| *** elmiko_afk is now known as elmiko | 00:10 | |
| *** Hao has joined #openstack-sahara | 00:25 | |
| *** RockKuo has joined #openstack-sahara | 00:28 | |
| *** RockKuo_ has joined #openstack-sahara | 00:28 | |
| *** RockKuo_ has quit IRC | 00:28 | |
| *** Hao has quit IRC | 00:29 | |
| *** ErikB1 has joined #openstack-sahara | 00:48 | |
| *** elmiko has quit IRC | 00:51 | |
| *** ErikB1 has quit IRC | 01:09 | |
| *** bradd1 has quit IRC | 01:55 | |
| *** Hao has joined #openstack-sahara | 02:07 | |
| *** Hao has quit IRC | 02:11 | |
| *** IlyaE has quit IRC | 02:43 | |
| *** Hao has joined #openstack-sahara | 02:50 | |
| *** Hao has quit IRC | 02:52 | |
| *** akuznetsov has joined #openstack-sahara | 05:05 | |
| *** IlyaE has joined #openstack-sahara | 05:25 | |
| openstackgerrit | Jenkins proposed a change to openstack/sahara: Imported Translations from Transifex https://review.openstack.org/82719 | 06:06 |
|---|---|---|
| *** IlyaE has quit IRC | 06:48 | |
| *** witlessb has joined #openstack-sahara | 07:31 | |
| *** tosky has joined #openstack-sahara | 07:33 | |
| *** skolekonov has joined #openstack-sahara | 07:43 | |
| *** akuznetsov has quit IRC | 07:46 | |
| *** akuznetsov has joined #openstack-sahara | 07:52 | |
| openstackgerrit | A change was merged to openstack/sahara-image-elements: Updates hadoop-hdp to use new project name in URL https://review.openstack.org/86065 | 09:26 |
| openstackgerrit | A change was merged to openstack/sahara-dashboard: Updated from global requirements https://review.openstack.org/85839 | 09:40 |
| *** akuznetsov has quit IRC | 09:43 | |
| openstackgerrit | Ivan Berezovskiy proposed a change to openstack/sahara-image-elements: Add ability to set DIB_IMAGE_SIZE parameter for CentOS images https://review.openstack.org/86270 | 09:49 |
| openstackgerrit | A change was merged to openstack/sahara: Imported Translations from Transifex https://review.openstack.org/82719 | 10:00 |
| *** akuznetsov has joined #openstack-sahara | 10:10 | |
| openstackgerrit | A change was merged to openstack/sahara: Cleanup of docs for integration tests https://review.openstack.org/86030 | 10:30 |
| openstackgerrit | Sergey Reshetnyak proposed a change to openstack/sahara-image-elements: Add inject hadoop native libs to vanilla hadoop 2 https://review.openstack.org/83036 | 11:00 |
| *** RockKuo has quit IRC | 11:09 | |
| *** _crobertsrh is now known as crobertsrh | 11:09 | |
| openstackgerrit | Sergey Kolekonov proposed a change to stackforge/sahara-ci-config: main configuration files were added https://review.openstack.org/74341 | 11:21 |
| *** witlessb has quit IRC | 11:27 | |
| tellesnobrega | hi, im trying to create a cluster with my devstack installation, but its failing, i found this error http://paste.openstack.org/show/75405/ at the heat-eng.log but its not clear what happened, where can i find more information? | 11:27 |
| *** witlessb has joined #openstack-sahara | 11:27 | |
| openstackgerrit | Sergey Kolekonov proposed a change to stackforge/sahara-ci-config: main configuration files were added https://review.openstack.org/74341 | 11:31 |
| dmitryme | tellesnobrega: amm, there is not much info the exception. Did you try to create cluster one more time? | 11:34 |
| tellesnobrega | dmitryme: yes, it gives me the same output, im going to clean my logs and start all over to see if i can get more error messages | 11:34 |
| dmitryme | did you try to just create instances manually? | 11:35 |
| tellesnobrega | dmitryme: no, i will do that | 11:35 |
| dmitryme | I have a feeling that the cause is instance entering 'error' state | 11:35 |
| tellesnobrega | dmitryme: ok | 11:35 |
| openstackgerrit | A change was merged to openstack/sahara-image-elements: [DIB] Added qemu-img dependency for CentOS host system https://review.openstack.org/86152 | 11:37 |
| openstackgerrit | Sergey Reshetnyak proposed a change to openstack/sahara-image-elements: Fix install Oozie for vanilla 2 plugin https://review.openstack.org/86294 | 11:51 |
| *** ErikB1 has joined #openstack-sahara | 11:53 | |
| *** _mattf is now known as mattf | 11:58 | |
| *** mattf has quit IRC | 11:59 | |
| *** mattf has joined #openstack-sahara | 11:59 | |
| *** sballe_ has quit IRC | 12:01 | |
| tellesnobrega | dmitryme: you were right, cant start an instance | 12:21 |
| tellesnobrega | dmitryme: is floating ips necessary? or can i have sahara running without it? | 12:45 |
| *** ErikB1 has quit IRC | 12:45 | |
| *** vrovachev has quit IRC | 12:51 | |
| *** elmiko has joined #openstack-sahara | 12:55 | |
| openstackgerrit | Sergey Lukjanov proposed a change to stackforge/sahara-ci-config: Initial skeleton https://review.openstack.org/86317 | 13:01 |
| openstackgerrit | A change was merged to stackforge/sahara-ci-config: Initial skeleton https://review.openstack.org/86317 | 13:03 |
| tellesnobrega | dmitryme: im getting this error http://paste.openstack.org/show/75409/ | 13:04 |
| *** tmckay has quit IRC | 13:15 | |
| *** Hao has joined #openstack-sahara | 13:22 | |
| openstackgerrit | Vadim Rovachev proposed a change to openstack/sahara-dashboard: Update dicts with plugin services https://review.openstack.org/86026 | 13:29 |
| crobertsrh | Anyone have any thoughts on how to best handle the merging of the sahara-dashboard into the horizon codebase? Should we have a temporary code-freeze for the dashboard until it is merged into horizon? | 13:43 |
| elmiko | that seems reasonable, reduce the amount of change while doing the merge | 13:45 |
| SergeyLukjanov | crobertsrh, hey | 13:49 |
| crobertsrh | howdy | 13:49 |
| crobertsrh | It's not like there's a ton of CRs open right now anyway, I suppose, so it might be a non-issue really | 13:50 |
| elmiko | strike while the iron is hot ;) | 13:51 |
| mattf | crobertsrh, how quickly can you get a cr into horizon in general / specific to sahara (they don't have sahara experts...)? is there relevant ci to hook up w/ as a dry run before merging? how much time would the merging take (potentially how long is the dashboard locked)? | 13:56 |
| * mattf asks and runs | 13:57 | |
| SergeyLukjanov | crobertsrh, I think that we could freeze s-dashboard till the j-1 for example | 13:57 |
| SergeyLukjanov | mattf, there is no CD in horizon | 13:57 |
| crobertsrh | Heh, all good questions. I talked briefly with some horizon peeps this morning. They didn't have a specific checklist for me to go through. I told them that I was planning to get a CR spun-up against he blueprint and we would go from there. | 13:57 |
| openstackgerrit | A change was merged to stackforge/sahara-ci-config: main configuration files were added https://review.openstack.org/74341 | 14:06 |
| elmiko | crobertsrh: have you ever seen horizon crap out with an importerror on a module named "theme"? | 14:06 |
| crobertsrh | I have not. Congrats though :) | 14:06 |
| elmiko | lol | 14:07 |
| crobertsrh | Any chance you've updated horizon recently? | 14:07 |
| elmiko | yea... the rdo install updated yesterday and completely destroyed my rhel install :( | 14:07 |
| crobertsrh | Hmm, so how are you running now? | 14:07 |
| crobertsrh | or, trying to run | 14:08 |
| crobertsrh | rdo install? | 14:08 |
| elmiko | rdo is the official rhel openstack install, i think | 14:08 |
| elmiko | but i'm running the icehouse devel stuff | 14:08 |
| crobertsrh | You're running horizon from a git clone? | 14:09 |
| elmiko | i can run horizon out of a virtualenv with trunk stuff, but when i try to hit the default installed version it craps out | 14:09 |
| crobertsrh | or from a package? | 14:09 |
| elmiko | i have both installed | 14:09 |
| crobertsrh | ah, ok, so the virtualenv works for you | 14:09 |
| crobertsrh | the packaged version is what is failing? | 14:09 |
| elmiko | the git clone is in a virtualenv, to keep it separate from the system installs | 14:09 |
| elmiko | yea, the packaged version is failing. i think it might be a django dependency but i can't find any module in django named theme | 14:10 |
| SergeyLukjanov | crobertsrh, re dashboard, it's very important for us to move it into horizon, so, we could freeze s-dashboard anyway | 14:10 |
| SergeyLukjanov | crobertsrh, additionally, I hope that it'll be not very painfull | 14:10 |
| openstackgerrit | Alexander Ignatov proposed a change to openstack/sahara: Updated validation section for Vanilla Plugin https://review.openstack.org/86338 | 14:10 |
| crobertsrh | Oh...also SergeyLukjanov, is there a way to accomplish our merge without losing our git history? | 14:11 |
| crobertsrh | +1 on hoping for minimal pain | 14:11 |
| SergeyLukjanov | crobertsrh, nope, I don't think that it's possible | 14:11 |
| crobertsrh | Ok :) That's one less thing to consider then! | 14:11 |
| SergeyLukjanov | crobertsrh, I think that we'll need at least to change some code related to the placement of sahara-dashboard as the new tab vs. as a subpanel | 14:12 |
| crobertsrh | Yeah, I'm looking at how that works right now. | 14:13 |
| SergeyLukjanov | crobertsrh, so, probably, it'll be better to start from simple commits like client addition / configuration, some list pages and etc. | 14:13 |
| SergeyLukjanov | native speakers, please, take a look at https://review.openstack.org/#/c/86338/1/doc/source/userdoc/vanilla_plugin.rst | 14:14 |
| crobertsrh | My current plan is to 1) Get the whole thing working locally, 2) Break up into some number of CRs that makes sense. | 14:14 |
| crobertsrh | oh...#3) Face a firing-squad of horizon zealots for weeks arguing the finer points of css styles. | 14:15 |
| SergeyLukjanov | crobertsrh, mm, IMO it'll be better to start pushing CRs earlier | 14:16 |
| SergeyLukjanov | crobertsrh, I can try to setup client-related stuff | 14:16 |
| SergeyLukjanov | for example | 14:16 |
| crobertsrh | ok, any help is greatly welcomed | 14:16 |
| SergeyLukjanov | crobertsrh, ok, I'll take a look on it | 14:18 |
| SergeyLukjanov | crobertsrh, if you feel that it's a too huge problem, than we could ask NikitaKonovalov with some stuff | 14:18 |
| crobertsrh | I'm just getting started now, so I'll see how it feels. | 14:18 |
| NikitaKonovalov | crobertsrh: ping me if the jQuery stuff in our dashboard becomes too much annoying | 14:23 |
| crobertsrh | ok, will do. Thanks | 14:23 |
| *** tmckay has joined #openstack-sahara | 14:27 | |
| dmitryme | tellesnobrega: floatings are not necessary, but configuring Sahara with them is much easier | 14:28 |
| tellesnobrega | dmitryme: i see, i think the problem is with nova-rootwrap | 14:29 |
| elmiko | crobertsrh: check out https://blueprints.launchpad.net/sahara/+spec/dashboard-enable-file it's small but related to our effort to integrate with horizon | 14:30 |
| crobertsrh | Yeah...that's the new way of turning on/off dashboard plugins | 14:30 |
| crobertsrh | should be much nicer | 14:30 |
| elmiko | it was suggested on the review for python-django-sahara | 14:31 |
| *** akuznetsov has quit IRC | 14:44 | |
| openstackgerrit | Sergey Reshetnyak proposed a change to openstack/sahara-image-elements: Fix install Oozie for vanilla 2 plugin https://review.openstack.org/86294 | 15:00 |
| *** IlyaE has joined #openstack-sahara | 15:01 | |
| *** bradd1 has joined #openstack-sahara | 15:12 | |
| *** akuznetsov has joined #openstack-sahara | 15:20 | |
| *** sballe has joined #openstack-sahara | 15:21 | |
| *** skolekonov has quit IRC | 15:24 | |
| tmckay | You've got to be kidding me. My pig job wasn't working because my cluster wasn't big enough, and there were not enough slots for Oozie launchers. Apparently, with Hadoop 2 and whatever build of Oozie we're using, pig runs differently than it did in 1.2.1 | 15:32 |
| tmckay | So, it just ran to a point and stopped, because it couldn't get another slot from the launcher (empirically). bigdata deadlock. | 15:33 |
| tmckay | adding another worker node fixed it | 15:34 |
| elmiko | ouch... | 15:37 |
| elmiko | but congrats :) | 15:37 |
| tmckay | elmiko, thanks. ouch is right. Been trying this since Monday morning, with an excurion into neutron just in case that had a bearing (thanks, by the way) | 15:39 |
| tmckay | excursion, that is | 15:39 |
| elmiko | and now you're running neutron too! | 15:39 |
| tmckay | yeah. Now, the question is whether or not I can wrap bigpetstore in an oozie job | 15:40 |
| tmckay | I don't know whether to laugh or cry | 15:41 |
| elmiko | lol, why not both! | 15:41 |
| tmckay | lol, <sniff> | 15:41 |
| elmiko | lol | 15:41 |
| tmckay | elmiko, underscores the need for a beefy devstack lab that we have free access to, so we can launch reasonable clusters | 15:43 |
| tmckay | crobertsrh, ^^ | 15:43 |
| elmiko | that would be cool | 15:44 |
| crobertsrh | Actually, just yesterday, I setup a sahara instance pointed at our OS1 internal setup. There is potential for some larger clusters there. | 15:45 |
| tmckay | crobertsrh, you'll have to take us to school on that. | 15:46 |
| aignatov | tmckay: interesting point you faced with EDP :0 | 15:57 |
| aignatov | :) | 15:58 |
| * aignatov thinks to open "Known issues" section in the docs | 15:58 | |
| tmckay | aignatov, yes. Under vanilla 1.2.1, I can run pig jobs no problem with a single worker node. Under vanilla 2.3.0, I need at least 2 it seems | 15:59 |
| tmckay | aignatov, I don't understand oozie logs enough to know exactly where it went wrong, but I think the oozie launcher failed when it tried to actually run the pig shell | 16:00 |
| crobertsrh | tmckay: how many nodes did you have when it failed? how many made it run? | 16:00 |
| aignatov | btw, what kind of flavor did you use? | 16:00 |
| tmckay | 2 masters, 1 worker failed (master processes were split). 2 masters, 2 workers worked. I think I will try 1 combined master, 3 workers and see how that does. | 16:00 |
| tmckay | small | 16:01 |
| tmckay | aignatov ^^ | 16:01 |
| aignatov | I think hadoop 2 can run on large flavor in single instance :) | 16:01 |
| tmckay | really? I could try that too. | 16:02 |
| aignatov | I think yes, if not you should play with Heap Sizes for all services | 16:02 |
| aignatov | probably, 1 GB per name node, resorcemanager, history server and datanode and 2GB per ooze and tasktracker | 16:03 |
| *** tmckay is now known as tmckay_bbl | 16:05 | |
| aignatov | actually all real mapreduce workflows depends on the data to be processed and active resources dedicated for that | 16:06 |
| aignatov | capacity planning is big theme in big data world :) | 16:06 |
| aignatov | and "cluster tuning" is another big topic | 16:07 |
| openstackgerrit | Alexander Ignatov proposed a change to openstack/sahara: Updated validation section for Vanilla Plugin https://review.openstack.org/86338 | 16:20 |
| *** IlyaE has quit IRC | 16:42 | |
| tellesnobrega | tmckay_bbl: hi, are you there? | 16:50 |
| openstackgerrit | A change was merged to openstack/sahara: Fix up Sahara UI installation guide https://review.openstack.org/85694 | 16:51 |
| *** Hao has quit IRC | 16:56 | |
| tellesnobrega | aignatov: hey, im using a localrc file that tmckay send to me, to start sahara, im getting an error about em1, i found in the localrc file that it is a flat_network, is there any automated whay that openstack creates this network adapter and configures it as bridge, or do i need to it by hand? | 17:08 |
| *** IlyaE has joined #openstack-sahara | 17:31 | |
| *** IlyaE has quit IRC | 17:33 | |
| *** bradd1 has quit IRC | 17:44 | |
| *** IlyaE has joined #openstack-sahara | 17:52 | |
| *** bradd1 has joined #openstack-sahara | 18:00 | |
| *** bradd1 has quit IRC | 18:07 | |
| *** tosky has quit IRC | 18:17 | |
| *** mattf is now known as _mattf | 18:19 | |
| *** _mattf is now known as mattf | 18:23 | |
| *** witlessb has quit IRC | 18:28 | |
| tmckay_bbl | tellesnobrega, I'm here now | 18:39 |
| *** tmckay_bbl is now known as tmckay | 18:39 | |
| tellesnobrega | tmckay: hi, i saw that you were using a em1 as FLAT_NETWORK, can i use eth0? or does it have anything special configured? | 18:39 |
| tmckay | tellesnobrega, the only thing I know of FLAT_NETWORK is that I believe it is supposed to be set to the interface which is your link to the outside world | 18:40 |
| tmckay | so, yes, eth0 should work | 18:40 |
| tellesnobrega | tmckay: thanks | 18:40 |
| tmckay | np | 18:40 |
| tellesnobrega | tmckay: for some reason there is a problem with swift, i couldnt get it working with swift. some problem between glance and swift | 18:42 |
| tmckay | tellesnobrega, that's unfortunate. No special config is necessary for glance/swift as far as I know, beyond the service enables in the localrc | 18:45 |
| tellesnobrega | tmckay: :/ | 18:45 |
| tellesnobrega | :( | 18:45 |
| tellesnobrega | tmckay: i will try and figure it out | 18:45 |
| tellesnobrega | see if i can run without swift | 18:45 |
| tmckay | sorry I don't have an answer | 18:45 |
| tellesnobrega | tmckay: np | 18:46 |
| openstackgerrit | Sergey Reshetnyak proposed a change to openstack/sahara-image-elements: Fix install Oozie for vanilla 2 plugin https://review.openstack.org/86294 | 19:13 |
| openstackgerrit | Dmitry Mescheryakov proposed a change to openstack/sahara: Fix up DevStack guide https://review.openstack.org/86426 | 19:14 |
| elmiko | dmitryme: i thought that devstack page was kind of odd too | 19:23 |
| *** bradd1 has joined #openstack-sahara | 19:28 | |
| ruhe | elmiko: i wrote that guide long time ago, mostly for myself :) | 19:28 |
| elmiko | ruhe: it made sense, it just had an odd link from the other page. something about OSX install, but the page was named devstack.rst. not a big deal :) | 19:30 |
| *** akuznetsov has quit IRC | 19:31 | |
| *** akuznetsov has joined #openstack-sahara | 19:32 | |
| *** akuznetsov has quit IRC | 19:33 | |
| *** openstackgerrit has quit IRC | 19:34 | |
| dmitryme | guys, I'd appreciate if you review the change today. As far as I understand, tomorrow is the last day to push changes into Icehouse rc2, and we want to make only absolutely necessary changes after that | 19:34 |
| ruhe | dmitryme: is it the last day for docs changes too? | 19:37 |
| crobertsrh | Question....are our pep8 rules just a bit more relaxed than most projects (at least horizon)? I'm working on merging our dashboard to horizon and wind-up with many more pep8 issues. Lots of "Import modules only" | 19:37 |
| tellesnobrega | tmckay: i got an error flavour disl is too small, i used the ubuntu 13.10 with a xlarge flavor, is this normal? | 19:37 |
| dmitryme | ruhe: right, SergeyLukjanov told that he would prefer not to make rc3 | 19:38 |
| dmitryme | and push rc2 as final release | 19:38 |
| ruhe | crobertsrh: i've got from SergeyLukjanov that there were problems with pep8 in the dashboard code and it could work incorrectly | 19:39 |
| dmitryme | crobertsrh: if you mean our dashboard code, there always was something weird in how pep8 worked in sahara-dashboard | 19:39 |
| crobertsrh | Ah, ok, so it's not just me :) | 19:39 |
| crobertsrh | Not a huge deal really, just a bit more work in the merging process. I'm well into taking care of the issues though. | 19:40 |
| SergeyLukjanov | yup, I'm afraid that it was not working at all for sahara-dashboard | 19:40 |
| SergeyLukjanov | and it was undebuggable *why* it's not working | 19:40 |
| crobertsrh | Undebuggable?? I haven't had problems debugging it in eclipse. | 19:41 |
| dmitryme | tellesnobrega: are you deploying HDP cluster? | 19:41 |
| SergeyLukjanov | crobertsrh, I mean why if we running pep8 in sahara-dashboard then it shows no errors | 19:41 |
| crobertsrh | Oh, right :) | 19:41 |
| tellesnobrega | dmitryme: vanilla | 19:41 |
| SergeyLukjanov | re rc2/rc3, yup, we need at least several days to final check that everything is ok | 19:42 |
| tellesnobrega | dmitryme: im trying with centos image | 19:42 |
| *** openstackgerrit has joined #openstack-sahara | 19:42 | |
| SergeyLukjanov | and rc2 @ Thu / Friday exactly means having Mon-Wed for final testin | 19:42 |
| tmckay | tellesnobrega, hmm, I usually use the "small" flavor. | 19:43 |
| tellesnobrega | tmckay: humm | 19:43 |
| tmckay | I can't imagine that xlarge is too small | 19:43 |
| tellesnobrega | tmckay: i first used tiny and got this message, so i tried xlarge | 19:44 |
| dmitryme | tellesnobrega: what disk size does xlarge have in your OpenStack? | 19:45 |
| tellesnobrega | dmitryme: 8vcpus, 160gb disk, 16gb ram | 19:46 |
| dmitryme | 160GB for root disk or for ephemeral one? | 19:48 |
| tellesnobrega | root, 0 for ephemeral | 19:48 |
| dmitryme | hmmm | 19:48 |
| dmitryme | strange | 19:48 |
| dmitryme | just checked our CI DevStack. It uses flavor of 40GB root and 0 ephemeral. So your xlarge should be enough | 19:49 |
| tellesnobrega | i might have found the problem, i left storage as ephemeral when creating the node group | 19:50 |
| dmitryme | tellesnobrega: I think this should be ok, it should use just root disk in that case | 19:51 |
| tellesnobrega | hmm] | 19:51 |
| dmitryme | besides, as far as I understand, the process stops at instance creation | 19:51 |
| dmitryme | try to just run an instance from that image | 19:51 |
| tellesnobrega | ok | 19:52 |
| *** bradd1 has quit IRC | 19:54 | |
| *** bradd1 has joined #openstack-sahara | 19:55 | |
| tellesnobrega | dmitryme: i will get back to you soon, just had a problem with the cloud here, lost my vm | 19:57 |
| dmitryme | tellesnobrega: I too leave my notebook for half an hour | 19:58 |
| tellesnobrega | dmitryme: np | 19:58 |
| tellesnobrega | dmitryme: thanks for the help | 19:58 |
| *** mattf is now known as _mattf | 20:22 | |
| *** bradd11 has joined #openstack-sahara | 20:22 | |
| *** bradd1 has quit IRC | 20:23 | |
| crobertsrh | ok, down to < 100 line too long pep8 issues. I can almost see the finish line. | 20:28 |
| elmiko | nice! | 20:30 |
| *** _mattf is now known as mattf | 20:43 | |
| *** bradd11 has quit IRC | 20:47 | |
| *** crobertsrh is now known as _crobertsrh | 21:02 | |
| *** elmiko is now known as elmiko_afk | 21:32 | |
| *** tmckay has left #openstack-sahara | 21:47 | |
| *** witlessb has joined #openstack-sahara | 22:14 | |
| *** openstack has joined #openstack-sahara | 22:34 | |
| *** witlessb has quit IRC | 23:25 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!