*** sacharya has joined #savanna | 00:21 | |
*** kui has quit IRC | 00:31 | |
*** sacharya1 has joined #savanna | 00:40 | |
*** sacharya has quit IRC | 00:40 | |
*** nosnos has joined #savanna | 00:43 | |
*** DennyZhang has joined #savanna | 01:04 | |
*** openstack has joined #savanna | 01:34 | |
-hobana.freenode.net- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp | 01:34 | |
*** DennyZhang has quit IRC | 01:47 | |
*** SergeyLukjanov has quit IRC | 01:58 | |
*** IlyaE has quit IRC | 02:15 | |
*** IlyaE has joined #savanna | 02:46 | |
*** IlyaE has quit IRC | 03:16 | |
*** openstackgerrit has quit IRC | 03:16 | |
*** openstackgerrit has joined #savanna | 03:17 | |
*** IlyaE has joined #savanna | 03:46 | |
*** DennyZhang has joined #savanna | 03:48 | |
*** IlyaE has quit IRC | 04:02 | |
*** macjack has quit IRC | 04:05 | |
*** IlyaE has joined #savanna | 04:17 | |
*** IlyaE has quit IRC | 04:20 | |
*** IlyaE has joined #savanna | 04:33 | |
*** IlyaE has quit IRC | 04:34 | |
*** sacharya1 has quit IRC | 05:00 | |
*** akuznetsov has joined #savanna | 05:13 | |
*** akuznetsov has quit IRC | 05:17 | |
*** DennyZhang has quit IRC | 05:17 | |
*** akuznetsov has joined #savanna | 05:20 | |
*** nadya has joined #savanna | 05:37 | |
openstackgerrit | A change was merged to openstack/savanna: Changed Integration tests https://review.openstack.org/53121 | 05:41 |
---|---|---|
*** nadya has quit IRC | 05:50 | |
*** akuznetsov has quit IRC | 06:02 | |
*** SergeyLukjanov has joined #savanna | 06:06 | |
*** nadya has joined #savanna | 06:15 | |
*** akuznetsov has joined #savanna | 06:17 | |
*** nadya has quit IRC | 06:19 | |
*** IlyaE has joined #savanna | 06:22 | |
*** nadya has joined #savanna | 06:30 | |
*** macjack has joined #savanna | 06:32 | |
*** nadya has quit IRC | 06:42 | |
*** akuznetsov has quit IRC | 06:47 | |
*** akuznetsov has joined #savanna | 06:48 | |
*** IlyaE has quit IRC | 06:51 | |
*** akuznetsov has quit IRC | 06:52 | |
*** akuznetsov has joined #savanna | 07:51 | |
*** SergeyLukjanov has quit IRC | 08:02 | |
*** xianghui has joined #savanna | 08:17 | |
*** nprivalova has joined #savanna | 08:36 | |
openstackgerrit | A change was merged to openstack/savanna: Include Vanilla Plugin *.sql files https://review.openstack.org/53162 | 08:44 |
*** dmitryme has joined #savanna | 08:48 | |
*** bogdando has quit IRC | 09:01 | |
*** bogdando has joined #savanna | 09:06 | |
openstackgerrit | Nikita Konovalov proposed a change to openstack/savanna-dashboard: Fixing importutils https://review.openstack.org/53298 | 09:09 |
*** nosnos has quit IRC | 09:12 | |
*** nosnos has joined #savanna | 09:13 | |
*** matsuhashi has quit IRC | 09:13 | |
*** matsuhashi has joined #savanna | 09:14 | |
openstackgerrit | Dmitry Mescheryakov proposed a change to openstack/savanna: Make 'ls' check threaded https://review.openstack.org/53300 | 09:15 |
*** nosnos has quit IRC | 09:17 | |
*** matsuhashi has quit IRC | 09:19 | |
*** xianghui has quit IRC | 09:34 | |
*** ruhe has joined #savanna | 10:11 | |
*** nprivalova has quit IRC | 10:25 | |
*** ruhe has quit IRC | 10:27 | |
*** ruhe has joined #savanna | 11:06 | |
*** nprivalova has joined #savanna | 11:12 | |
*** nprivalova has quit IRC | 11:28 | |
*** dmitryme has quit IRC | 11:32 | |
*** nprivalova has joined #savanna | 11:35 | |
*** xianghui has joined #savanna | 11:36 | |
*** _crobertsrh is now known as crobertsrh | 11:38 | |
openstackgerrit | Yaroslav Lobankov proposed a change to openstack/savanna: Added a check for Oozie configs https://review.openstack.org/53329 | 12:02 |
*** dmitryme has joined #savanna | 12:07 | |
*** ruhe has quit IRC | 12:25 | |
*** nprivalova has quit IRC | 12:31 | |
*** dmitryme has quit IRC | 12:31 | |
openstackgerrit | Matthew Farrellee proposed a change to openstack/savanna: Include Vanilla Plugin *.sql files https://review.openstack.org/53338 | 12:34 |
*** nprivalova has joined #savanna | 12:37 | |
*** kui has quit IRC | 12:39 | |
*** akuznetsov has quit IRC | 12:39 | |
*** akuznetsov has joined #savanna | 12:43 | |
*** mattymo has joined #savanna | 12:45 | |
*** akuznetsov has quit IRC | 12:46 | |
*** ruhe has joined #savanna | 12:47 | |
*** ruhe has quit IRC | 12:47 | |
*** akuznetsov has joined #savanna | 12:49 | |
*** nprivalova has quit IRC | 12:52 | |
*** ruhe has joined #savanna | 13:00 | |
*** ruhe has quit IRC | 13:04 | |
*** tmckay has quit IRC | 13:14 | |
*** sacharya has joined #savanna | 13:41 | |
*** sacharya has quit IRC | 13:48 | |
*** akuznetsov has quit IRC | 13:50 | |
*** akuznetsov has joined #savanna | 13:53 | |
*** nprivalova has joined #savanna | 13:54 | |
*** dmitryme has joined #savanna | 13:56 | |
*** IlyaE has joined #savanna | 13:58 | |
openstackgerrit | Yaroslav Lobankov proposed a change to openstack/savanna: Integration test for Swift has changes https://review.openstack.org/53357 | 14:04 |
*** tmckay has joined #savanna | 14:08 | |
*** nprivalova has quit IRC | 14:10 | |
*** crobertsrh has quit IRC | 14:28 | |
*** sacharya has joined #savanna | 14:29 | |
*** crobertsrh has joined #savanna | 14:31 | |
*** xianghui has quit IRC | 14:34 | |
*** sreshetnyak has quit IRC | 14:40 | |
*** akuznetsov has quit IRC | 14:52 | |
*** macjack has quit IRC | 14:52 | |
*** nprivalova has joined #savanna | 14:57 | |
*** dmitryme has quit IRC | 15:09 | |
*** rnirmal has joined #savanna | 15:16 | |
*** akuznetsov has joined #savanna | 15:27 | |
*** akuznetsov has quit IRC | 15:28 | |
*** mattymo has quit IRC | 15:31 | |
*** mattymo has joined #savanna | 15:32 | |
*** akuznetsov has joined #savanna | 15:33 | |
*** SergeyLukjanov has joined #savanna | 15:51 | |
openstackgerrit | Yaroslav Lobankov proposed a change to openstack/savanna: Added a check for Oozie configs https://review.openstack.org/53329 | 15:59 |
*** DennyZhang has joined #savanna | 16:07 | |
*** isviridov is now known as d | 16:11 | |
*** d is now known as do | 16:11 | |
*** do is now known as tests | 16:11 | |
*** tests is now known as do | 16:11 | |
*** IlyaE has quit IRC | 16:18 | |
*** mattf is now known as _mattf | 16:37 | |
*** IlyaE has joined #savanna | 16:44 | |
*** _mattf is now known as mattf | 16:44 | |
*** IlyaE has quit IRC | 16:57 | |
*** DennyZhang has quit IRC | 17:02 | |
*** akuznetsov has quit IRC | 17:11 | |
*** SergeyLukjanov has quit IRC | 17:21 | |
mattf | aignatov, what's the word on https://bugs.launchpad.net/savanna/+bug/1240511 ? | 17:38 |
mattf | we had to abort using my install for a demo today. we fell back to a devstack crobertsrh has working from weeks ago. | 17:38 |
*** SergeyLukjanov has joined #savanna | 17:40 | |
*** IlyaE has joined #savanna | 17:54 | |
*** mattf is now known as _mattf | 18:05 | |
*** _mattf is now known as mattf | 18:06 | |
*** SergeyLukjanov has quit IRC | 18:07 | |
*** mattymo has quit IRC | 18:11 | |
*** SergeyLukjanov has joined #savanna | 18:13 | |
mattf | crobertsrh, i'm only seeing connection refused, not classnotfound | 18:17 |
mattf | Launcher ERROR, reason: Main class [org.apache.oozie.action.hadoop.MapReduceMain], main() threw exception, POST http://127.0.0.1:35357/v2.0/tokens/ failed on exception: java.net.ConnectException: Connection refused; For more details see: http://wiki.apache.org/hadoop/ConnectionRefused | 18:17 |
crobertsrh | Yes, that might be from another job I just tried running. | 18:17 |
crobertsrh | My jar job throws that. Pig job was CNF | 18:18 |
mattf | you know what, that exception is coming from the swift plugin | 18:20 |
mattf | at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) | 18:20 |
mattf | at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAcces | 18:20 |
mattf | sorImpl.java:57) | 18:20 |
mattf | at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstruc | 18:20 |
mattf | torAccessorImpl.java:45) | 18:20 |
mattf | at java.lang.reflect.Constructor.newInstance(Constructor.java:526) | 18:20 |
mattf | at org.apache.hadoop.fs.swift.http.ExceptionDiags.wrapWithMessage(ExceptionDiag | 18:20 |
mattf | s.java:90) | 18:20 |
mattf | it's trying to reach localhost:35357 (35357 is keystone) | 18:23 |
mattf | keystone is not running on the instance | 18:23 |
crobertsrh | :q | 18:23 |
* mattf scratches head | 18:24 | |
mattf | crobertsrh, my guess is the pig classnotfound error is a race between oozie starting and hdfs starting (justification is the replication error) | 18:26 |
mattf | so i got lucky w/ this cluster starting up | 18:26 |
mattf | so yeah, a little progress. | 18:26 |
crobertsrh | could be. I did try manually restarting oozie at one point yesterday....didn't seem to have an effect. | 18:27 |
mattf | nprivalova, you did a youtube video of EDP, did you use devstack for it? | 18:27 |
mattf | SergeyLukjanov, who was it that made the EDP youtube video? | 18:31 |
mattf | crobertsrh, want to shutdown my cluster and spin one up w/ the mirantis image you downloaded? | 18:38 |
crobertsrh | I have already spun up a cluster with the mirantis image...seems to be enough room for both. | 18:38 |
mattf | way ahead of me | 18:38 |
crobertsrh | Job run on that cluster gives the NoClassDefFound stack | 18:39 |
mattf | err, they're both mirantis images | 18:39 |
* mattf goofed | 18:39 | |
crobertsrh | Doh...I suppose I should have checked before I spun another | 18:39 |
crobertsrh | but we're getting 2 different errors with the same image | 18:40 |
mattf | will you set the root pw on your cluster for me? | 18:40 |
crobertsrh | will do | 18:41 |
crobertsrh | done | 18:41 |
mattf | tkx | 18:41 |
tmckay | mattf, on the EDP youtube, that was aignatov I believe | 18:43 |
mattf | 2013-10-23 18:32:29,995 ERROR DataNucleus.Plugin (Log4JLogger.java:error(115)) - Bundle "org.eclipse.jdt.core" requires "org.eclipse.text" but it cannot be resolved. | 18:44 |
mattf | ^^ in /etc/hadoop/hive.log | 18:44 |
mattf | so fs.swift.service.savanna.auth.url is providing the bad url to keystone, it's in /etc/hadoop/core-site.xml (expected) | 18:48 |
crobertsrh | Anything in the RDO packaging that sets that value? On my devstack install, that value comes out correctly. | 18:52 |
mattf | not that i can see. the instance can access keystone on the host env, checked w/ curl http://10.16.44.108:35357/v2.0/tokens | 18:54 |
mattf | the swift_helper is making the decision to use localhost | 18:54 |
crobertsrh | mattf: looks like os_auth_host is defaulted to 127.0.0.1 in savanna.conf | 18:54 |
*** nadya has joined #savanna | 18:54 | |
mattf | maybe misconfig on my part | 18:55 |
* mattf frowns | 18:55 | |
mattf | i'm not very happy about that value being burned into the instance | 18:56 |
mattf | it's a legit value for the services running on the system | 18:56 |
mattf | i know some folks like lots of config steps, but i'm trying to drive the manual (mishaps) out | 18:57 |
crobertsrh | noble cause | 18:59 |
mattf | sed -i -e "s/#os_auth_host.*/os_auth_host=$(host $(hostname) | cut -f4 -d\ )/" /etc/savanna/savanna.conf | 19:00 |
crobertsrh | bounce and refire cluster? | 19:01 |
mattf | it's a pita because keystonerc only has OS_AUTH_*URL* | 19:01 |
mattf | crobertsrh, please do, leave my cluster tho | 19:01 |
crobertsrh | will do | 19:01 |
mattf | i'm going to manually change c0 | 19:01 |
crobertsrh | Did you bounce savanna? | 19:01 |
mattf | yes, just now | 19:02 |
crobertsrh | ok, thanks | 19:02 |
mattf | crobertsrh, fyi, launching a job does not bring you to the "job executions" tab, disconnect from "launch cluster" behavior | 19:06 |
openstackgerrit | Jonathan Maron proposed a change to openstack/savanna: Enable network operations over neutron private nets https://review.openstack.org/52997 | 19:06 |
crobertsrh | mattf: Yes...I already submitted a bug for that one | 19:06 |
crobertsrh | simple fix...wish I would have noticed before release | 19:06 |
crobertsrh | same CNF on new cluster. | 19:08 |
* mattf wonders how pandora got on "just like honey" | 19:10 | |
mattf | certainly is psychocandy | 19:10 |
* mattf shakes fist hard as he discovers selinux is in permissive mode | 19:11 | |
mattf | org.apache.oozie.command.CommandException: E0604: Job does not exist [None] | 19:13 |
mattf | ^^ new one | 19:13 |
crobertsrh | I've seen that one sporadically over the past couple of days | 19:14 |
mattf | i also have a job that's apparently stuck pending | 19:15 |
mattf | crobertsrh, you get a new cluster up yet? | 19:19 |
mattf | a non-CNF cluster | 19:19 |
crobertsrh | No cluster without CNF. | 19:19 |
crobertsrh | New cluster though, yes. core-site.xml is correct this time. | 19:19 |
mattf | i can't rule out that i messed something up when i hand edited | 19:25 |
mattf | all i know is the error changed | 19:25 |
crobertsrh | Yeah, no arguing that. | 19:25 |
crobertsrh | There is a note on the bug for oozie not starting that mentions that we ought to be able to do "oozie sharelib" multiple times. | 19:27 |
crobertsrh | sharelib comes up a lot when I google around about the CNF exception. I wonder if something went wacky (race?) when that is being installed. | 19:28 |
crobertsrh | I might just be lucking out on my slow machine and avoiding any races. | 19:28 |
mattf | i'm going to take down c0 | 19:30 |
crobertsrh | ok | 19:31 |
*** IlyaE has quit IRC | 19:32 | |
mattf | crobertsrh, i keep having thoughts that jobs should be job templates | 19:35 |
* mattf stops | 19:35 | |
crobertsrh | I think I know what you mean. I'd like to be able to quickly rerun a job with a set config, but change the data sources. | 19:35 |
crobertsrh | I have a bug for that too. | 19:36 |
crobertsrh | I like the "job template" term though. I might add that. | 19:36 |
* mattf fights symmetryyrtemmys | 19:37 | |
mattf | so, just like only i can build sesame | 19:37 |
mattf | apparently only i can start a cluster | 19:37 |
mattf | my last job succeeded | 19:38 |
crobertsrh | wow | 19:38 |
mattf | (course, i can't get to the coozie console) | 19:38 |
mattf | try running against c1 | 19:38 |
crobertsrh | I can get there...yep...it succeeded | 19:38 |
crobertsrh | ok | 19:38 |
mattf | just lagging for me | 19:40 |
crobertsrh | my job worked too | 19:41 |
mattf | ok, so back to one bug | 19:42 |
crobertsrh | Did you do anything different on c1? | 19:42 |
mattf | i'm making mental note of the config change and secgroup change | 19:42 |
mattf | well, for c1 i ran my disk image | 19:42 |
mattf | i could try again w/ the mirantis one, but i'm 2-0 atm! | 19:43 |
crobertsrh | heh | 19:43 |
crobertsrh | I'll try a fresh cluster and see what I get. | 19:43 |
* mattf considers quitting while he's ahead | 19:43 | |
*** IlyaE has joined #savanna | 19:47 | |
mattf | it's awkward that i have to create a source for the destination | 19:51 |
mattf | esp since the actual dest is a container/dir and the dest i make gets touched and nothing else | 19:52 |
crobertsrh | Yes, I agree. | 19:52 |
crobertsrh | Ok, jobs are also now succeeding on my new cluster | 19:55 |
crobertsrh | what have we done differently? | 19:55 |
mattf | we had pathologically bad config before | 19:56 |
mattf | it manifested as a connection refused | 19:56 |
mattf | we still have the non-deterministic CNF bug imho | 19:56 |
crobertsrh | Hard to say. I wonder if I'll ever see it again. | 19:57 |
mattf | even odds you will if you run 3 more clusters | 19:57 |
crobertsrh | Clusters are cheap. I'll give it a try. | 19:58 |
*** nadya has quit IRC | 19:58 | |
*** rnirmal_ has joined #savanna | 20:02 | |
*** rnirmal_ has quit IRC | 20:03 | |
*** rnirmal_ has joined #savanna | 20:03 | |
*** rnirmal has quit IRC | 20:04 | |
*** rnirmal_ is now known as rnirmal | 20:04 | |
*** do is now known as isviridov | 20:05 | |
*** SergeyLukjanov is now known as _SergeyLukjanov | 20:08 | |
*** _SergeyLukjanov has quit IRC | 20:09 | |
crobertsrh | Yep, hit the CNF again. | 20:09 |
*** sacharya has quit IRC | 20:12 | |
*** SergeyLukjanov has joined #savanna | 20:12 | |
*** sacharya has joined #savanna | 20:19 | |
mattf | crobertsrh, how many did it take? | 20:22 |
crobertsrh | 3rd one failed. | 20:23 |
mattf | crobertsrh, what secgroup rules do you think are necessary? | 20:26 |
mattf | i see you added one for all ports | 20:26 |
crobertsrh | Yeah, that was a shot in the dark. | 20:26 |
mattf | what the minimum set? | 20:26 |
crobertsrh | I think the others there are necessary except for the last 2 that I added. | 20:27 |
mattf | ok | 20:27 |
crobertsrh | 11000 is probably not totally mandatory (only for oozie web ui) | 20:27 |
crobertsrh | I think 11443 was needed to be able to submit to oozie. | 20:28 |
mattf | mind reducing the set and verifying? | 20:28 |
crobertsrh | sure | 20:28 |
crobertsrh | Ok, looks like 11000 is sufficient | 20:34 |
mattf | great | 20:35 |
*** kui has joined #savanna | 20:36 | |
*** crobertsrh is now known as _crobertsrh | 20:38 | |
openstackgerrit | John Speidel proposed a change to openstack/savanna: Add support for Hive related services https://review.openstack.org/52923 | 20:51 |
tmckay | cool, I got a test wrapper extracted from Ambari to run with the javasqlite rpm sources instead of sqlite-jdbc from xerial. Replaced the driver class loaded from SQLiteConnectionFactory. | 21:01 |
tmckay | of course, I need a real test but this is a start :) | 21:01 |
tmckay | my goodness, I can't keep track of my windows | 21:02 |
* mattf nods | 21:03 | |
*** tmckay has quit IRC | 21:12 | |
*** IlyaE has quit IRC | 21:16 | |
*** kushi_ has joined #savanna | 21:21 | |
*** kui has quit IRC | 21:22 | |
*** ChanServ has joined #savanna | 21:32 | |
*** services. sets mode: +o ChanServ | 21:32 | |
*** IlyaE has joined #savanna | 21:33 | |
*** mattf is now known as _mattf | 21:44 | |
*** SergeyLukjanov is now known as _SergeyLukjanov | 22:02 | |
*** _SergeyLukjanov has quit IRC | 22:02 | |
*** sacharya has quit IRC | 22:15 | |
*** SergeyLukjanov has joined #savanna | 22:41 | |
*** rnirmal has quit IRC | 22:43 | |
*** openstack has joined #savanna | 22:47 | |
*** kushi_ has quit IRC | 22:55 | |
*** sacharya has joined #savanna | 23:27 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!