, mwhudson: hi19:34
mwhudsonzyga-nc10: hang on, are you talking about a createToken xml-rpc method?19:34
zyga-nc10yes19:35
mwhudsonis that a good idea?  presumably the createToken call will be authenticated with a token, i'm not sure having a token should let you create more19:35
zyga-nc10well, the idea is that it is NOT authenticated at al19:35
zyga-nc10then you go to the web and agree or not19:35
zyga-nc10(it returns an URL)19:35
mwhudsonah19:36
zyga-nc10oauth all over again, without all the crypto19:36
mwhudsonah right19:38
mwhudsonthen so long as it's all over ssl, that sounds reasonably secure19:38
mwhudsonactually even if not i guess, so long as it's a https url that's returned :-)19:39
zyga-nc10my point was subtly different: if you use ssl it is secure, otherwise it was never secure in the first place19:40
mwhudsonwe really should start insisting on ssl soon i think19:41
mwhudsonplars: hi19:41
zyga-nc10mwhudson: anyway, I think that makes sense19:41
plarsmwhudson: welcome back :) have a good vacation?19:41
zyga-nc10mwhudson: I was thinking that using a token over non-ssl should blacklist the token eventually19:41
mwhudsonplars: i guess we have a call scheduled today19:41
mwhudsonplars: does that work for you?19:42
plarsmwhudson: yeah, would like to move it earlier, or later if that's ok19:42
plarsactually...19:42
mwhudsonplars: yes, good vacation (no cell phone reception for 8 days...)19:42
plarsouch19:42
plarsintentionally?19:43
mwhudsonyes19:43
plarsah, good then :)19:43
plarsI was cell phone only for quite a lot of it, and barely getting a 2g signal19:43
plarszyga-nc10: you up for a staging deployment attempt still?19:44
zyga-nc10plars: soon19:44
michaelh1plars: hey, has conmux on the farm changed?  'conmux tcpanda04' fails with a 'No such file or directory'19:49
=== jkridner_ is now known as jkridner
plarsmichaelh1: shouldn't be, let me take a look19:53
michaelh1plars: oh, sorry - my fault, I'm in the wrong user account19:54
plarsmichaelh1: the command is conmux-console, works fine for me :)19:54
plarsmichaelh1: though it seems to be hung or something, not getting a prompt19:54
michaelh1plars: yeah, it needs a reset.  Doing a PGO build kills the board.19:55
plarsmichaelh1: ouch19:55
michaelh1plars: it's an unreasonable build that uses lots of swap.  The kernel should survive but I'll let it away with this one :)19:56
plarsmichaelh1: would it help to change the swap location? are we swapping on usb or sd?19:57
michaelh1SD19:57
michaelh1I didn't want to reserve 2 GB of the USB drive and had a big SD sitting unused.  SD is unreliable though.19:58
plarsmichaelh1: I'm not sure which SD cards those tcpanda boards have, but we identified a problem with some of the SDs being outrageously slow - working to replace them all19:58
=== robclark_ is now known as robclark
* zyga-nc10 commits stuff to l-d-t21:12
zyga-nc10plars: almost ready21:14
=== nytowl_away is now known as nytowl
zyga-nc10has anyone here used git-flow?21:18
joeyJesseBarker: I can go earlier if you want to21:20
joeyI'm at a stopping spot now21:20
JesseBarkerjoey, i'm here. mumble, phone?21:20
joeymumble is good, let me just plug back in21:20
tgall_foorsalveti, did you try out the new panda kernel ?21:25
joeyJesseBarker: http://www.sno-drift.org/21:26
rsalvetitgall_foo: yup, it's working as before, but I'm still installing all extra packages needed for video decoding 21:29
rsalvetitgall_foo: give me a few and I'll ping you back21:29
tgall_foonp ?  if it's working for you, then I'll get the UCM update going so that's ready to go with the updated hwpack21:29
rsalvetitgall_foo: ok, if you want to check the kernel you can download it from the staging ppa21:30
rsalvetionce I'm able to test the video decoding I'll copy it over to the overlay21:30
rsalvetiand respin the hwpack21:30
tgall_foosounds great21:31
rsalvetijcrigby: hey, are you around today?21:32
jcrigbyrsalveti, yes I am21:33
rsalvetijcrigby: need to talk with you about the blueprints and such for 12.01, will get something to eat but should be back in 1h21:33
rsalvetiping you back21:33
jcrigbyrsalveti, ok21:33
=== nytowl is now known as nytowl_away
=== nytowl_away is now known as nytowl
zyga-nc10mwhudson: plars: I'll be done soon, sorry, tv distractes me22:29
=== nytowl is now known as nytowl_away
plarszyga-nc10: are you trying to land everything? or just the critical pieces we discussed?23:27

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!