| *** clopez has quit IRC | 00:00 | |
| *** rkukura has joined #openstack | 00:05 | |
| *** jdg has quit IRC | 00:05 | |
| *** bsza has joined #openstack | 00:06 | |
| *** Blackavar has quit IRC | 00:06 | |
| *** heckj has joined #openstack | 00:07 | |
| *** llang629 has joined #openstack | 00:07 | |
| *** ahasenack has quit IRC | 00:07 | |
| *** dachary has quit IRC | 00:07 | |
| *** heckj has quit IRC | 00:07 | |
| *** dwcramer has joined #openstack | 00:09 | |
| *** MarkAtwood has joined #openstack | 00:09 | |
| *** gyee has quit IRC | 00:10 | |
| *** vitiho has quit IRC | 00:10 | |
| *** albert23 has left #openstack | 00:11 | |
| *** ayoung has joined #openstack | 00:14 | |
| *** dwcramer has quit IRC | 00:15 | |
| *** rafaduran has quit IRC | 00:16 | |
| *** bsza has quit IRC | 00:18 | |
| *** LarsErikP has quit IRC | 00:19 | |
| *** clopez has joined #openstack | 00:19 | |
| *** mrjazzcat has quit IRC | 00:20 | |
| *** bhall has quit IRC | 00:22 | |
| Crshman | do security groups work when using a flat network config? | 00:23 |
|---|---|---|
| *** warik has joined #openstack | 00:25 | |
| *** maplebed has quit IRC | 00:27 | |
| *** vitiho has joined #openstack | 00:28 | |
| *** dwcramer has joined #openstack | 00:28 | |
| *** joearnol_ has quit IRC | 00:28 | |
| *** spiffxp has quit IRC | 00:30 | |
| *** straylyon has quit IRC | 00:30 | |
| *** krow1 has quit IRC | 00:30 | |
| *** Guest23981 has quit IRC | 00:32 | |
| *** nelson1234 has quit IRC | 00:34 | |
| *** localhost has quit IRC | 00:35 | |
| *** nelson1234 has joined #openstack | 00:35 | |
| *** localhost has joined #openstack | 00:36 | |
| *** vitiho_ has joined #openstack | 00:40 | |
| *** vitiho has quit IRC | 00:40 | |
| *** tmichael has quit IRC | 00:49 | |
| *** wariola has quit IRC | 00:49 | |
| *** vizsla has joined #openstack | 00:49 | |
| *** troytoman-away is now known as troytoman | 00:50 | |
| *** alrs has quit IRC | 00:50 | |
| *** johnpostlethwait has quit IRC | 00:52 | |
| *** johnpostlethwait has joined #openstack | 00:53 | |
| agonella | do you know if openstack has something like vcenter server from vmware? | 00:54 |
| agonella | a management tool able to clone and snapshot virtual machines. | 00:54 |
| Spirilis | agonella: the "horizon" web GUI can do that. | 00:57 |
| *** miclorb__ has joined #openstack | 00:57 | |
| *** vincentricci has quit IRC | 00:59 | |
| agonella | has it a link? | 01:00 |
| agonella | to read about it | 01:00 |
| *** miclorb_ has quit IRC | 01:01 | |
| agonella | got it | 01:01 |
| *** miclorb__ has quit IRC | 01:02 | |
| *** pixelbeat has quit IRC | 01:02 | |
| *** jakkudanieru has joined #openstack | 01:04 | |
| *** livemoon has joined #openstack | 01:04 | |
| *** whenry has joined #openstack | 01:06 | |
| *** rnorwood has joined #openstack | 01:07 | |
| *** warik has left #openstack | 01:07 | |
| *** FACEFOX has quit IRC | 01:13 | |
| *** vincentricci has joined #openstack | 01:14 | |
| *** vizsla has quit IRC | 01:14 | |
| *** llang629 has left #openstack | 01:14 | |
| *** hub_cap has joined #openstack | 01:17 | |
| *** vincentricci has quit IRC | 01:19 | |
| *** adam__ has joined #openstack | 01:22 | |
| *** krow has joined #openstack | 01:22 | |
| *** shaon has quit IRC | 01:24 | |
| *** aspiers has quit IRC | 01:25 | |
| *** shevek_ has quit IRC | 01:27 | |
| *** ywu has joined #openstack | 01:27 | |
| *** hermy|away is now known as hermy | 01:31 | |
| *** davlaps has quit IRC | 01:31 | |
| *** dendro-afk is now known as dendrobates | 01:32 | |
| *** leifmadsen has quit IRC | 01:32 | |
| *** rclewis has quit IRC | 01:33 | |
| *** johnpostlethwait has quit IRC | 01:35 | |
| *** halfss has joined #openstack | 01:38 | |
| halfss | hi,is there anyone use swift as store of glance? | 01:39 |
| *** natea|dinner is now known as natea | 01:40 | |
| *** agonella has quit IRC | 01:42 | |
| *** Ryan_Lane has quit IRC | 01:45 | |
| *** vizsla has joined #openstack | 01:46 | |
| *** linmin has joined #openstack | 01:47 | |
| *** halfss_ has joined #openstack | 01:50 | |
| *** halfss has quit IRC | 01:54 | |
| *** egant has joined #openstack | 01:54 | |
| *** FACEFOX has joined #openstack | 01:59 | |
| *** machine2 has joined #openstack | 02:00 | |
| *** gohko_air has quit IRC | 02:00 | |
| *** jj0hns0n has joined #openstack | 02:02 | |
| *** mnewby has quit IRC | 02:02 | |
| *** gohko_air has joined #openstack | 02:03 | |
| *** jj0hns0n has quit IRC | 02:04 | |
| *** egant has quit IRC | 02:06 | |
| *** flaviamissi has joined #openstack | 02:06 | |
| *** sandywalsh has quit IRC | 02:07 | |
| *** dpippenger has quit IRC | 02:08 | |
| *** anotherjesse_zz is now known as anotherjesse | 02:09 | |
| *** adam__ is now known as aspiers | 02:09 | |
| *** flaviamissi has quit IRC | 02:11 | |
| *** flaviamissi has joined #openstack | 02:12 | |
| *** mdrnstm has quit IRC | 02:12 | |
| *** agonella has joined #openstack | 02:12 | |
| *** dwcramer has quit IRC | 02:14 | |
| *** agonella has quit IRC | 02:14 | |
| *** FACEFOX has quit IRC | 02:15 | |
| *** tmichael has joined #openstack | 02:16 | |
| *** jdurgin has quit IRC | 02:18 | |
| *** machine2 has quit IRC | 02:22 | |
| *** erkules has quit IRC | 02:25 | |
| *** erkules has joined #openstack | 02:26 | |
| *** MarkAtwood has quit IRC | 02:26 | |
| *** paulmillar has joined #openstack | 02:26 | |
| *** dwcramer has joined #openstack | 02:27 | |
| *** anotherjesse is now known as anotherjesse_zz | 02:28 | |
| *** miclorb_ has joined #openstack | 02:28 | |
| *** dendrobates is now known as dendro-afk | 02:28 | |
| *** rods has quit IRC | 02:28 | |
| *** halfss__ has joined #openstack | 02:29 | |
| *** issackelly has quit IRC | 02:29 | |
| *** phorgh has quit IRC | 02:29 | |
| *** emorisse has joined #openstack | 02:30 | |
| *** halfss_ has quit IRC | 02:33 | |
| *** reed has quit IRC | 02:33 | |
| *** anotherjesse_zz is now known as anotherjesse | 02:33 | |
| *** zynzel has quit IRC | 02:34 | |
| *** zynzel has joined #openstack | 02:34 | |
| *** mrjazzcat has joined #openstack | 02:37 | |
| *** natea has quit IRC | 02:39 | |
| *** mdomsch has joined #openstack | 02:42 | |
| *** mikal has quit IRC | 02:44 | |
| *** mikal has joined #openstack | 02:48 | |
| *** Blackavar has joined #openstack | 02:50 | |
| *** troytoman is now known as troytoman-away | 02:54 | |
| *** Karmaon has quit IRC | 02:58 | |
| *** sandywalsh has joined #openstack | 02:59 | |
| *** ywu has quit IRC | 03:00 | |
| *** anotherjesse is now known as anotherjesse_zz | 03:03 | |
| *** mdrnstm has joined #openstack | 03:08 | |
| *** paulstallworth has joined #openstack | 03:10 | |
| *** scoph has quit IRC | 03:12 | |
| *** gekko_ has quit IRC | 03:14 | |
| *** Ruetobas has joined #openstack | 03:16 | |
| *** whenry has quit IRC | 03:16 | |
| *** jmckenty has joined #openstack | 03:18 | |
| *** spiffxp has joined #openstack | 03:18 | |
| *** sandywalsh has quit IRC | 03:23 | |
| *** linmin has quit IRC | 03:24 | |
| uvirtbot | New bug: #960823 in horizon "caption use on table header cause layout weirdness" [Undecided,New] https://launchpad.net/bugs/960823 | 03:26 |
| *** pixelbeat has joined #openstack | 03:27 | |
| *** hub_cap has quit IRC | 03:28 | |
| *** miclorb_ has quit IRC | 03:29 | |
| *** pixelbeat has quit IRC | 03:33 | |
| *** miclorb_ has joined #openstack | 03:34 | |
| *** krow has quit IRC | 03:34 | |
| *** supriya has joined #openstack | 03:36 | |
| *** sandywalsh has joined #openstack | 03:36 | |
| *** wilmoore has joined #openstack | 03:36 | |
| *** llang629 has joined #openstack | 03:38 | |
| *** vitiho__ has joined #openstack | 03:41 | |
| *** vitiho_ has quit IRC | 03:42 | |
| *** krow has joined #openstack | 03:44 | |
| *** natea has joined #openstack | 03:49 | |
| *** supriya has quit IRC | 03:51 | |
| *** vizsla has quit IRC | 03:51 | |
| *** mrjazzcat has quit IRC | 03:57 | |
| *** dolphm has quit IRC | 04:01 | |
| *** cafe_ngam_mua has joined #openstack | 04:04 | |
| cafe_ngam_mua | hi | 04:08 |
| cafe_ngam_mua | i'm setting up openstack on centos | 04:08 |
| cafe_ngam_mua | in the glance-api.conf, | 04:08 |
| cafe_ngam_mua | there are some lines for s3 configuration, | 04:09 |
| cafe_ngam_mua | i don't have s3, | 04:09 |
| cafe_ngam_mua | so what can i set at this? | 04:09 |
| cafe_ngam_mua | thanks | 04:10 |
| *** cw has quit IRC | 04:10 | |
| *** kentbye has quit IRC | 04:10 | |
| *** cw has joined #openstack | 04:11 | |
| *** mnewby has joined #openstack | 04:16 | |
| *** dwcramer has quit IRC | 04:19 | |
| *** dolphm has joined #openstack | 04:19 | |
| *** mnewby has quit IRC | 04:20 | |
| *** supriya has joined #openstack | 04:21 | |
| *** mnewby has joined #openstack | 04:26 | |
| *** miclorb_ has quit IRC | 04:27 | |
| *** hermy is now known as hermy|away | 04:28 | |
| *** kayyagari has joined #openstack | 04:30 | |
| dweimer | cafe_ngam_mua: Unless your storage backend is set to use s3 you don't need to set the s3 configuration parameters. Is your 'default_store' entry set to file or swift? | 04:32 |
| *** vitiho__ has quit IRC | 04:32 | |
| *** vitiho has joined #openstack | 04:32 | |
| cafe_ngam_mua | hi dweimer | 04:33 |
| cafe_ngam_mua | i see the log in the glance-api log file | 04:34 |
| *** dolphm has quit IRC | 04:34 | |
| cafe_ngam_mua | ERROR [glance.store.s3] Could not find s3_store_access_key in configuration options. 2012-03-12 09:27:16 ERROR [glance.store.base] Failed to configure store correctly. Disabling add method. | 04:34 |
| cafe_ngam_mua | if it disable add method, we can not add images | 04:35 |
| *** dolphm has joined #openstack | 04:36 | |
| *** cryptk|offline is now known as cryptk | 04:37 | |
| dweimer | cafe_ngam_mua: In your glance-api.conf file is the default_store set? | 04:38 |
| *** DigitalFlux has quit IRC | 04:38 | |
| *** reed has joined #openstack | 04:39 | |
| *** omgub has joined #openstack | 04:39 | |
| *** dolphm has quit IRC | 04:39 | |
| cafe_ngam_mua | hi, i deleted the default config, | 04:39 |
| cafe_ngam_mua | sorry for the dump question | 04:39 |
| cafe_ngam_mua | :) | 04:39 |
| cafe_ngam_mua | it seems ok now | 04:39 |
| cafe_ngam_mua | thanks | 04:39 |
| dweimer | cafe_ngam_mua: Glad it fixed it. I'm working on setting up openstack on CentOS myself. Are you using the EPEL packages? | 04:40 |
| *** llang629 has left #openstack | 04:40 | |
| cafe_ngam_mua | no, i'm using the openstack source file | 04:40 |
| cafe_ngam_mua | i'm using diable 2011.3.1 | 04:41 |
| cafe_ngam_mua | :) | 04:41 |
| cafe_ngam_mua | diablo | 04:41 |
| *** DigitalFlux has joined #openstack | 04:42 | |
| *** DigitalFlux has joined #openstack | 04:42 | |
| dweimer | cafe_ngam_mua: Interesting, are you just installing each of the python module depenencies using pip or easy_install? | 04:43 |
| *** randomubuntuguy has joined #openstack | 04:44 | |
| *** stewart has joined #openstack | 04:49 | |
| *** miclorb_ has joined #openstack | 04:51 | |
| cafe_ngam_mua | i used yum for some modules, and easy_install for anothers | 04:51 |
| cafe_ngam_mua | :) | 04:51 |
| cafe_ngam_mua | like in the tutorial on openstackwiki | 04:51 |
| *** jmckenty has quit IRC | 04:51 | |
| *** refactor has joined #openstack | 04:52 | |
| cafe_ngam_mua | i'm going for lunch now, | 04:53 |
| cafe_ngam_mua | see u soon | 04:53 |
| cafe_ngam_mua | thanks | 04:53 |
| *** MarkAtwood has joined #openstack | 04:55 | |
| halfss__ | is there anybody user keystone+swift+glance? | 05:02 |
| *** paulstallworth has left #openstack | 05:03 | |
| *** hattwick has quit IRC | 05:03 | |
| *** randomubuntuguy has quit IRC | 05:05 | |
| *** guynaor has joined #openstack | 05:07 | |
| *** guynaor has quit IRC | 05:12 | |
| *** linmin has joined #openstack | 05:17 | |
| halfss__ | is there anybody user keystone+swift+glance? | 05:17 |
| uvirtbot | New bug: #960864 in tempest "Testcase: Write Testcases for the action list Volumes and list Volumes with Detail" [Undecided,New] https://launchpad.net/bugs/960864 | 05:21 |
| *** guynaor has joined #openstack | 05:28 | |
| *** miclorb_ has quit IRC | 05:30 | |
| *** wilmoore has quit IRC | 05:30 | |
| *** miclorb_ has joined #openstack | 05:30 | |
| *** guynaor has quit IRC | 05:30 | |
| uvirtbot | New bug: #960866 in horizon "row actions affects whole table" [Undecided,New] https://launchpad.net/bugs/960866 | 05:31 |
| *** miclorb_ has quit IRC | 05:32 | |
| *** deshantm has quit IRC | 05:36 | |
| *** Karmaon has joined #openstack | 05:37 | |
| uvirtbot | New bug: #960870 in openobject-addons/trunk "[Fetchmail]: UnboundLocalError: local variable 'imap_server' referenced before assignment" [Undecided,New] https://launchpad.net/bugs/960870 | 05:46 |
| *** flaviamissi has quit IRC | 05:51 | |
| *** rnorwood has quit IRC | 05:52 | |
| *** reed has quit IRC | 05:59 | |
| *** MarkAtwood has quit IRC | 06:00 | |
| *** MarkAtwood has joined #openstack | 06:00 | |
| *** tmichael has quit IRC | 06:04 | |
| *** natea has quit IRC | 06:07 | |
| *** GlenBaars has joined #openstack | 06:09 | |
| uvirtbot | New bug: #960884 in nova "os-hosts extension accepts invalid host status" [Undecided,New] https://launchpad.net/bugs/960884 | 06:10 |
| *** Me2 has joined #openstack | 06:11 | |
| Me2 | SUP | 06:11 |
| *** wilmoore has joined #openstack | 06:14 | |
| *** kayyagari has quit IRC | 06:15 | |
| *** Scottt has joined #openstack | 06:16 | |
| *** Daniel has joined #openstack | 06:16 | |
| *** Daniel is now known as XuchieX | 06:17 | |
| *** XuchieX has quit IRC | 06:18 | |
| *** Daniel has joined #openstack | 06:18 | |
| *** linmin has quit IRC | 06:19 | |
| *** Me2 has quit IRC | 06:20 | |
| *** Me2 has joined #openstack | 06:21 | |
| *** Daniel has left #openstack | 06:21 | |
| *** Me2 has quit IRC | 06:21 | |
| *** Me2 has joined #openstack | 06:21 | |
| *** Me2 has quit IRC | 06:22 | |
| *** supriya has quit IRC | 06:22 | |
| *** miclorb_ has joined #openstack | 06:26 | |
| *** clopez has quit IRC | 06:26 | |
| *** Person123 has joined #openstack | 06:28 | |
| *** Person123 has quit IRC | 06:29 | |
| *** cafe_ngam_mua has quit IRC | 06:31 | |
| *** Person123 has joined #openstack | 06:32 | |
| halfss__ | #chmouel hi | 06:32 |
| *** juergh has joined #openstack | 06:32 | |
| *** bepernoot has joined #openstack | 06:33 | |
| *** Ash has quit IRC | 06:35 | |
| *** koolhead17 has quit IRC | 06:35 | |
| *** Person123 has quit IRC | 06:37 | |
| *** Person123 has joined #openstack | 06:37 | |
| *** Daniel has joined #openstack | 06:45 | |
| *** Daniel is now known as xuchiex | 06:46 | |
| *** Person123 has quit IRC | 06:47 | |
| *** Scottt is now known as DerpyDerp | 06:47 | |
| *** warik has joined #openstack | 06:54 | |
| *** guynaor has joined #openstack | 06:54 | |
| *** miclorb_ has quit IRC | 06:55 | |
| *** guynaor has quit IRC | 06:55 | |
| *** bepernoot has quit IRC | 06:55 | |
| *** guynaor has joined #openstack | 06:56 | |
| *** guynaor has left #openstack | 06:57 | |
| *** krow has quit IRC | 07:00 | |
| *** linmin has joined #openstack | 07:02 | |
| *** xuchiex has quit IRC | 07:02 | |
| *** Daniel has joined #openstack | 07:02 | |
| *** almaisan-away is now known as al-maisan | 07:03 | |
| *** superjudge has joined #openstack | 07:07 | |
| *** longhudou has joined #openstack | 07:09 | |
| *** mnewby has quit IRC | 07:10 | |
| *** gekko_ has joined #openstack | 07:10 | |
| *** dachary has joined #openstack | 07:10 | |
| *** Grimdin has joined #openstack | 07:12 | |
| *** guynaor has joined #openstack | 07:14 | |
| *** mindpixel has joined #openstack | 07:16 | |
| *** Daniel is now known as RawR | 07:17 | |
| *** Daniel has joined #openstack | 07:19 | |
| *** Daniel has quit IRC | 07:20 | |
| *** uchie has joined #openstack | 07:23 | |
| *** praefect has quit IRC | 07:23 | |
| *** kayyagari has joined #openstack | 07:24 | |
| longhudou | I can not start an instance of m1.small,but can start an instance of m1.tiny | 07:25 |
| longhudou | anyone met that before? | 07:25 |
| *** ghe_ is now known as GheRivero | 07:25 | |
| *** miclorb_ has joined #openstack | 07:29 | |
| halfss__ | check log | 07:30 |
| *** superjudge_ has joined #openstack | 07:32 | |
| longhudou | Error: [Errno 13] Permission denied: '/var/lib/nova/instances/_base | 07:35 |
| *** superjudge has quit IRC | 07:35 | |
| *** superjudge_ is now known as superjudge | 07:35 | |
| longhudou | halfss_, | 07:35 |
| longhudou | not sure how com this | 07:35 |
| *** Madkiss has joined #openstack | 07:36 | |
| *** GheRivero has quit IRC | 07:38 | |
| *** ghe_ has joined #openstack | 07:38 | |
| ghe_ | longhudou, check perms in /var/lib/nova | 07:38 |
| *** LarsErikP has joined #openstack | 07:39 | |
| ghe_ | btw, can someone please remove the ban of GheRivero in this channel and #openstack-dev? Thnx | 07:40 |
| *** dieterdemeyer has joined #openstack | 07:41 | |
| halfss__ | ? | 07:42 |
| halfss__ | is there anybody use keystone+swift+glance? | 07:43 |
| *** livemoon has quit IRC | 07:43 | |
| *** livemoon has joined #openstack | 07:46 | |
| *** wilmoore has quit IRC | 07:49 | |
| *** MarkAtwood has left #openstack | 07:49 | |
| longhudou | ghe_ | 07:50 |
| longhudou | which one do u mean？ | 07:50 |
| *** guigui1 has joined #openstack | 07:51 | |
| longhudou | cacert.pem or crl.pem | 07:51 |
| ghe_ | /var/lib/nova usually have 750 and user nova, group nova, this menas libvirt-qemu (if you are using kvm) can not access files inside /var/lib/nova | 07:52 |
| ghe_ | solutions: chmod /var/lib/nova with more perms, or add libvirt-qemu to group nova | 07:53 |
| longhudou | I just chmod of /var/lib/nova/instances to 777 | 07:54 |
| *** ChanServ sets mode: +o mtaylor | 07:54 | |
| *** mtaylor sets mode: -b GheRivero!*@* | 07:54 | |
| mtaylor | ghe_: done | 07:54 |
| ghe_ | it's better if you chmod /var/lib/nova | 07:54 |
| ghe_ | thnx mtaylor | 07:54 |
| *** ohnoimdead has quit IRC | 07:54 | |
| *** zigo has joined #openstack | 07:54 | |
| ghe_ | longhudou, later dnsmasq will need to access net information too :/ | 07:55 |
| *** davepigott has joined #openstack | 07:55 | |
| longhudou | ok,let me have a try | 07:55 |
| LarsErikP | how do I properly register nova-volume in keystone? i get ServiceCatalogException Invalid service catalog service: volume in horizon | 07:58 |
| *** robix has quit IRC | 07:58 | |
| *** hattwick has joined #openstack | 07:59 | |
| *** bepernoot has joined #openstack | 08:03 | |
| *** littleidea has joined #openstack | 08:03 | |
| *** RawR has quit IRC | 08:04 | |
| *** superjudge has quit IRC | 08:04 | |
| *** eirikbb has quit IRC | 08:05 | |
| GlenBaars | Has anyone figured out how to install xentools on the cloud routers? | 08:08 |
| longhudou | ghe_ | 08:08 |
| longhudou | Error: [Errno 13] Permission denied: '/var/lib/nova/instances/_base/da4b9237bacccdf19c0760cab7aec4a8359010b0.part' | 08:08 |
| *** RAR has joined #openstack | 08:09 | |
| *** NickWingate has joined #openstack | 08:10 | |
| ghe_ | longhudou, try to be more aggresive, chmod 777 /var/lib/nova -R :) | 08:12 |
| ghe_ | and once working, try to narrow the perms as much as possible | 08:13 |
| *** heyho has joined #openstack | 08:14 | |
| longhudou | I think I have already done that | 08:14 |
| longhudou | :( | 08:14 |
| *** livemoon1 has joined #openstack | 08:15 | |
| *** livemoon1 has left #openstack | 08:15 | |
| ghe_ | if you are using kvm, try to add libvirt-qemu user to nova group | 08:15 |
| *** livemoon has quit IRC | 08:15 | |
| LarsErikP | in keystone, when I have multiple compute nodes, running in multi_host, menaning i have nova-api on every compute-node, do I have to register on compute service per node in keystone? | 08:16 |
| longhudou | ghe_, | 08:16 |
| longhudou | also done that | 08:16 |
| longhudou | something more? | 08:17 |
| ghe_ | not really, that should be enough, maybe try to repeat it (chmod) | 08:18 |
| longhudou | ok,its odd | 08:19 |
| *** livemoon has joined #openstack | 08:21 | |
| *** reidrac has joined #openstack | 08:21 | |
| *** milner has quit IRC | 08:22 | |
| *** ArionInger-Kirkp has joined #openstack | 08:22 | |
| *** RAR is now known as nick | 08:24 | |
| *** ArionInger-Kirkp is now known as Nourok | 08:24 | |
| *** nick is now known as Guest3599 | 08:24 | |
| *** Guest3599 is now known as goddamnick | 08:25 | |
| *** nacx has joined #openstack | 08:25 | |
| *** ghe_ is now known as GheRivero | 08:25 | |
| *** hugokuo has joined #openstack | 08:28 | |
| Madkiss | A | 08:31 |
| Madkiss | oops. | 08:31 |
| *** shaon has joined #openstack | 08:32 | |
| *** kayyagari has quit IRC | 08:33 | |
| *** milner has joined #openstack | 08:36 | |
| *** shevek_ has joined #openstack | 08:36 | |
| *** dxd828 has joined #openstack | 08:37 | |
| LarsErikP | Kiall: here? | 08:38 |
| *** goddamnick has quit IRC | 08:38 | |
| *** uksysadmin has joined #openstack | 08:38 | |
| *** wariola has joined #openstack | 08:39 | |
| *** Johm has joined #openstack | 08:39 | |
| *** asavu has joined #openstack | 08:40 | |
| *** Johm has quit IRC | 08:40 | |
| *** darraghb has joined #openstack | 08:41 | |
| *** GAWD has joined #openstack | 08:42 | |
| *** shaon has quit IRC | 08:45 | |
| *** Free_maN has joined #openstack | 08:47 | |
| *** guigui1 has quit IRC | 08:53 | |
| *** ohnoimdead has joined #openstack | 08:54 | |
| Dieterbe | can anyone explain what somebody would mean with "Swift is designed with multi-tenancy in mind, where HDFS has no notion of multi-tenancy" | 08:54 |
| Dieterbe | hdfs can have multiple clients at the same time, or at least that's what i thought, so i must be missing something | 08:54 |
| *** GAWD has quit IRC | 08:55 | |
| *** karbonick has joined #openstack | 08:56 | |
| *** DerpyDerp has quit IRC | 08:57 | |
| *** dachary has quit IRC | 09:00 | |
| uksysadmin | Dieterbe: swift is controlled through user, group and tenancy access... many members can belong to a tenant, members can belong to multiple tenants, and if you're not on the list, you're not coming in. | 09:02 |
| *** oneiroi has joined #openstack | 09:03 | |
| *** oneiroi has quit IRC | 09:03 | |
| *** oneiroi has joined #openstack | 09:03 | |
| Dieterbe | oh so it's something like ACL's | 09:04 |
| Dieterbe | i thought it had something to do with concurrent access | 09:04 |
| *** derekh has joined #openstack | 09:06 | |
| *** arBmind has joined #openstack | 09:06 | |
| *** bbcmicrocomputer has joined #openstack | 09:07 | |
| *** Nourok has quit IRC | 09:07 | |
| *** miclorb_ has quit IRC | 09:08 | |
| uvirtbot | New bug: #960977 in keystone "Can't create user" [Undecided,New] https://launchpad.net/bugs/960977 | 09:11 |
| *** oneiroi has quit IRC | 09:11 | |
| *** oneiroi has joined #openstack | 09:12 | |
| *** guigui3 has joined #openstack | 09:13 | |
| LarsErikP | uksysadmin: hi! got some questions... Horizon gived HTTP 400 malformed request when trying to fetch info from nova.. any clue? | 09:14 |
| *** karbonick has left #openstack | 09:14 | |
| LarsErikP | guess i've defined the service and/or the endpoint wrong... http://paste.openstack.org/show/11622/ I'm running multi-host (nova-api) on all compute-nodes. No compute on the controller (which horizon is running on). 192.168.10.2 is the controller | 09:17 |
| *** dachary has joined #openstack | 09:19 | |
| LarsErikP | nova client also gived Malformed request url (HTTP 400) | 09:28 |
| uksysadmin | LarsErikP: your endpoint URL looks wrong: http://192.168.10.2:8774/v2/1 | 09:30 |
| uksysadmin | you have /v2/1 on the end | 09:30 |
| uksysadmin | should be http://$ENDPOINT:8774/v2/\$(tenant_id)s | 09:30 |
| Dieterbe | uksysadmin: so multi-tenancy is something like ACL's? versus something to do with concurrent access | 09:32 |
| uksysadmin | Dieterbe: multi-tenancy = multi-project and if you are in that project, you have access, if you're not, then you can't get access. so customer1 can't see customer2's data | 09:34 |
| uksysadmin | yes, nothing to do with concurrency | 09:34 |
| Dieterbe | oh but hdfs has a permission system IIRC | 09:36 |
| *** vitiho has quit IRC | 09:36 | |
| *** vitiho has joined #openstack | 09:37 | |
| Dieterbe | so that hides the data as well (but not the metadata) | 09:37 |
| *** guigui3 has quit IRC | 09:38 | |
| *** pixelbeat has joined #openstack | 09:38 | |
| uksysadmin | hdfs is not swift = hdfs is Hadoop File System used by Hadoop. This is Swift. Similar - and yes based around ACLs | 09:39 |
| uksysadmin | you can set ACLs on containers in Swift | 09:39 |
| *** dneary has joined #openstack | 09:40 | |
| *** dneary has joined #openstack | 09:40 | |
| uksysadmin | so customer1 can give access to customer2's containers if need be | 09:40 |
| *** zul has quit IRC | 09:40 | |
| *** scoph has joined #openstack | 09:40 | |
| LarsErikP | uksysadmin: literally? | 09:41 |
| *** zul has joined #openstack | 09:42 | |
| *** thinrhino has joined #openstack | 09:43 | |
| *** guigui1 has joined #openstack | 09:43 | |
| thinrhino | I am a little confused of configuring nova-manage to authenticate with keystone on diablo | 09:44 |
| thinrhino | any pointers? | 09:44 |
| thinrhino | currently nova-manage user create command creates a user on a different database and does not authenticate on keystone | 09:44 |
| *** __fc has joined #openstack | 09:45 | |
| *** __fc is now known as fc__ | 09:45 | |
| *** thinrhino has quit IRC | 09:46 | |
| *** vizsla has joined #openstack | 09:47 | |
| *** livemoon has left #openstack | 09:48 | |
| uksysadmin | LarsErikP: replace $ENDPOINT with 192.168.10.2... see https://github.com/uksysadmin/OpenStackInstaller/blob/essex/keystone-services.sh for a verbose script on creating the keystone entries | 09:50 |
| *** rkukura has quit IRC | 09:51 | |
| *** vizsla has quit IRC | 09:52 | |
| *** thinrhino has joined #openstack | 09:53 | |
| *** rkukura has joined #openstack | 09:53 | |
| LarsErikP | uksysadmin: sorry, I'm in noob mode to day.. the "\$(tenant_id)s" part, what should I replace that with.. i dont't have any tenant_id command or so :s | 09:55 |
| uksysadmin | that's how it is written | 09:56 |
| uksysadmin | its not a command, the string in your endpoint literally looks like this http://192.168.10.2:8774/v2/$(tenant_id)s | 09:57 |
| LarsErikP | ah | 09:57 |
| *** Karmaon has quit IRC | 09:59 | |
| uksysadmin | so when you authenticate, it adds in the required tenant_id which is a generated id string that matches your tenant. | 09:59 |
| *** thinrhino has quit IRC | 09:59 | |
| *** guynaor has left #openstack | 10:00 | |
| *** melmoth has joined #openstack | 10:00 | |
| melmoth | Hola ! is there a way to download all of essex rc1 in one tarball, or shall i get all separate project from the respective revision control system ? | 10:01 |
| uksysadmin | melmoth: seperately | 10:03 |
| melmoth | thanks. | 10:04 |
| *** jakkudanieru has quit IRC | 10:05 | |
| *** livemoon has joined #openstack | 10:07 | |
| LarsErikP | uksysadmin: thanks. Now i got the fampus HTTP 500 server has either erred.. when trying nova list | 10:08 |
| *** jakkudanieru has joined #openstack | 10:08 | |
| *** livemoon has left #openstack | 10:08 | |
| uvirtbot | New bug: #961021 in openstack-ci "First tests after milestone-proposed is cut always fail" [High,Confirmed] https://launchpad.net/bugs/961021 | 10:11 |
| *** refactor has quit IRC | 10:13 | |
| *** taihen has joined #openstack | 10:13 | |
| LarsErikP | uksysadmin: do you have any link on which env to set, for nova-client to work? :s | 10:14 |
| uksysadmin | LarsErikP: https://github.com/uksysadmin/OpenStackInstaller/blob/essex/create-user#L91 | 10:15 |
| *** supriya has joined #openstack | 10:18 | |
| *** tryggvil_ has joined #openstack | 10:21 | |
| *** linmin has quit IRC | 10:22 | |
| LarsErikP | thanks... still http 500 :S | 10:22 |
| LarsErikP | essex rc1 this is.. I forgot | 10:22 |
| LarsErikP | if that makes any difference | 10:23 |
| *** wariola has quit IRC | 10:23 | |
| *** arBmind has quit IRC | 10:27 | |
| *** alexn6 has joined #openstack | 10:30 | |
| *** supriya has quit IRC | 10:30 | |
| *** halfss__ has quit IRC | 10:32 | |
| *** po has joined #openstack | 10:37 | |
| *** GlenBaars has quit IRC | 10:39 | |
| *** arBmind has joined #openstack | 10:39 | |
| heyho | Hi everybody.. based on your experiences and attemps, what would be the best method to install and deploy essex-rc1 on ubuntu 12.04 beta? I mean, bleeding-edge ppas, github, others? Also, should I follow the trunk version docs? | 10:42 |
| *** dxd828 has quit IRC | 10:44 | |
| LarsErikP | uksysadmin: this weird. It works just fine if I just run "nova-api" in the shell. But it doesn't work at all when I do "service nova-api start" | 10:45 |
| *** oNeToWn has joined #openstack | 10:47 | |
| soren | LarsErikP: If you run nova-api in the shell (as yourself) you probably can't read your nova.conf (due to fs perms). | 10:47 |
| soren | LarsErikP: So it uses some defaults. | 10:48 |
| soren | LarsErikP: "service nova-api start" will probably run it as a different user which may read a config file (which may have an error in it). | 10:48 |
| uvirtbot | New bug: #961050 in glance "Creating bucket via region-specific S3 endpoint fails" [Medium,Confirmed] https://launchpad.net/bugs/961050 | 10:50 |
| *** rods has joined #openstack | 10:54 | |
| *** thinrhino has joined #openstack | 10:55 | |
| LarsErikP | soren: it works when i run it as root in the shell | 10:55 |
| LarsErikP | service ... does not | 10:56 |
| LarsErikP | hmm.. all nova.conf has 640 and root:nova | 10:57 |
| *** al-maisan is now known as almaisan-away | 11:00 | |
| *** hugokuo has quit IRC | 11:00 | |
| *** dendro-afk is now known as dendrobates | 11:01 | |
| LarsErikP | nova-api starts via service, but not nova-network :S | 11:04 |
| *** mdrnstm has quit IRC | 11:04 | |
| uksysadmin | LarsErikP: any clues in /var/log/syslog? | 11:05 |
| uksysadmin | heyho: essex rc1 (or v close to it) in ubuntu 12.04 b1 is good enough | 11:06 |
| uksysadmin | heyho: http://uksysadmin.wordpress.com/2012/02/09/updated-openstackinstaller-script-for-precise-and-essex-installs/ | 11:07 |
| heyho | uksysadmin, ok, thanks! | 11:08 |
| uksysadmin | heyho: np | 11:10 |
| *** kayyagari has joined #openstack | 11:12 | |
| oneiroi | And _just_ incase you're using RHEL 6.x: http://saiweb.co.uk/openstack/rhel-6-openstack-via-epel-nova-and-glance-on-kvm/ /endplug | 11:13 |
| LarsErikP | anyone knows why nova-compute insist this directory should exist? /usr/lib/python2.7/dist-packages/instances | 11:17 |
| *** davepigott has quit IRC | 11:17 | |
| LarsErikP | :s | 11:17 |
| *** davepigott has joined #openstack | 11:17 | |
| *** kayyagari has quit IRC | 11:18 | |
| *** kayyagari has joined #openstack | 11:19 | |
| LarsErikP | uksysadmin: syslog; Mar 21 12:19:49 cardiff kernel: [566861.804223] init: nova-network main process (30864) terminated with status 1 | 11:20 |
| alexn6 | uksysadmin, Hi again. I received answer yesterday, that they fix routing VM to itself publicIP bug in essex. So is not a problem anymore :) | 11:20 |
| *** saju_ma has joined #openstack | 11:21 | |
| *** dneary has quit IRC | 11:25 | |
| uksysadmin | alexn6: great! :) | 11:25 |
| *** sandywalsh has quit IRC | 11:25 | |
| uksysadmin | LarsErikP: try adding the following to your nova.conf...? | 11:26 |
| uksysadmin | --state_path=/var/lib/nova | 11:26 |
| uksysadmin | --lock_path=/var/lock/nova | 11:26 |
| uksysadmin | though if anything it's the state_path... but I believe it should default to that dir anyway...? | 11:26 |
| *** chrisfer has joined #openstack | 11:31 | |
| *** longhudou has quit IRC | 11:33 | |
| *** msavy has joined #openstack | 11:33 | |
| *** vizsla has joined #openstack | 11:34 | |
| lxu | how do I configure my credentials to use euca2ools with keystone? | 11:38 |
| *** sandywalsh has joined #openstack | 11:38 | |
| *** TREllis has quit IRC | 11:43 | |
| *** praefect has joined #openstack | 11:44 | |
| *** tkleinscea has joined #openstack | 11:48 | |
| *** dwcramer has joined #openstack | 11:49 | |
| *** guigui1 has quit IRC | 11:51 | |
| uksysadmin | lxu: keystone ec2-credentials-* commands | 11:52 |
| uksysadmin | or do you mean the envvars? | 11:52 |
| uksysadmin | I still get this when I try and use euca2ools | 11:52 |
| uksysadmin | euca-describe-instances | 11:52 |
| uksysadmin | Unauthorized: Failure communicating with keystone | 11:52 |
| *** tkleinscea has quit IRC | 11:54 | |
| *** scoph has quit IRC | 11:56 | |
| *** danpb has joined #openstack | 11:56 | |
| *** guigui1 has joined #openstack | 11:56 | |
| uksysadmin | just reminded me to fill in supporting info in a related bug | 11:57 |
| *** wonk has quit IRC | 11:57 | |
| *** wonk has joined #openstack | 11:58 | |
| danpb | ttx: what's the policy on submitting patches to gerrit for Essex milestone-proposed branch ? I presume I should submit to master/ first, and wait for it to be merged before submitting to milestone-proposed ? | 11:58 |
| lxu | uksysadmin: well, I want to use euca2ools in a openstack with keystone installation. I tried to export the ec2 variables but without success | 11:59 |
| lxu | euca-describe-instances Warning: failed to parse error message from AWS: <unknown>:1:0: syntax error None: None | 11:59 |
| lxu | Im using diablo from managedit | 12:00 |
| *** siwos has joined #openstack | 12:02 | |
| *** marrusl has joined #openstack | 12:05 | |
| *** tkleinscea has joined #openstack | 12:05 | |
| *** leifmadsen has joined #openstack | 12:06 | |
| uksysadmin | speak to Kiall - I always got that with the managedit ppa | 12:06 |
| uksysadmin | danpb: Essex is at RC status now - milestones precede this... | 12:06 |
| danpb | uksysadmin: well which branch should essex fixes go to then - GIT shows a choice of only master, milestone-proposed & stable/diablo | 12:08 |
| *** ahasenack has joined #openstack | 12:09 | |
| LarsErikP | uksysadmin: nova.conf http://paste.openstack.org/show/11662/ added what you said. It stopped bragging about that directory.. nova-network won't start from ther service command.. | 12:11 |
| *** arBmind has quit IRC | 12:12 | |
| *** maploin has joined #openstack | 12:13 | |
| *** cloudvirt has joined #openstack | 12:13 | |
| *** maploin has quit IRC | 12:13 | |
| *** maploin has joined #openstack | 12:13 | |
| uksysadmin | danpb: pass. I'm, sure ttx will pipe up as soon as he can. | 12:13 |
| *** vizsla has quit IRC | 12:13 | |
| LarsErikP | also getting alot of theese: Exception AssertionError: AssertionError() in <module 'threading' from '/usr/lib/python2.7/threading.pyc'> ignored | 12:13 |
| LarsErikP | both from nova-api and network | 12:14 |
| uksysadmin | LarsErikP: what versions are you using again and what distro? | 12:14 |
| LarsErikP | precise, essex rc1 | 12:14 |
| ttx | danpb: you should file bug, tag essex-rc-potential and push fix to master | 12:14 |
| ttx | danpb: if deemed RC-worthy it will get backported to milestone-proposed branch and included in the next RC | 12:14 |
| *** cloudvirt has left #openstack | 12:15 | |
| tkleinscea | Does anyone know where I can find a feature list for Essex compared to Diablo? | 12:15 |
| danpb | ttx: sorry, what do you mean by "tag essex-rc-potential" in this context ? | 12:16 |
| ttx | danpb: add a tag to the bug in launchpad | 12:16 |
| danpb | ah ok | 12:16 |
| *** dingdeng has quit IRC | 12:18 | |
| *** clopez has joined #openstack | 12:19 | |
| *** MrHeat has joined #openstack | 12:19 | |
| lxu | Kiall: are you there? | 12:19 |
| *** flaviamissi has joined #openstack | 12:20 | |
| *** arBmind has joined #openstack | 12:22 | |
| uksysadmin | LarsErikP: I just use the steps that are in my essex scripts found in git hub... best looking through that to find some clues - maybe some steps you've missed, or some config entries missing? | 12:22 |
| *** oubiwann has quit IRC | 12:23 | |
| *** TREllis has joined #openstack | 12:23 | |
| *** dneary has joined #openstack | 12:24 | |
| *** dneary has joined #openstack | 12:24 | |
| *** gekko_ has quit IRC | 12:24 | |
| *** lts has joined #openstack | 12:25 | |
| *** dingdeng has joined #openstack | 12:26 | |
| *** kbringard has joined #openstack | 12:26 | |
| hwestman | any horizon-pros here? | 12:28 |
| *** kbringard1 has joined #openstack | 12:30 | |
| *** roge has joined #openstack | 12:30 | |
| uksysadmin | TREllis: how are you? | 12:30 |
| *** msavy has quit IRC | 12:31 | |
| *** kbringard has quit IRC | 12:31 | |
| uksysadmin | TREllis: have you got euca2ools working with keystone/nova? I get the feeling I've made a fundamental mistake and want to compare notes. | 12:31 |
| *** msavy has joined #openstack | 12:32 | |
| *** roge has quit IRC | 12:35 | |
| melmoth | uksysadmin: i got it working with help from TREllis. Here is the env i used http://pastebin.com/h13DnUGQ | 12:36 |
| melmoth | just in case it helps :) | 12:36 |
| *** thinrhino has quit IRC | 12:37 | |
| *** roge has joined #openstack | 12:37 | |
| *** ccorrigan has quit IRC | 12:38 | |
| *** flamboz has joined #openstack | 12:39 | |
| *** dendrobates is now known as dendro-afk | 12:39 | |
| *** dendro-afk is now known as dendrobates | 12:42 | |
| melmoth | Hmm, with a git clone of keystone, i cannot create endpoint anymore. I have some 'TemplatedCatalog' object has no attribute 'service_exists' error. | 12:43 |
| *** kbringard1 has quit IRC | 12:43 | |
| melmoth | http://pastebin.com/yAX03uKN | 12:44 |
| *** aliguori has joined #openstack | 12:44 | |
| melmoth | does it ring a bell to anyone ? | 12:44 |
| *** mdomsch has quit IRC | 12:45 | |
| LarsErikP | uksysadmin: found out. had to kill dnsmasq, delete old bridges, and change som perms. works now :-) | 12:45 |
| LarsErikP | strange problem though... in horizon, i see my ubuntu-image in admin-> images, but not in project-> images and snapshots :s | 12:46 |
| *** cloudvirt has joined #openstack | 12:47 | |
| uksysadmin | cheers melmoth I have the same set up... will see what bug squashing ppl do | 12:48 |
| *** CrazyThinker has quit IRC | 12:48 | |
| *** kbringard has joined #openstack | 12:51 | |
| *** vitiho has quit IRC | 12:51 | |
| *** vitiho has joined #openstack | 12:52 | |
| *** dwcramer has quit IRC | 12:52 | |
| *** jastr has quit IRC | 12:53 | |
| *** olafont has quit IRC | 12:54 | |
| *** CrazyThinker has joined #openstack | 12:55 | |
| *** guigui1 has left #openstack | 12:55 | |
| *** lctmz has joined #openstack | 12:56 | |
| *** gekko_ has joined #openstack | 12:56 | |
| uksysadmin | LarsErikP: my horizon doesn't work... but check your glance index and whether the images are "public" or not | 12:56 |
| *** ldlework has joined #openstack | 12:56 | |
| Kiall | uksysadmin, see that fix i pushed for horizon? | 12:57 |
| Kiall | (For your issue..) | 12:57 |
| *** 65MAAEVEN has joined #openstack | 12:57 | |
| *** cryptk is now known as cryptk|offline | 12:58 | |
| Dieterbe | does swift chunk each individual object into chunks, or are objects generally kept as 1 piece unless you enable segmenting? | 12:59 |
| LarsErikP | uksysadmin: horizon says its public... | 13:01 |
| chmouel | Dieterbe: you need to do the chunking yourself and one object will be place on one object server | 13:01 |
| chmouel | Dieterbe: object can be a chunk | 13:01 |
| lctmz | hey guys, I'm trying to enable the nova-billing plugin on Horizon, but I'm not sure what I need configure. I already added ‘horizon_billing’ to INSTALLED_APPS tuple and the other part of configuration is add ‘billing’ to ‘dashboards’ key in HORIZON_CONFIG. But I can't find the HORIZON_CONFIG in local_settings.py . I'm doing it right? | 13:02 |
| *** dprince has joined #openstack | 13:05 | |
| TREllis | uksysadmin: I'm still hitting that issue from before with ec2 creds, I think melmoth solved it though... I've not had the time to look again yet | 13:05 |
| melmoth | i was able to use euc tools (list instance, run them and stuff), i had problems with bundling images and sshing to my vm though. | 13:06 |
| *** cloudvirt has quit IRC | 13:07 | |
| *** scoph has joined #openstack | 13:07 | |
| *** Glacee has joined #openstack | 13:07 | |
| *** SusanneB has quit IRC | 13:07 | |
| lxu | Kiall: I'm having error using ec2 with keystone, using the magedit packages, is this normal? | 13:08 |
| Kiall | No - the EC2 stuff works fine with those packages (with the exception of image upload, since that is not supported by diablo at all) | 13:08 |
| *** cloudvirt has joined #openstack | 13:09 | |
| lxu | Kiall: so how do I configure that? I'm able to use dashboard fine. I tried to configure like that : export EC2_SECRET_KEY=secretword export EC2_URL=http://192.168.1.120:8773/services/Cloud export EC2_ACCESS_KEY=adminUser | 13:10 |
| *** catintheroof has joined #openstack | 13:10 | |
| lxu | and when I run some command, it gives me Warning: failed to parse error message from AWS: <unknown>:1:0: syntax error None: None | 13:10 |
| Kiall | have you added EC2 credentials into keystone? or are you just tyring with your normal username and password? | 13:11 |
| Kiall | trying* | 13:11 |
| lxu | I have used the command sudo keystone-manage credentials add adminUser EC2 adminUser secretword adminTenant | 13:11 |
| lxu | like the openstack manual show | 13:12 |
| Kiall | Then thats all you should need to get it working.. | 13:12 |
| *** livemoon has joined #openstack | 13:12 | |
| Kiall | My env var template: https://github.com/managedit/openstack-setup/blob/master/openrc.tmpl#L26 | 13:12 |
| Kiall | and the credentials add command: https://github.com/managedit/openstack-setup/blob/master/keystone_data.sh.tmpl#L36 | 13:12 |
| Kiall | I've never needed to do anything else to get it working.. | 13:13 |
| lxu | Im pretty sure already tried that, but I will try again | 13:13 |
| Kiall | re the error message, yea, euca2tools gives some *really* *god* *damn* *terrible* error messages ;) | 13:14 |
| Kiall | try with --debug (or was it --verbose, I can never remember) | 13:14 |
| *** refactor has joined #openstack | 13:14 | |
| *** paulmillar has quit IRC | 13:18 | |
| LarsErikP | just uploaded a new image to glance, and trying to spawn it. Never used it before.. How long time should it take? | 13:19 |
| Kiall | depends on the speed of your network and disks.. | 13:20 |
| *** hub_cap has joined #openstack | 13:21 | |
| LarsErikP | ok.. so, i shouldn'r worry if it takes ~10 minutes? :p | 13:21 |
| *** hub_cap has quit IRC | 13:22 | |
| *** ayoung has quit IRC | 13:24 | |
| *** freeflyi1g has quit IRC | 13:25 | |
| *** freeflying has joined #openstack | 13:25 | |
| *** deshantm has joined #openstack | 13:27 | |
| Kiall | LarsErikP, 10 mins would be a tad too long | 13:28 |
| *** troytoman-away is now known as troytoman | 13:29 | |
| LarsErikP | heh.. okey | 13:30 |
| LarsErikP | i guess it's too long now then.. hmm.. | 13:30 |
| LarsErikP | cause, it worked with the tty-linux.. | 13:31 |
| *** koolhead11 has joined #openstack | 13:31 | |
| LarsErikP | but not with natty | 13:31 |
| *** hggdh has quit IRC | 13:31 | |
| Kiall | Time to check the error logs, the transfer could have failed or something | 13:31 |
| *** paulmillar has joined #openstack | 13:33 | |
| *** ppradhan has joined #openstack | 13:34 | |
| *** tkleinscea has quit IRC | 13:34 | |
| lctmz | anyone can help with nova-billing? I'm trying to use this tutorial http://openstackgd.wordpress.com/2012/01/24/billing-plugin-for-horizon/ , but it's not clear. I have already added ‘horizon_billing’ to the INSTALLED_APPS tuple in (nova_path)/dashboard/settings.py . But I didn't found HORIZON_CONFIG in local_settings.py . Help? | 13:35 |
| Kiall | lctmz, I've not used nova-billing.. but.. | 13:37 |
| Kiall | 1) Never edit stuff in (nova_path).. ;) | 13:37 |
| Kiall | 2) everything you set it settings.py can also be set in local_settings.py | 13:37 |
| Kiall | just copy it over | 13:37 |
| *** ayoung has joined #openstack | 13:37 | |
| Kiall | So, if settings.py has a HORIZON_CONFIG copt it over to local_settings.py and add whatever you need to it | 13:38 |
| *** hggdh has joined #openstack | 13:38 | |
| *** Guest10799 is now known as tobin | 13:39 | |
| *** tobin is now known as Guest70213 | 13:39 | |
| lxu | Kiall: I have used your file, and i'm still have error. with debug it gives me code 400 Bad Request. Using a novarc file of other installation, I only changed the ips and still the same error. Maybe the server is not working? how do I debug that | 13:40 |
| Kiall | lxu, have a look at the logs on the server, they may have something interesting in them | 13:40 |
| lxu | Kiall: the service that listen the port is the nova-api right? | 13:41 |
| uvirtbot | New bug: #961162 in nova "Failing to launch a instance in two node setup in Ubuntu 11.10 [ devstack]" [Undecided,New] https://launchpad.net/bugs/961162 | 13:41 |
| Kiall | lxu, not sure what you mean... | 13:41 |
| Kiall | the EC2 url is usually port 8773 | 13:41 |
| *** tryggvil_ has quit IRC | 13:42 | |
| Kiall | lxu, Actually - I forgot about another setting you may need | 13:42 |
| Kiall | https://github.com/managedit/openstack-setup/blob/master/nova.conf.tmpl | 13:42 |
| Kiall | keystone_ec2_url | 13:42 |
| lctmz | Kiall, thanks for the advice, I'm trying use this following the tutorial bellow and the github page, but these tutorials are very vague | 13:42 |
| *** cloudvirt has quit IRC | 13:43 | |
| Kiall | Yea - they tend to assume a certain level of familiarity with all the core parts to begin with | 13:43 |
| lxu | Kiall: ty, but I already have that | 13:44 |
| *** tkleinscea has joined #openstack | 13:44 | |
| *** sagar_nikam has joined #openstack | 13:44 | |
| *** marrusl has quit IRC | 13:45 | |
| Kiall | lxu, well, the scripts and templates in that repo take a blank 11.10 server, and do absolutely everything necessary to get keystone+glance+nova+horizon working.. including the EC2 api.. | 13:45 |
| Kiall | Have a look through them, you may find something I'm forgetting | 13:45 |
| Dieterbe | does swift perform checksums on all data it reads to make sure you only get the correct data? | 13:45 |
| *** lborda has joined #openstack | 13:46 | |
| MrHeat | Someone here can explain the deep of script prepare_guest.sh (Xen/XCP install of DevStack)? | 13:47 |
| lxu | Kiall: Ok, I will take a look =), ty for your help | 13:47 |
| *** mdomsch has joined #openstack | 13:48 | |
| LarsErikP | someone have experience with windwos images? | 13:52 |
| *** dendrobates is now known as dendro-afk | 13:52 | |
| *** CrazyThinker has quit IRC | 13:52 | |
| lctmz | Kiall, if neither settings.py and local_settings.py have INSTALLED_APPS and HORIZON_CONFIG, you know if can I simple add it? https://github.com/griddynamics/horizon-billing | 13:52 |
| *** andrewsmedina has joined #openstack | 13:53 | |
| Kiall | lctmz, one of those two should defiantly have INSTALLED_APPS | 13:53 |
| Kiall | Its kinda "core" to Django.. | 13:53 |
| Kiall | definitely* | 13:54 |
| lctmz | Kiall, yep the settings.py have INSTALLED_APPS, but neither settings.py or local_settings.py have HORIZON_CONFIG | 13:54 |
| *** marrusl has joined #openstack | 13:54 | |
| Kiall | Oh.. I'm not sure.. | 13:54 |
| Kiall | what version of horizon are you using? | 13:54 |
| Kiall | essex or diablo? | 13:54 |
| lctmz | diablo | 13:55 |
| *** michael_tn has joined #openstack | 13:55 | |
| *** GlenBaars has joined #openstack | 13:55 | |
| *** rkukura has quit IRC | 13:55 | |
| Kiall | Ah.. Is the nova-billion addon for diablo or essex? | 13:56 |
| Kiall | I think that option was added on essex | 13:56 |
| GlenBaars | it there a paid support? I've got an issue that I have been working with for 6 months now! I feel it is a xenserver issue not a cloudstack one. | 13:56 |
| *** AlanClark has joined #openstack | 13:56 | |
| Kiall | GlenBaars, wrong channel for anything cloudstack related ;) | 13:57 |
| GlenBaars | soz | 13:57 |
| Kiall | no hassle.. openstack != cloudstack even though they sound pretty similar1 | 13:58 |
| Kiall | no hassle.. openstack != cloudstack even though they sound pretty similar!* | 13:58 |
| GlenBaars | yeah, I have a company called opencloud too .... | 13:58 |
| *** CrazyThinker has joined #openstack | 13:59 | |
| lctmz | I don't think so, in the github page it's not explicit, but in the tutorials that I saw they comment about diablo | 13:59 |
| *** siwos has quit IRC | 13:59 | |
| Kiall | lctmz, strange.. I could be wrong.. Either way, its getting into specifics I dont know anything about :) I just pointed out the generic stuff i could help with | 13:59 |
| *** troytoman is now known as troytoman-away | 14:00 | |
| *** andrewsmedina has quit IRC | 14:00 | |
| lctmz | Kiall, ok. Thank you anyway (: | 14:01 |
| *** taihen has quit IRC | 14:01 | |
| *** jastr has joined #openstack | 14:02 | |
| *** taihen has joined #openstack | 14:02 | |
| *** swiftindia has quit IRC | 14:04 | |
| *** swiftindia has joined #openstack | 14:04 | |
| notmyname | Dieterbe: yes. swift audits the data in the background and quarantines any file not matching its checksum | 14:06 |
| *** dtroyer_zzz is now known as dtroyer | 14:06 | |
| *** andrewsmedina has joined #openstack | 14:07 | |
| Madkiss | Can somebody provide me with a howto on how to use RBD devices in nova-volume | 14:08 |
| Dieterbe | notmyname: but, since there are $replication_factor-1 other copies around, why not try to read those? | 14:08 |
| *** stewart has quit IRC | 14:08 | |
| Dieterbe | notmyname: it seems likely one of the other machines will have a valid version | 14:09 |
| annegentle | Madkiss: I don't know of one, sorry. | 14:09 |
| notmyname | Dieterbe: we do. after a file is quarantined, replication will make sure that the nth replica gets replaced with a good copy | 14:09 |
| *** dwcramer has joined #openstack | 14:09 | |
| Dieterbe | notmyname: and the client will get a read error, or the correct copy? | 14:10 |
| lxu | Kiall: I was able to run ec2 now | 14:10 |
| Kiall | what was it? | 14:11 |
| *** osier has joined #openstack | 14:11 | |
| Madkiss | annegentle: okay | 14:12 |
| *** clopez has quit IRC | 14:12 | |
| annegentle | Madkiss: I can see there's a nova flag for it, I wonder if libvirt has instructions you could follow? | 14:12 |
| *** tkleinscea has quit IRC | 14:12 | |
| annegentle | Madkiss: (sorry, I know that's not a great answer) | 14:13 |
| *** justfine has quit IRC | 14:13 | |
| lxu | Kiall: Well, I'm not really sure, but the docs manual create ec2 users credential using only the user name. The templates at managedit used "credentials add admin EC2 'admin:admin' " | 14:14 |
| lxu | I just create a new project and add this credential like the template uses, and then it works | 14:15 |
| Kiall | lxu, yea my template uses username:tenant as the key, so that the same user can have EC2 creds for multiple tenants | 14:15 |
| Kiall | annegentle, ^ lxu's thing might be a doc bug? | 14:15 |
| Kiall | anyway - gotta run | 14:15 |
| *** dolphm has joined #openstack | 14:16 | |
| *** dolphm_ has joined #openstack | 14:16 | |
| notmyname | Dieterbe: auditing and replication run in the background. we also check the checksum on GET requests and quarantine if necessary. | 14:17 |
| sstent_ | madkiss: i dont think nova-volume has that option ? though nova-compute does | 14:17 |
| notmyname | Dieterbe: if the checksum on GET fails, the file is quarantined, but that request may have received a bad copy | 14:17 |
| *** littleidea has quit IRC | 14:17 | |
| Dieterbe | thanks | 14:19 |
| Madkiss | sstent_: can you point me to some documentation to that? :) | 14:19 |
| *** dwcramer has quit IRC | 14:19 | |
| annegentle | lxu: which docs manual does the user name only? | 14:20 |
| sstent_ | madkiss: griddynamics had some notes on using lvm for compute, which should be close http://openstackgd.wordpress.com/2012/01/20/using-lvm-as-disk-storage-for-openstack/ Dare I ask why you want to use RBDs? | 14:20 |
| *** dolphm has quit IRC | 14:21 | |
| *** GlenBaars has left #openstack | 14:21 | |
| *** dendro-afk is now known as dendrobates | 14:22 | |
| Madkiss | sstent_: i'm trying to fiddle out how to make the nova-volume architecture HA aware | 14:22 |
| *** stewart has joined #openstack | 14:22 | |
| *** dspano has joined #openstack | 14:22 | |
| uksysadmin | Kiall: thanks for the patch to horizon | 14:23 |
| uksysadmin | as well as your responses | 14:23 |
| lxu | annegentle: diablo keystone manual http://docs.openstack.org/diablo/openstack-compute/install/content/identity-define-services-endpoints.html | 14:24 |
| *** robbiew has joined #openstack | 14:25 | |
| Madkiss | sstent_: I think I was looking for https://code.launchpad.net/~jdurgin/nova/rbd_volume/+merge/45091 | 14:25 |
| *** oubiwann has joined #openstack | 14:25 | |
| sstent_ | madkiss: cool :) I think that's for compute not volume btw | 14:27 |
| lxu | annegentle: seen again, the problem is a little different, is missing some parameter. I used "keystone-manage credentials add admin EC2 'admin:admin' %ADMIN_PASSWORD% admin" and the manual uses "sudo keystone-manage credentials add adminUser EC2 'secretword' adminTenant" | 14:27 |
| *** tkleinscea has joined #openstack | 14:27 | |
| lxu | (different users and projects) | 14:27 |
| Madkiss | sstent_: obviously, yes, hmm. | 14:28 |
| *** almaisan-away is now known as al-maisan | 14:28 | |
| *** ryanpetrello has joined #openstack | 14:28 | |
| *** CrazyThinker has quit IRC | 14:28 | |
| annegentle | lxu: oo. hm. Yes I think the doc may be wrong there. I'll file a doc bug | 14:29 |
| annegentle | lxu: so your command worked (where you use admin:admin) right? | 14:29 |
| lxu | annegentle: yes | 14:29 |
| annegentle | lxu: where admin is the name of the user and tenant both | 14:29 |
| lxu | annegentle: yes | 14:29 |
| annegentle | lxu: ok, thanks, filing bug now | 14:29 |
| hwestman | im having this exact issue: https://lists.launchpad.net/openstack/msg07268.html anyone? | 14:31 |
| *** stuntmachine has joined #openstack | 14:32 | |
| *** CrazyThinker has joined #openstack | 14:33 | |
| *** krow has joined #openstack | 14:34 | |
| *** troytoman-away is now known as troytoman | 14:35 | |
| *** melmoth has quit IRC | 14:36 | |
| *** melmoth has joined #openstack | 14:36 | |
| *** tryggvil_ has joined #openstack | 14:40 | |
| *** shall_ has joined #openstack | 14:40 | |
| *** rkukura has joined #openstack | 14:41 | |
| *** dtroyer is now known as dtroyer_zzz | 14:43 | |
| *** zzed has joined #openstack | 14:47 | |
| sagar_nikam | Hi, I am documenting the dashboard. have a question. what is the default login user name of the instance launched. | 14:47 |
| sagar_nikam | in instances launched on trystack.org, the user is ubuntu | 14:48 |
| uvirtbot | New bug: #961220 in openstack-manuals "In Diablo install doc, EC2 credentials setup is missing user:tenant (has just user)" [Undecided,New] https://launchpad.net/bugs/961220 | 14:48 |
| *** natea has joined #openstack | 14:48 | |
| koolhead11 | sagar_nikam: are you documenting it for essex? | 14:49 |
| *** lctmz has quit IRC | 14:49 | |
| sagar_nikam | koolhead11: yes for essex | 14:49 |
| sagar_nikam | just launching instances part of documentation | 14:50 |
| *** thinrhino has joined #openstack | 14:50 | |
| koolhead11 | sagar_nikam: when you say Dashboard/Horizon you need to talk about keystone as well | 14:50 |
| sagar_nikam | for this bug https://bugs.launchpad.net/openstack-manuals/+bug/952302 | 14:51 |
| koolhead11 | a user associated with a particular tenant with admin enabled will be able to log in | 14:51 |
| sagar_nikam | ok | 14:52 |
| *** mdomsch has quit IRC | 14:53 | |
| *** n0ano has quit IRC | 14:53 | |
| *** GheRivero_ has joined #openstack | 14:54 | |
| *** dtroyer_zzz is now known as dtroyer | 14:55 | |
| uvirtbot | New bug: #961237 in nova "Instance remains in scheduling state if RabbitMQ is down during instance creation" [Undecided,New] https://launchpad.net/bugs/961237 | 14:56 |
| *** imsplitbit has joined #openstack | 14:58 | |
| *** rnorwood has joined #openstack | 15:02 | |
| *** nhm has quit IRC | 15:02 | |
| *** mattray has joined #openstack | 15:03 | |
| *** justfine has joined #openstack | 15:03 | |
| melmoth | uksysadmin: i would you have an /etc/swift tarball of a swift setting working with keystone by any chance ? | 15:06 |
| *** sjhon has joined #openstack | 15:09 | |
| sjhon | Hi all. I have some questions about Swift. Is there any channel for it ? | 15:10 |
| *** Blackavar has quit IRC | 15:10 | |
| *** tkleinscea has quit IRC | 15:10 | |
| notmyname | sjhon: you're in it | 15:10 |
| *** joesavak has joined #openstack | 15:15 | |
| uksysadmin | melmoth: its not that simple | 15:15 |
| uksysadmin | on the other hand it is simple | 15:16 |
| melmoth | i fail to see the simplicity of it :) | 15:16 |
| *** thinrhino has quit IRC | 15:16 | |
| * koolhead11 kicks uksysadmin | 15:16 | |
| uksysadmin | melmoth: I feel your pain - I've seen the light though. | 15:16 |
| * uksysadmin kicks koolhead11 back! | 15:16 | |
| *** mrjazzcat has joined #openstack | 15:16 | |
| * koolhead11 gets back to life. | 15:17 | |
| koolhead11 | uksysadmin: howdy sir :P | 15:17 |
| sjhon | I'm reading the Swift documentation at: http://swift.openstack.org/deployment_guide.html. And I have some doubts. They talk about drives and it is not clear for me what does "drive" means. Is it a HArd Disk ? | 15:18 |
| uksysadmin | melmoth: I'll make some notes for you | 15:18 |
| notmyname | sjhon: yes. more generally it's "storage volumes" where a volume is uniquely identified by (IP, port, mount point) | 15:18 |
| *** osier has quit IRC | 15:19 | |
| *** saju_ma has quit IRC | 15:19 | |
| notmyname | sjhon: where IP and port are the IP and port of the swift process (one of the account, container, or object servers running in the cluster) | 15:20 |
| sjhon | notmyname, I physical partition in a hard disk can be drive ? | 15:21 |
| *** smulcahy has joined #openstack | 15:21 | |
| notmyname | sjhon: generally you should use drives with one filesystem partition on them (ie use the whole drive) | 15:22 |
| sjhon | thanks. Another doubt is about the term "partition" in the documentation. What is a "partition" ? Is it like a physical partition in a HD ? | 15:23 |
| *** tkleinscea has joined #openstack | 15:24 | |
| *** cw has quit IRC | 15:24 | |
| *** cw has joined #openstack | 15:24 | |
| *** pmatulis has joined #openstack | 15:25 | |
| *** pmatulis has left #openstack | 15:25 | |
| smulcahy | is anyone using nova-volumes with the managedit ppa? | 15:26 |
| smulcahy | i'm running into some iscsi errors and wondering is it a config issue or a buglet | 15:27 |
| uksysadmin | melmoth: have a look at this: http://paste.openstack.org/show/11702/ | 15:29 |
| *** CrazyThinker has quit IRC | 15:29 | |
| melmoth | thanks uksysadmin | 15:31 |
| uksysadmin | np | 15:31 |
| *** robix has joined #openstack | 15:32 | |
| *** tkleinscea has quit IRC | 15:32 | |
| *** swill has joined #openstack | 15:34 | |
| *** clauden has joined #openstack | 15:35 | |
| swill | has anyone else been getting this recently? STDOUT: /swift/proxy/server.py:804: DeprecationWarning: Response.request and Response.environ are deprecated#012 source.getheader('x-timestamp') | 15:35 |
| *** 65MAAEVEN has quit IRC | 15:35 | |
| *** clauden has quit IRC | 15:35 | |
| *** CrazyThinker has joined #openstack | 15:38 | |
| swill | (looking at the current code on github, the line numbers are not identical). | 15:38 |
| *** supriya has joined #openstack | 15:39 | |
| swill | the reason i am looking into this is because when this occures, slogging stops working. | 15:39 |
| Kyle__ | Can you mix hypervisor types on the same compute nodes? | 15:41 |
| *** supriya has quit IRC | 15:41 | |
| *** supriya has joined #openstack | 15:42 | |
| *** gekko_ has quit IRC | 15:42 | |
| *** nOStahl has joined #openstack | 15:42 | |
| *** msavy has quit IRC | 15:43 | |
| swill | notmyname: do you have monitoring setup on slogging to verify that it is working correctly? | 15:44 |
| *** vitiho has quit IRC | 15:44 | |
| *** gekko_ has joined #openstack | 15:44 | |
| *** CrazyThinker has quit IRC | 15:44 | |
| *** supriya has quit IRC | 15:45 | |
| *** al-maisan is now known as almaisan-away | 15:45 | |
| smulcahy | has anyone run into "Error: Could not find iSCSI export for volume" ? | 15:45 |
| *** alperkanat has joined #openstack | 15:45 | |
| *** supriya has joined #openstack | 15:45 | |
| zynzel | smulcahy: check nova-volume: tgtadm --lld iscsi --op ^Cow --mode target | 15:46 |
| zynzel | check if target name exist, backing store path should match target name | 15:47 |
| zynzel | tgtadm --lld iscsi --op show --mode target * | 15:47 |
| smulcahy | tgtadm isn't installed | 15:47 |
| smulcahy | does that suggest I'm using a different version of iscsi? | 15:47 |
| smulcahy | or do I just need to go ahead and install it? | 15:48 |
| *** tkleinscea has joined #openstack | 15:48 | |
| zynzel | grep iscsi_helper /etc/nova/nova.conf | 15:48 |
| smulcahy | nada | 15:48 |
| *** shevek_ has quit IRC | 15:49 | |
| zynzel | smulcahy: default is ietadm, but i cant help with that ;) | 15:49 |
| smulcahy | zynzel: :-) | 15:50 |
| smulcahy | zynzel: thanks for trying | 15:50 |
| zynzel | but you can try ietadm --op show | 15:50 |
| smulcahy | Invalid argument. | 15:50 |
| zynzel | ;) | 15:50 |
| smulcahy | any suggestions on how I can see what iscsi targetname openstack is using on the compute node? | 15:51 |
| *** davepigott has quit IRC | 15:51 | |
| smulcahy | I have manually connected to an iscsi target on volume server | 15:51 |
| smulcahy | but can't see what target is failing when I use euca-attach-volume | 15:51 |
| zynzel | smulcahy: in log of compute node | 15:51 |
| *** wilmoore has joined #openstack | 15:52 | |
| smulcahy | zynzel: iscsiadm -m discovery -t sendtargets -p nimbus05 is the last command listed | 15:52 |
| smulcahy | and that gives me output | 15:52 |
| *** CrazyThinker has joined #openstack | 15:52 | |
| smulcahy | but I guess not the output that the nova-compute is expecting | 15:53 |
| *** alperkanat has quit IRC | 15:53 | |
| *** supriya has quit IRC | 15:53 | |
| *** supriya has joined #openstack | 15:53 | |
| zynzel | i can see in logs smth like: sudo nova-rootwrap iscsiadm -m node -T iqn.2010-10.org.openstack:volume-cc1-000000c3 -p 10.189.0.31:3260 --login | 15:53 |
| smulcahy | yeah, I never get to see that | 15:54 |
| smulcahy | I get the error before-hand | 15:54 |
| *** michael_tn has quit IRC | 15:55 | |
| smulcahy | so I guess its not finding the volume name in the format or with the address it is expecting | 15:55 |
| smulcahy | but it doesn't log what its looking for | 15:55 |
| *** guigui3 has joined #openstack | 15:55 | |
| *** sandywalsh has quit IRC | 15:56 | |
| *** asavu has quit IRC | 15:57 | |
| alexn6 | have you nova-volume && lvm2 installed?and do all instruction for set up volume for it? | 15:58 |
| smulcahy | yup | 15:58 |
| smulcahy | I think I'll try hacking some extra log messages into /usr/lib/python2.7/dist-packages/nova/volume/driver.py and see if that helps | 15:58 |
| *** natea has quit IRC | 15:58 | |
| *** almaisan-away is now known as al-maisan | 15:59 | |
| *** natea has joined #openstack | 15:59 | |
| alexn6 | does it create volume in lvm? | 15:59 |
| zynzel | smulcahy: you can match volume id with target name in database | 15:59 |
| zynzel | nova volume-list | 15:59 |
| zynzel | +-----+-----------+--------------+------+-------------+ | 15:59 |
| zynzel | | ID | Status | Display Name | Size | Attached to | | 15:59 |
| zynzel | +-----+-----------+--------------+------+-------------+ | 15:59 |
| zynzel | | 196 | available | ads | 1 | None | | 15:59 |
| zynzel | +-----+-----------+--------------+------+-------------+ | 15:59 |
| zynzel | mysql> select id,provider_location from volumes where id=196; | 15:59 |
| zynzel | +-----+------------------------------------------------------------------+ | 15:59 |
| zynzel | | id | provider_location | | 15:59 |
| zynzel | +-----+------------------------------------------------------------------+ | 15:59 |
| zynzel | | 196 | 10.189.0.31:3260,7 iqn.2010-10.org.openstack:volume-cc1-000000c4 | | 15:59 |
| zynzel | +-----+------------------------------------------------------------------+ | 15:59 |
| *** joesavak has quit IRC | 15:59 | |
| zynzel | sorry for pasting it here :) | 15:59 |
| Kyle__ | Hum. I guess before even thinking of mixing, has anyone here ever used openstack with hypervisors OTHER than kvm? | 16:00 |
| *** supriya has quit IRC | 16:00 | |
| smulcahy | zynzel: Thanks, my provider_location is listed as NULL | 16:00 |
| smulcahy | so I guess the discovery is failing for some reason | 16:01 |
| *** supriya has joined #openstack | 16:01 | |
| notmyname | swill: that's a webob version mismatch | 16:01 |
| *** chevin has quit IRC | 16:01 | |
| zynzel | smulcahy: provider_location is updated after creating volume imho | 16:01 |
| zynzel | but i dont know for sure. | 16:01 |
| swill | notmyname: really? i am runnign 1.0.8 think. (i had that issue before). | 16:01 |
| * swill checking | 16:01 | |
| alexn6 | did you set --iscsi_ip_prefix to a useful value? | 16:01 |
| *** supriya has quit IRC | 16:02 | |
| alexn6 | launchpad answers have your problem :) | 16:02 |
| smulcahy | yes, set --iscsi_ip_prefix ok | 16:02 |
| *** vitiho has joined #openstack | 16:03 | |
| alexn6 | smoser, in that way - "if your local network is 192.168.1.0/24 for example, try setting the prefix to 192.168.1" ? | 16:03 |
| *** rnirmal has joined #openstack | 16:04 | |
| swill | notmyname: do you remember how to check the version of webob? i did it before and i cant find how i did it. | 16:04 |
| alexn6 | or mb this one https://answers.launchpad.net/nova/+question/169359 | 16:04 |
| *** joesavak has joined #openstack | 16:04 | |
| notmyname | swill: dpkg -l | grep webob ;-) | 16:04 |
| annegentle | koolhead11: around? | 16:05 |
| *** maplebed has joined #openstack | 16:05 | |
| *** jsavak has joined #openstack | 16:06 | |
| *** clopez has joined #openstack | 16:06 | |
| *** msavy has joined #openstack | 16:07 | |
| *** fc__ has quit IRC | 16:08 | |
| *** deshantm_ has joined #openstack | 16:08 | |
| *** joesavak has quit IRC | 16:09 | |
| *** lctmz has joined #openstack | 16:10 | |
| *** al-maisan is now known as almaisan-away | 16:11 | |
| *** clauden_ has joined #openstack | 16:11 | |
| *** sandywalsh has joined #openstack | 16:11 | |
| *** deshantm has quit IRC | 16:12 | |
| *** dieterdemeyer has left #openstack | 16:12 | |
| *** almaisan-away is now known as al-maisan | 16:13 | |
| *** freeflying has quit IRC | 16:14 | |
| *** freeflying has joined #openstack | 16:16 | |
| uvirtbot | New bug: #961325 in melange "nova instance IDs in interfaces table" [Undecided,New] https://launchpad.net/bugs/961325 | 16:16 |
| sjhon | I have a doubt about Swift. I was reading the documentation http://swift.openstack.org/deployment_guide.html, but I did not understant what does "partition" means. I thought it was like a common partition in a HD, but it stange for me to have 100 partitions, for example. Does anyone could help me ? | 16:16 |
| notmyname | sjhon: I can, but I'm headed to lunch. I can answer that when I get back | 16:16 |
| sjhon | thanks, notmyname. | 16:17 |
| *** swill_ has joined #openstack | 16:18 | |
| *** vincentricci has joined #openstack | 16:19 | |
| koolhead11 | annegentle: hello there | 16:20 |
| annegentle | koolhead11: hey I had to reinstall bzr to get latest cssoss doc | 16:20 |
| *** swill has quit IRC | 16:20 | |
| *** swill_ is now known as swill | 16:21 | |
| koolhead11 | annegentle: gosh. why :( | 16:21 |
| annegentle | koolhead11: for image management, how have you documented it? | 16:21 |
| annegentle | koolhead11: oh because I upgraded to max 10.7 | 16:21 |
| *** Blackavar has joined #openstack | 16:21 | |
| annegentle | koolhead11: I will get the latest branch but also thought I'd ask | 16:21 |
| koolhead11 | annegentle: cool | 16:21 |
| *** smulcahy has quit IRC | 16:22 | |
| *** msavy has quit IRC | 16:22 | |
| *** livemoon has left #openstack | 16:22 | |
| *** msavy has joined #openstack | 16:22 | |
| koolhead11 | image management will remain same, just need to add keystone in picture. guys are working on testing part | 16:22 |
| annegentle | koolhead11: ok. it used nova-manage commands, which will be deprecated in Folsom, so I was going to use all glance add commands | 16:22 |
| koolhead11 | annegentle: makes sense. We will be using glance command only 4 image management on Essex | 16:23 |
| *** shaon has joined #openstack | 16:24 | |
| *** swill_ has joined #openstack | 16:24 | |
| *** deshantm_ is now known as deshantm | 16:24 | |
| koolhead11 | annegentle: sounds good :) | 16:25 |
| annegentle | koolhead11: thanks | 16:26 |
| *** nati has joined #openstack | 16:26 | |
| koolhead11 | annegentle: so folsem wont have nova-objecstore :P | 16:27 |
| annegentle | koolhead11: doesn't sound like it, does it? I'm still unclear on the pre-reqs for bundling to work with Essex though. | 16:27 |
| koolhead11 | annegentle: i donno am asking. :) | 16:28 |
| *** bepernoot has quit IRC | 16:28 | |
| *** CrazyThinker has quit IRC | 16:28 | |
| *** heckj has joined #openstack | 16:28 | |
| *** swill has quit IRC | 16:28 | |
| *** swill_ is now known as swill | 16:28 | |
| *** popux has joined #openstack | 16:28 | |
| *** davepigott has joined #openstack | 16:29 | |
| *** reed has joined #openstack | 16:29 | |
| *** no`x has joined #openstack | 16:29 | |
| koolhead11 | hello heckj | 16:30 |
| annegentle | koolhead11: yeah unclear to me at this point | 16:30 |
| koolhead11 | annegentle: probably after design summit it will be more clear | 16:31 |
| *** jdurgin has joined #openstack | 16:31 | |
| heckj | morning koolhead11 | 16:31 |
| *** chevin has joined #openstack | 16:32 | |
| koolhead11 | heckj: howdy? | 16:32 |
| *** reidrac has quit IRC | 16:32 | |
| * no`x not able to find a changelog for essex, can anyone help me here? | 16:32 | |
| no`x | we are right now starting a poc and are not sure if it is not the better idea to start with rc1 | 16:34 |
| heckj | no`x: from my point of view (then, again, I'm deep in openstack dev), essex rc1 is much better than diablo release. | 16:34 |
| heckj | we'll be finding some bugs, but if you're willing to work through rejiggering docs, the end result is better | 16:35 |
| heckj | a lot higher quality release and focus on testing in RC/final stage of the 6mo cycle this time through | 16:35 |
| no`x | nice | 16:35 |
| *** CrazyThinker has joined #openstack | 16:36 | |
| no`x | thats what i thought. any paper to lure my workmates? | 16:37 |
| *** joearnold has joined #openstack | 16:38 | |
| *** joearnold has quit IRC | 16:39 | |
| *** shall_ has quit IRC | 16:39 | |
| alexn6 | sjhon, patrition is not a hdaNN disk/OS thing, it`s a swift term, which just a group of chunks or data or files, that managed by swift together. Read about ring if you want | 16:40 |
| *** kebray has joined #openstack | 16:40 | |
| sjhon | thanks alexn6 | 16:40 |
| *** mindpixel has quit IRC | 16:41 | |
| *** kebray has quit IRC | 16:41 | |
| uvirtbot | New bug: #961354 in keystone "python-keystoneclient job fails to build a tarball" [Critical,In progress] https://launchpad.net/bugs/961354 | 16:41 |
| heckj | no`x: nothing that I can easily point to, although I expect some preso and papers are getting generated now that RC1 is about cut | 16:41 |
| *** spiffxp has quit IRC | 16:42 | |
| *** joearnold has joined #openstack | 16:42 | |
| sjhon | alexn6, what does 100 partitions mean ? | 16:42 |
| no`x | btw big praise doc enhanced great the last month | 16:43 |
| alexn6 | sjhon, try to read about Ring | 16:43 |
| annegentle | no`x: thanks that type of encouragement will get us through the next month :) | 16:43 |
| no`x | heh | 16:44 |
| koolhead11 | annegentle: your rocking!! :) | 16:44 |
| alexn6 | sjhon, where from did you get 100 partitions? | 16:46 |
| sjhon | alexn6, I read bout it. But the documentation is not so clear. I got it from http://swift.openstack.org/deployment_guide.html | 16:46 |
| sjhon | alexn6, they state that you need at least 100 partitions per drive. I just don't undestand why this number and the term partition is not clear. | 16:47 |
| alexn6 | sjhon, there is only weight term of 100 per 1Tb drive | 16:48 |
| *** swill has quit IRC | 16:48 | |
| alexn6 | sjhon, let get it in other way :) - how many 1TB drives do you plan to use? | 16:49 |
| *** CrazyThinker has quit IRC | 16:49 | |
| sjhon | alexn, suppose I have 8TB drives. How can I compute the partitions ? | 16:50 |
| sjhon | ops | 16:50 |
| *** swill has joined #openstack | 16:50 | |
| sjhon | 8 x 1TB drives | 16:50 |
| *** al-maisan is now known as almaisan-away | 16:50 | |
| uvirtbot | New bug: #961367 in openstack-ci "Some things appear too large in gerrit new style" [Undecided,New] https://launchpad.net/bugs/961367 | 16:50 |
| *** davepigott has quit IRC | 16:51 | |
| sjhon | alexn, in the documentation the say: "We recommend that there be a minimum of 100 partitions per drive to insure even distribution across the drives. A good starting point might be to figure out the maximum number of drives the cluster will contain, and then multiply by 100, and then round up to the nearest power of two." | 16:52 |
| *** swill has quit IRC | 16:52 | |
| *** spiffxp has joined #openstack | 16:52 | |
| alexn6 | well, wait I need to think about it | 16:53 |
| sjhon | ok | 16:53 |
| *** gyee has joined #openstack | 16:54 | |
| uvirtbot | New bug: #961370 in openstack-manuals "tenantid/tenantname use should be documented in v2.0/tokens" [Undecided,New] https://launchpad.net/bugs/961370 | 16:56 |
| *** joearnold has quit IRC | 16:56 | |
| *** joearnold has joined #openstack | 16:57 | |
| *** warik has joined #openstack | 16:57 | |
| *** sandywalsh has quit IRC | 16:58 | |
| *** CrazyThinker has joined #openstack | 16:58 | |
| alexn6 | sjhon, how I understand it - the value parameters is partition power(which depends of your total disk numbers and your choice of segment size) and disk weigh(later in docs), which is relative | 16:59 |
| *** vincentricci has quit IRC | 17:00 | |
| *** vincentricci has joined #openstack | 17:00 | |
| *** dendrobates is now known as dendro-afk | 17:00 | |
| alexn6 | sjhon, 1Tb=10^30 which leave for partition power(config flag) 18 partition size of 2^12 what is 4 kb. for 8 Tb if you want say 3 replicas it only 2 useful Tb - what mean that if you chose part_power of 18 - you will have 8kb segments | 17:00 |
| *** sagar_nikam has quit IRC | 17:01 | |
| *** dendro-afk is now known as dendrobates | 17:02 | |
| sjhon | alexn6, if I have 8 drives of 8TB. Suppose I have 100 partitions, my ring would have 100*8=800 and need to round it to 2^10. 2^10 would be the max of partitions on the ring ? | 17:02 |
| alexn6 | 100 is drive(or more exactly node weight) - what means that if you have 4 zones you may add 2x1Tb disk to one zone, what give you for 3 replicas about 2,33 Tb total real space | 17:03 |
| *** nacx has quit IRC | 17:03 | |
| alexn6 | sjhon, you may do it so, or may increase this value | 17:04 |
| *** uksysadmin has quit IRC | 17:04 | |
| *** nati has quit IRC | 17:06 | |
| sjhon | alexn6, the total of partitions is the maximum the ring can support ? If the ring is in the maximum, can I create anothe ring ? | 17:06 |
| alexn6 | sjhon, if you have a 8kb segment and 2^18 of partitions - it`s a quite large work for swift services. You may decrease it to 2^10 and have a 2Mb segment. Think about it like a FS smallest segment, that file system will "defragment" and other things | 17:07 |
| sjhon | alexn6, What do you mean as "segment' ? | 17:08 |
| alexn6 | sjhon, no. As I understand, it define a segment size. Of course you can create another ring, but you will need to rebalnce old data_segments and replicas to fit your new ring | 17:09 |
| alexn6 | you know fat32 FS segment? | 17:09 |
| sjhon | yes | 17:09 |
| alexn6 | kinda of that | 17:10 |
| alexn6 | the smallest data block, that swift would manage | 17:10 |
| alexn6 | at once | 17:10 |
| alexn6 | i.e -replicate, move, read or write | 17:10 |
| *** swill has joined #openstack | 17:10 | |
| alexn6 | and address | 17:11 |
| alexn6 | more partitions - then more "index" but less expensive operations of data block operation. less partition - larger its size and more expensive operations. RElative of course | 17:12 |
| uvirtbot | New bug: #961394 in openstack-ci "Reviewed checkmark in gerrit is missing" [High,New] https://launchpad.net/bugs/961394 | 17:12 |
| *** maploin has quit IRC | 17:13 | |
| sjhon | alexn6 thanks. If I have a senario like that: 4 zones, 8 drives of 1TB each, and 3 replicas. How could I compute the number of partitions/ segments ? | 17:13 |
| *** sandywalsh has joined #openstack | 17:13 | |
| uvirtbot | New bug: #961385 in openstack-ci "Jenkins breadcrumbs transparancy scroll issue" [Undecided,New] https://launchpad.net/bugs/961385 | 17:13 |
| alexn6 | sjhon, I`ve already give you an example above - for 1Tb and 8 also | 17:14 |
| *** swill has quit IRC | 17:15 | |
| sjhon | alexn6, I just missed the relation partition/segment. You stated that "8 Tb if you want say 3 | 17:16 |
| sjhon | replicas it only 2 useful Tb - what mean that if you chose | 17:16 |
| sjhon | part_power of 18 - you will have 8kb segments" | 17:16 |
| *** souza has joined #openstack | 17:16 | |
| alexn6 | they are the same | 17:16 |
| sjhon | but how many partitions would be necessary for these segments ? | 17:16 |
| souza | Hello guys i'm having a problem with glance registry, i try up the service and it stops some seconds before | 17:17 |
| *** oneiroi has quit IRC | 17:17 | |
| alexn6 | sjhon, power of partition - is your real number of them. And 100 per drive is relative weight to other drives | 17:17 |
| *** dendrobates is now known as dendro-afk | 17:17 | |
| souza | Hello guys i'm having a problem with glance registry, i try up the service and it stops some seconds after* | 17:17 |
| alexn6 | sjhon, if you build ring with first parameter(power of partitions) 18 - you will have 2^18 of them. You may make this smaller | 17:18 |
| uvirtbot | New bug: #961395 in openstack-ci "Use Jenkins blue links in Gerrit" [Undecided,In progress] https://launchpad.net/bugs/961395 | 17:19 |
| praefect | hey guys, I have 2 environments down currently, just updated some packages (openstack on centos) only openstack-nova novaclient, stuff like that, now I can't launch VMs on both clouds! it all fails with image_size exceeded allowed size.. | 17:19 |
| praefect | the allowed size is 10GB I don't know where that is coming from! | 17:19 |
| sjhon | alexn6, I'm using 2^13 and supposing 50 drives maximum. | 17:19 |
| praefect | dammit.. MAXIMUM_ROOT_SIZE... just noticed that from the nova-compute log! | 17:20 |
| *** Ard1t has quit IRC | 17:21 | |
| Kyle__ | souza: Have you watched the logs while starting it, seeing what it complains about? | 17:21 |
| praefect | it's minimum_root_size | 17:21 |
| praefect | ;( | 17:21 |
| *** bepernoot has joined #openstack | 17:21 | |
| sjhon | alexn6, many thanks. | 17:21 |
| alexn6 | sjhon, 50/3=17 is between 2^4 and 2^5 - i.e max segment size 2^35/2^13=2^23 - 8Mb | 17:22 |
| *** clauden_ has quit IRC | 17:22 | |
| *** clauden has joined #openstack | 17:22 | |
| sjhon | alexn6, it it good enough ? | 17:23 |
| *** Ard1t has joined #openstack | 17:23 | |
| souza | Kyle__: yeah, but have any change in it! :/ | 17:23 |
| *** qazwsx has joined #openstack | 17:24 | |
| Kyle__ | souza: it's not logging anything? Then change your log level to verbose or debug in your glance settings and try again. | 17:24 |
| *** asavu_ has joined #openstack | 17:24 | |
| alexn6 | sjhon, think of it like if you choose very small partition power and enlarge your disk cluster too much - there will be too large segements, and rebuild all the ring from scratch os not so easy if it full of data | 17:24 |
| alexn6 | sjhon, I have power of 18 for 4x1Tb only - and this is not good, not bad. | 17:26 |
| uvirtbot | New bug: #961412 in keystone "endpoint-create fails" [Undecided,New] https://launchpad.net/bugs/961412 | 17:26 |
| *** gohko_nao has quit IRC | 17:26 | |
| *** gohko_nao has joined #openstack | 17:27 | |
| *** bepernoot has quit IRC | 17:27 | |
| sjhon | alexn6, is better to have a bigger power ? | 17:27 |
| alexn6 | sjhon, as for my own case mb better smaller power :) | 17:28 |
| *** shaon has quit IRC | 17:28 | |
| sjhon | alexn6, 13 is ok I suppose :) | 17:29 |
| alexn6 | sjhon, and I give you examples and what it all means above | 17:29 |
| *** popux has quit IRC | 17:29 | |
| alexn6 | sjhon, I see - you like this very number - 13 ;) | 17:30 |
| *** chasing`Sol has joined #openstack | 17:30 | |
| sjhon | :) No I used it to calculate, but I'm not sure it is good enough | 17:30 |
| *** guigui3 has quit IRC | 17:30 | |
| alexn6 | sjhon, in the end of this page they say about history of it all http://swift.openstack.org/overview_ring.html | 17:31 |
| *** clopez has quit IRC | 17:32 | |
| sjhon | thanks :) | 17:32 |
| *** Ryan_Lane has joined #openstack | 17:32 | |
| *** Ryan_Lane has quit IRC | 17:32 | |
| *** Ryan_Lane has joined #openstack | 17:32 | |
| *** jj0hns0n has joined #openstack | 17:36 | |
| souza | Kyle__: i've changed the log level, but it was not modified. | 17:38 |
| sjhon | alexn6, the documentation is not so clear. | 17:38 |
| alexn6 | sjhon, :) docs is quite good, though rather complex to understand | 17:39 |
| *** Guest70213 is now known as tobin | 17:39 | |
| *** llang629 has joined #openstack | 17:39 | |
| *** alrs has joined #openstack | 17:40 | |
| alexn6 | sjhon, they say something like - there is no statistic and comparison of usage different values, but we give you our thought about that | 17:40 |
| koolhead11 | alexn6: it would be complex for those who come with 5 mins cloud ready without knowing building blocks of cloud i suppose :) | 17:41 |
| alexn6 | sjhon, mb when notmyname come back, he tell you smth more concrete | 17:41 |
| *** llang629 has left #openstack | 17:41 | |
| alexn6 | koolhead11, :))) | 17:41 |
| sjhon | alexn6, "50/3=17 is between 2^4 and 2^5 - i.e max segment size 2^35/2^13=2^23 - 8Mb". why did you use 2^35 and 2^13 ? | 17:41 |
| *** danpb has quit IRC | 17:42 | |
| alexn6 | sjhon, did you really hear me and read all previous messages? :) | 17:42 |
| koolhead11 | alexn6: i call them cloud hippies and i have seen many here :) | 17:42 |
| sjhon | alexn6, yes. I'm trying to understant everything now. | 17:43 |
| alexn6 | koolhead11, btw - what is in your swift? | 17:43 |
| sjhon | koolhead11, sorry for the basic questions. | 17:43 |
| koolhead11 | sjhon: no don`t worry. It was meant for you :) | 17:44 |
| sjhon | :) | 17:44 |
| koolhead11 | alexn6: this week am focusing on glance. :D | 17:44 |
| koolhead11 | swift nest week | 17:44 |
| alexn6 | sjhon, 35=1Tb+power of 5 for 17 of them | 17:44 |
| alexn6 | koolhead11, clear :) | 17:45 |
| * koolhead11 rushes home | 17:45 | |
| sjhon | alexn6, and 2^13 ? | 17:46 |
| *** clopez has joined #openstack | 17:48 | |
| *** alrs has quit IRC | 17:48 | |
| *** Ryan_Lane has quit IRC | 17:48 | |
| sjhon | ops, I got it :) thanks :_ | 17:48 |
| sjhon | :) | 17:48 |
| alexn6 | sjhon, you say it by yourself :) | 17:49 |
| *** alrs has joined #openstack | 17:49 | |
| sjhon | alexn6, many thanks and sorry for these questions. | 17:49 |
| Kyle__ | souza: Then you need to start watching other log files while you start glance. Until you see one that's throwing an error. | 17:49 |
| *** zigo has quit IRC | 17:49 | |
| alexn6 | sjhon, well, I leaving this topic now. Good conf ;) | 17:49 |
| *** koolhead11 has quit IRC | 17:50 | |
| *** dpippenger has joined #openstack | 17:50 | |
| *** rodlabs has joined #openstack | 17:54 | |
| *** dneary has quit IRC | 17:55 | |
| rodlabs | anyone got a sec for a swift noob question? | 17:55 |
| notmyname | sjhon: alexn6: I don't think your explanation of the swift partitions is accurate | 17:55 |
| *** scotticus has joined #openstack | 17:56 | |
| alexn6 | notmyname, will listen you carefully ;) | 17:56 |
| *** asadpanda has quit IRC | 17:56 | |
| *** asadpanda has joined #openstack | 17:56 | |
| sjhon | notmyname. Please, go ahead :) | 17:56 |
| notmyname | sjhon: alexn6: the "partitions" in the swift ring have nothing to do with the size of data you can store in the cluster | 17:57 |
| *** llang629_ has joined #openstack | 17:57 | |
| notmyname | sjhon: swift manages storage by distributing the data throughout the cluster based on the has of it's name | 17:57 |
| *** catintheroof has quit IRC | 17:58 | |
| notmyname | sjhon: the hash function used (md5) gives the entire keyspace for placement | 17:58 |
| notmyname | sjhon: the swift "partitions" are segments of that keyspace | 17:58 |
| sjhon | sjhon: swift manages storage by distributing the data throughout | 17:58 |
| alexn6 | notmyname, I didn`t say "can store", I said minimal block of data, which swift would operate | 17:58 |
| *** llang629_ has left #openstack | 17:59 | |
| notmyname | alexn6: and while that is sort of correct, it's not the whole picture (and IMO confuses the issue a little) | 17:59 |
| sjhon | notmyname, how to calculate the number of partitions ? | 17:59 |
| alexn6 | notmyname, mb it is :) say him the answer | 18:00 |
| sjhon | notmyname, alexn6 already explained me | 18:01 |
| notmyname | sjhon: swift balances the data by ensuring all the storage volumes have the right number of swift partitions on them (in this way, alexn6's explanation is ok) | 18:01 |
| *** joearnol_ has joined #openstack | 18:01 | |
| notmyname | sjhon: so we recommend that you target about 100 partitions per drive (assuming you have all the same drive sizes) in a full cluster | 18:02 |
| notmyname | sjhon: that allows you to control roughly 1% of the storage on each storage volume by adjusting the weight | 18:02 |
| *** joesavak has joined #openstack | 18:03 | |
| sjhon | notmyname, if I have drives of diferent sizes. How can I define this ? | 18:03 |
| notmyname | sjhon: so since the number of partitions in the cluster is set at cluster creation time, you have to determine the maximum size of your cluster up front (based on space, budget, etc) | 18:03 |
| notmyname | sjhon: ok, I'll get to different sized drives :-) | 18:04 |
| alexn6 | notmyname, is that mean that for 8x1Tb drives 800=2^10 and partition power of 10 well? | 18:04 |
| *** jsavak has quit IRC | 18:04 | |
| *** armaan has joined #openstack | 18:04 | |
| *** joearnold has quit IRC | 18:04 | |
| notmyname | so if your max cluster size is 8 drives, then you need 800 partitions. so use a ring power of 10 (for 1024 total partitions) | 18:04 |
| notmyname | sjhon: FWIW, that's really small and I wouldn't recommend that | 18:05 |
| alexn6 | and partition will operate max block of 8Mb? | 18:05 |
| notmyname | no | 18:05 |
| sjhon | notmyname. I suppose I will have maximum of 50 drives. I'm using 2^13 | 18:05 |
| souza | MrHeat: Alexandre? | 18:05 |
| notmyname | sjhon: the balance to maintain is max cluster size against the time it takes to do ring operations (add/remove drives, rebalance, etc) | 18:06 |
| *** robix has left #openstack | 18:06 | |
| notmyname | and the size in memory the ring uses | 18:06 |
| notmyname | so that's why you don't want to set it to some big number up front | 18:06 |
| *** aliguori has quit IRC | 18:06 | |
| notmyname | I think all the docs (and the defaults) set the ring power to 18. this will give you a moderate-sized cluster and still be fast enough | 18:07 |
| sjhon | ok | 18:07 |
| sjhon | sjhon: the balance to maintain is max cluster size against the | 18:07 |
| sjhon | time it takes to do ring operations (add/remove drives, rebalance, | 18:07 |
| sjhon | etc) | 18:07 |
| alexn6 | do you think power of 18 bad for 4x1Tb of 4 zones and 3 replicas? | 18:07 |
| *** joesavak has quit IRC | 18:08 | |
| *** johnpostlethwait has joined #openstack | 18:08 | |
| *** darraghb has quit IRC | 18:08 | |
| alexn6 | notmyname, do you think power of 18 bad for 4x1Tb of 4 zones and 3 replicas? | 18:08 |
| notmyname | alexn6: swift doesn't balance the cluster based on data stored. it balances by the number of objects in each partition. as clusters get larger, this reduces the variance between the partitions | 18:08 |
| notmyname | alexn6: is that your max cluster size or just the current size? | 18:09 |
| *** jj0hns0n has quit IRC | 18:09 | |
| alexn6 | notmyname, current, but will not change very soon or very dramatically | 18:09 |
| sjhon | notmyname. I'm planning to use 4 zones, 3 replicas; 8 drives of 500GB each and 1 drive of 5TB. Can I use 2 ^18 for this ? | 18:10 |
| notmyname | alexn6: a higher partition power will make the ring files larger and make the ring operations take longer. it will also make the cluster store more metadata (more inodes on disk for directory trees) | 18:11 |
| *** armaan has quit IRC | 18:11 | |
| alexn6 | notmyname, so what is incoorect in saying that partition power define kinda smallest block for separate I/O? | 18:11 |
| notmyname | sjhon: ah yes. different drive types | 18:11 |
| *** armaan has joined #openstack | 18:11 | |
| uvirtbot | New bug: #961451 in nova "ImageNotFound until euca-describe-images" [Undecided,New] https://launchpad.net/bugs/961451 | 18:11 |
| *** derekh has quit IRC | 18:11 | |
| notmyname | alexn6: because swift still operates on the actual object stored in the cluster. it only balances the cluster based on the partitions | 18:11 |
| sjhon | notmyname, I need to use the "weights" for balance the cluster ? | 18:12 |
| notmyname | sjhon: yes. the weights are dimentionless numbers that only make sense relative to one another | 18:12 |
| *** arBmind has quit IRC | 18:13 | |
| notmyname | sjhon: so we recommend that you set the weight to the number of GB (eg 2000 = 2TB, etc) | 18:13 |
| *** rkukura has quit IRC | 18:13 | |
| sjhon | notmyname, in the setup that described, I would have 100 partitions per drive ? | 18:13 |
| *** GheRivero_ has quit IRC | 18:14 | |
| notmyname | sjhon: no, because of the different weights, you would have a different number of partitons. however, you would have roughly the same fullness of each drive (eg all drives about 40% full) | 18:14 |
| *** wabat has joined #openstack | 18:15 | |
| notmyname | where 40% is whatever fullness you have | 18:15 |
| sjhon | ahh, ok for the bigger drive, I need more partitions. So, I can apply more weights for the bigger one | 18:16 |
| notmyname | right | 18:16 |
| notmyname | sjhon: also the weights can be used to drain or fill a drive (or set of drives) gradually | 18:17 |
| *** koolhead17|away has joined #openstack | 18:17 | |
| notmyname | sjhon: generally you should try to keep your zones equally sized. so you either add equally to all zones or add another zone of equal size | 18:18 |
| notmyname | sjhon: this will give you the best space utilization in your cluster | 18:18 |
| *** jaypipes has joined #openstack | 18:19 | |
| *** bhall has joined #openstack | 18:19 | |
| *** bhall has joined #openstack | 18:19 | |
| notmyname | alexn6: make sense why I don't like saying that the partition power doesn't define a chunk of data that swift deals with? | 18:19 |
| sjhon | notmyname, for tests, can I use virtual machines with 500GB HD each of them, where each machine would be a drive ? | 18:19 |
| notmyname | sjhon: ya. (with the normal caveats of testing a disk IO system in a VM). my dev environment is a single VM that uses loopback devices to simulate drives | 18:21 |
| *** jayfordddd has joined #openstack | 18:22 | |
| sjhon | notmyname, one last question. What is the definition of partition in the swift ? | 18:23 |
| *** litejk has joined #openstack | 18:23 | |
| *** litejk has left #openstack | 18:24 | |
| notmyname | sjhon: hmm...I thought I just explained all that ;-) | 18:24 |
| notmyname | sjhon: what part is unclear? | 18:24 |
| *** bepernoot has joined #openstack | 18:25 | |
| sjhon | The first time a read about it I imagined that it was like a partition. Then alenx6 said me "it`s a swift term, which just a "group of chunks or data or files" | 18:25 |
| *** joearnol_ has quit IRC | 18:27 | |
| *** msavy has quit IRC | 18:27 | |
| *** bepernoot has quit IRC | 18:27 | |
| notmyname | sjhon: it's a segment of the md5 keyspace. the more partitions, the smaller the pieces. the less partitions, the larger the pieces | 18:27 |
| *** dolphm_ has quit IRC | 18:28 | |
| *** joearnold has joined #openstack | 18:28 | |
| *** My1 has joined #openstack | 18:28 | |
| *** My1 has left #openstack | 18:28 | |
| sjhon | notmyname, but it groups as alexn6 said ? | 18:29 |
| *** heckj has quit IRC | 18:29 | |
| *** refactor has quit IRC | 18:29 | |
| *** dieterdemeyer has joined #openstack | 18:29 | |
| *** My1 has joined #openstack | 18:30 | |
| *** My1 has left #openstack | 18:30 | |
| *** rubiojr has joined #openstack | 18:31 | |
| *** koolhead17|away is now known as koolhead17 | 18:32 | |
| *** jct has quit IRC | 18:32 | |
| alexn6 | sjhon, sorry for misguidance | 18:33 |
| notmyname | sjhon: I don't like either "yes" or "no" for that. It does group things in some sense, but a simple "yes" doesn't seem to cover how it works | 18:33 |
| *** lvaughn has quit IRC | 18:33 | |
| sjhon | alexn6, don't worry | 18:33 |
| notmyname | alexn6: I'm happy you jumped in to answer earlier. thanks :-) | 18:33 |
| sjhon | thanks alexn6 | 18:34 |
| alexn6 | sjhon, notmyname it would be better if entropy would decrease and common sense increases only :) | 18:34 |
| notmyname | sjhon: alexn6: for a very technical intro to the swift ring, check out http://www.tlohg.com/p/building-consistent-hashing-ring.html | 18:35 |
| sjhon | thanks :) | 18:35 |
| *** mnewby has joined #openstack | 18:37 | |
| alexn6 | notmyname, kinda understand what you say about weight and chunks, but I think only practice would help :) Last link very-very techincal :))) | 18:37 |
| *** joearnold has quit IRC | 18:38 | |
| notmyname | yes it is :-) | 18:38 |
| *** dachary has quit IRC | 18:38 | |
| *** n0ano has joined #openstack | 18:40 | |
| *** dolphm has joined #openstack | 18:40 | |
| alexn6 | notmyname, thanks for great story! | 18:40 |
| sjhon | notmyname. Many thanks. | 18:41 |
| *** alrs has quit IRC | 18:41 | |
| *** alrs has joined #openstack | 18:41 | |
| *** jj0hns0n has joined #openstack | 18:41 | |
| *** eirikbb has joined #openstack | 18:42 | |
| *** armaan has left #openstack | 18:42 | |
| *** alexn6 has left #openstack | 18:42 | |
| *** juergh has quit IRC | 18:46 | |
| *** albert23 has joined #openstack | 18:46 | |
| *** fukushim_ has joined #openstack | 18:46 | |
| *** fikus-kukis^TP1 has joined #openstack | 18:46 | |
| *** arBmind has joined #openstack | 18:48 | |
| *** fukushima has quit IRC | 18:48 | |
| *** villep has quit IRC | 18:48 | |
| *** fikus-kukis^TP has quit IRC | 18:48 | |
| *** fikus-kukis^TP1 is now known as fikus-kukis^TP | 18:48 | |
| *** fesnel has quit IRC | 18:48 | |
| *** fesnel has joined #openstack | 18:48 | |
| *** Deathvalley122 has quit IRC | 18:49 | |
| *** villep has joined #openstack | 18:50 | |
| *** deshantm has quit IRC | 18:50 | |
| *** Deathvalley122 has joined #openstack | 18:51 | |
| *** aliguori has joined #openstack | 18:51 | |
| *** dolphm has quit IRC | 18:52 | |
| *** heckj has joined #openstack | 18:53 | |
| *** asavu_ has quit IRC | 18:53 | |
| *** Kiall has quit IRC | 18:54 | |
| *** epim has joined #openstack | 18:54 | |
| *** clopez has quit IRC | 18:59 | |
| *** dachary has joined #openstack | 19:00 | |
| *** jct has joined #openstack | 19:00 | |
| *** MrHeat has quit IRC | 19:01 | |
| *** Kiall has joined #openstack | 19:01 | |
| *** po has quit IRC | 19:01 | |
| *** lctmz has quit IRC | 19:03 | |
| *** Ryan_Lane has joined #openstack | 19:04 | |
| manu-db | noob question: Installing OpenStack for the first time, single machine configuration, RedHat 6.2-based install, followed OpenStack tutorial as much as I could... I'm now attempting to install a VM image using the euca2ools stuff: euca-upload-bundle -b kernels -m /tmp/centos.5-3.x86-64.img.manifest.xml | 19:04 |
| manu-db | however, I'm getting an exception from the server, I believe: File "/usr/lib/python2.6/site-packages/nova/objectstore/s3server.py", line 128, in __call__ | 19:05 |
| manu-db | f(**params) --- TypeError: get() got an unexpected keyword argument 'action' | 19:05 |
| manu-db | So, seems like the Nova objectstore is up... but perhaps misconfigured? Is there a command I can run to check to see if the basic config stuff is correct for the nova objectstore? | 19:06 |
| melmoth | manu-db: noob too, but 1) what about looking in /var/log/nova/nova-* ? | 19:06 |
| melmoth | with essex, i never managed to upload images with euca tool, but it worked with glance, you may want to try it this way if you are using essex | 19:07 |
| *** dtroyer is now known as dtroyer_zzz | 19:07 | |
| *** mrmartin has joined #openstack | 19:07 | |
| manu-db | got a link to how to upload w/ glance? | 19:08 |
| *** oneiroi has joined #openstack | 19:08 | |
| melmoth | http://glance.openstack.org/glance.html | 19:08 |
| manu-db | thx, melmoth | 19:09 |
| *** po has joined #openstack | 19:09 | |
| melmoth | my own notes: http://pastebin.com/tBacvap5 | 19:09 |
| uvirtbot | New bug: #961507 in nova "xenapi tries to shutdown rescue instance even if it's already shutdown" [Undecided,In progress] https://launchpad.net/bugs/961507 | 19:11 |
| uvirtbot | New bug: #961509 in horizon "horizon.js is overloaded" [Low,Confirmed] https://launchpad.net/bugs/961509 | 19:11 |
| *** clopez has joined #openstack | 19:13 | |
| *** dbuzz has joined #openstack | 19:15 | |
| *** manlymatt83 has left #openstack | 19:16 | |
| *** dbuzz has quit IRC | 19:18 | |
| *** dbuzz has joined #openstack | 19:18 | |
| *** dtroyer_zzz is now known as dtroyer | 19:19 | |
| uvirtbot | New bug: #961524 in nova "xenapi cannot unplug VBDs when rescue instance is halted" [Undecided,In progress] https://launchpad.net/bugs/961524 | 19:20 |
| *** oneiroi has quit IRC | 19:22 | |
| *** dbuzz is now known as oneiroi | 19:22 | |
| *** llang629_ has joined #openstack | 19:23 | |
| *** llang629_ is now known as llang629 | 19:23 | |
| *** llang629 has left #openstack | 19:24 | |
| *** oneiroi is now known as Oneiroi | 19:24 | |
| *** dolphm has joined #openstack | 19:25 | |
| *** Free_maN has quit IRC | 19:26 | |
| *** rkukura has joined #openstack | 19:26 | |
| *** jayfordddd has quit IRC | 19:27 | |
| *** Maroo has joined #openstack | 19:28 | |
| *** lborda has quit IRC | 19:29 | |
| Maroo | hi | 19:29 |
| *** lxu has quit IRC | 19:30 | |
| *** bepernoot has joined #openstack | 19:30 | |
| Maroo | i'm getting 401 unauthorized with the command curl -v -H 'X-Auth-Token: <x-auth-token>' <url-x-storage> | 19:30 |
| *** shevek_ has joined #openstack | 19:31 | |
| Maroo | any solution please ? | 19:31 |
| *** vitiho has quit IRC | 19:31 | |
| *** robix has joined #openstack | 19:33 | |
| *** bepernoot has quit IRC | 19:33 | |
| *** lborda has joined #openstack | 19:33 | |
| *** natea_ has joined #openstack | 19:34 | |
| robix | What the best way to clean out 'nova-manage vm list' entires that are stuck in building for days? | 19:35 |
| kbringard | if terminating them doesn't do it then you'll need to go into the db | 19:36 |
| *** ppradhan has left #openstack | 19:36 | |
| *** natea has quit IRC | 19:37 | |
| *** natea_ is now known as natea | 19:37 | |
| robix | ok, that's what I figured. thanks. | 19:38 |
| notmyname | any chance florian haas is in here? | 19:39 |
| *** mdrnstm has joined #openstack | 19:40 | |
| *** nOStahl has quit IRC | 19:44 | |
| *** sZoz has quit IRC | 19:45 | |
| uvirtbot | New bug: #961545 in keystone ""ValueError: unsupported format character" in catalog" [Undecided,New] https://launchpad.net/bugs/961545 | 19:45 |
| uvirtbot | New bug: #961547 in swift "Containers should expose change-indicator" [Undecided,New] https://launchpad.net/bugs/961547 | 19:46 |
| *** souza has quit IRC | 19:48 | |
| *** asavu has joined #openstack | 19:50 | |
| *** GheRivero_ has joined #openstack | 19:52 | |
| *** jastr has quit IRC | 19:53 | |
| *** rnorwood has quit IRC | 19:56 | |
| *** msavy has joined #openstack | 19:56 | |
| *** rnorwood has joined #openstack | 19:59 | |
| *** Maroo has quit IRC | 20:00 | |
| *** natea has quit IRC | 20:00 | |
| *** RichardRaseley has joined #openstack | 20:00 | |
| *** vincentricci has quit IRC | 20:01 | |
| *** clopez has quit IRC | 20:02 | |
| *** natea has joined #openstack | 20:02 | |
| *** dwcramer has joined #openstack | 20:03 | |
| RichardRaseley | Hello all. I am interested in finding out more about using Ubuntu Server & OpenStack for creating a scalable infrastructure for hosting VMs (combination of *nix and Windows). I am having trouble understanding how the Ubuntu OS, OpenStack, and the Hypervisor sit in relation to each other. Is there an "executive summary" or some like document that could help me understand better? | 20:04 |
| *** jastr has joined #openstack | 20:05 | |
| *** msavy has quit IRC | 20:11 | |
| uvirtbot | New bug: #961574 in glance "pip glance==2011.3.1 missing dependencies" [Undecided,New] https://launchpad.net/bugs/961574 | 20:12 |
| *** shevek_ has quit IRC | 20:12 | |
| *** MarkAtwood has joined #openstack | 20:12 | |
| *** msavy has joined #openstack | 20:13 | |
| *** koolhead17 has quit IRC | 20:14 | |
| *** clopez has joined #openstack | 20:16 | |
| kbringard | anyone remember the value to use when uploading an image with no kernel or ramdisk for kern_id | 20:17 |
| kbringard | kernel_id | 20:17 |
| kbringard | is it nokernel | 20:17 |
| kbringard | ? | 20:17 |
| *** vricci has joined #openstack | 20:19 | |
| *** sjhon has quit IRC | 20:20 | |
| Glacee | Anyone is running swift on Ubuntu precise? We are trying it and running into greenlet issues with swift-dispersion and swift-bench so far | 20:21 |
| uvirtbot | New bug: #961588 in nova "Fail-fast for invalid read_deleted values" [Medium,In progress] https://launchpad.net/bugs/961588 | 20:21 |
| uvirtbot | New bug: #961590 in glance "pip glance-2012.1~rc1 missing dependency iso8601" [Undecided,New] https://launchpad.net/bugs/961590 | 20:21 |
| Glacee | evenlet also | 20:22 |
| *** vitiho has joined #openstack | 20:23 | |
| *** warik_ has joined #openstack | 20:24 | |
| melmoth | RichardRaseley: i guess http://docs.openstack.org/diablo/openstack-compute/starter/content/Introduction_to_OpenStack_and_its_components-d1e59.html | 20:26 |
| RichardRaseley | melmoth: Thank you for the link. I have already read through that but it doesn't make some things clear. Does the Hypervisor sit on top of OpenStack or the other way around? | 20:28 |
| *** warik has quit IRC | 20:28 | |
| RichardRaseley | Or do they both run on top of the base operating system (i.e. Ubuntu Server). | 20:28 |
| melmoth | not sure i understand "on top", but i would say "below", you can use several hypervisor (though if i understand correctly, kvm is the only "supported" choice) | 20:28 |
| *** vricci is now known as vincentricci | 20:29 | |
| *** joearnold has joined #openstack | 20:29 | |
| melmoth | http://wiki.openstack.org/HypervisorSupportMatrix | 20:29 |
| *** miclorb_ has joined #openstack | 20:30 | |
| RichardRaseley | Hmm.. | 20:30 |
| *** dprince has quit IRC | 20:30 | |
| *** manlymatt83 has joined #openstack | 20:31 | |
| manlymatt83 | What's the link to the countdown of RC1 again? | 20:31 |
| RichardRaseley | So, imagine I have HP hardware. Upon that hardware I install Ubuntu Server. Upon the Ubuntu Server instances I install OpenStack. Correct? | 20:31 |
| melmoth | sounds correct to me. | 20:31 |
| RichardRaseley | I would assume that the Hypervisor is also installed atop the Ubuntu Server instance, not atop some abstraction layer created by OpenStack? | 20:32 |
| melmoth | correct. it s the same code that you woud use without openstack. | 20:32 |
| RichardRaseley | OK | 20:32 |
| melmoth | you actually need the hypervisor bit only on the "compute" nodes. | 20:32 |
| melmoth | the one you allocate to run vms. | 20:33 |
| RichardRaseley | So, in order to use (for example) Hyper-V (not saying I would) but you would run OpenStack on Windows? | 20:33 |
| melmoth | the openstack is "just" some python code that let you manage those vms easyly if you want. | 20:33 |
| *** natea has quit IRC | 20:33 | |
| melmoth | i m not sure hyper v is supported, but as far as i have understdood, the whole stack should be os agnostic. | 20:34 |
| melmoth | it s python code mostly (at least all the part i saw), there s no need to talk directly to the os with some kernel module. | 20:34 |
| RichardRaseley | OK, you're right. Hyper-V isn't currently supported. Let's take VMware ESXi for example. | 20:34 |
| melmoth | but my feeling is, it s already complicated enough not to try it out on something exotic :) | 20:34 |
| RichardRaseley | It runs "below" the OS (being a bare metal hypervisor). So, how would that work with OpenStack? | 20:35 |
| melmoth | hmmmm. i would not know how to fit it in the picture. | 20:35 |
| melmoth | i only used openstack with kvm , qemu and lxc. | 20:35 |
| RichardRaseley | Yeah, that is somewhat confusing to me. | 20:35 |
| RichardRaseley | OK | 20:35 |
| *** natea has joined #openstack | 20:37 | |
| RichardRaseley | So, my scenario would be as follows: 3 physical servers (16 Cores, 64GB of RAM, 1TB of RAID10 Storage). I could install Ubuntu Server on all 3 nodes, Install the OpenStack compute components along with KVM to allow me to host VMs and then install the OpenStack stroage component to create three different availability zones (one for each server) and use that space to store the VM configruation and virutal disk files? | 20:37 |
| *** dwcramer has quit IRC | 20:38 | |
| RichardRaseley | I would then be able to have a highly available clustered configuration that would allow me to move VMs between the nodes, and the data would be replicated between the nodes as well? | 20:38 |
| melmoth | i am assuming that by "storage component", you mean swift. | 20:39 |
| RichardRaseley | Yes | 20:39 |
| melmoth | you do not _need_ swift. But, usually, people use it for : 1) storing the images of the vms. | 20:39 |
| melmoth | read only images. that is, the bit you need to boot them up. | 20:39 |
| melmoth | where things are forgotten and lost when you shut them down | 20:39 |
| melmoth | 2) storing "file" | 20:39 |
| melmoth | but it s not ablock device. | 20:40 |
| melmoth | it s not a filesystem | 20:40 |
| melmoth | you can attach volumes ot your vms (proper block devices) if you want. but it s more costly in network (if i understand correctly) | 20:40 |
| melmoth | so , if your applications are not already using some swift storage, and you do not plan to use it, there is no need (except if you plan to host a _lot_ of vm images) | 20:41 |
| RichardRaseley | I do plan on hosting quite a few - but I am mainly looking to take advantage of the resiliancy of the swift system. | 20:41 |
| melmoth | you can use it, it s just one more thing to grok. | 20:42 |
| andrewbogott | Vek: Does that guideline about string encoding strike you as common knowledge or obscure? Wondering if I should/need to schedule a summit session to proselytize. | 20:42 |
| RichardRaseley | As opposed to throwing a $500K SAN behind a more traditional VMware cluster. | 20:42 |
| andrewbogott | oop, wrong room, sorry | 20:42 |
| melmoth | RichardRaseley: alos, as far as i remember, do _not_ use any raid on the swift nodes. | 20:43 |
| melmoth | let swift handle all redondancy. | 20:43 |
| RichardRaseley | OK, but I was just thinking that if I was using swift to store the virtual hard disks for my VMs, that I would need more performance than a single drive could offer. | 20:44 |
| RichardRaseley | BRB | 20:45 |
| melmoth | you do not use swift as harddrive, only as a container. | 20:45 |
| RichardRaseley | I guess that is the part that is confusing to me. So I have a physical device that gives me block level access (hard drive, RAID volume, LUN, etc.) | 20:48 |
| RichardRaseley | Upon that device I care out one or more logical volumes. | 20:48 |
| RichardRaseley | Those volumes are then formated with a file system (say ext4 - or whatever). | 20:48 |
| melmoth | you can use those as block device for your vms, but it s part of nova-volume, not swift. | 20:49 |
| melmoth | and the vms access them via iscsi | 20:49 |
| melmoth | if i understand correctly, the + aspect of swift is "high availablility/redondancy", the - one are it s slow (compare to iscsi), it s not a block device, you have bucket and that s about it. | 20:50 |
| melmoth | not proper directlry that can be embedded one into another | 20:51 |
| RichardRaseley | Hmm... | 20:51 |
| RichardRaseley | I appreciate your help - it is just somewhat difficult to understand how all the components work together (never actually having worked with OpenStack). | 20:51 |
| RichardRaseley | I will have to continue to investigate / learn about it. | 20:52 |
| Madkiss | can somebody point me to a good list of basic accounts that I should create in keystone? | 20:52 |
| melmoth | i have the same feeling since i have looked at it :) | 20:52 |
| RichardRaseley | Well, I appreciate your help. | 20:53 |
| *** dwcramer has joined #openstack | 20:53 | |
| dolphm | Madkiss: https://github.com/openstack-dev/devstack/blob/master/files/keystone_data.sh | 20:53 |
| Madkiss | dolphm: that's full of non-existing variables for those not using devstack ... | 20:54 |
| *** bruges_ has joined #openstack | 20:54 | |
| dolphm | Madkiss: non-existing? are there any values you can't determine? | 20:55 |
| Madkiss | and the only thing it actually tells me is "You must provide a username via either --username or env[OS_USERNAME]" plus a shitload of other error messages when being executed | 20:55 |
| dolphm | Madkiss: do you have an admin user in keystone yet? | 20:55 |
| *** imsplitbit has quit IRC | 20:55 | |
| uvirtbot | New bug: #961624 in horizon "Billing feature" [Undecided,New] https://launchpad.net/bugs/961624 | 20:56 |
| dolphm | Madkiss: there's two auth methods for the client at the moment, described here http://keystone.openstack.org/configuration.html#adding-users-tenants-and-roles-with-python-keystoneclient | 20:56 |
| Madkiss | dolphm: i have an admin token set in keystone.conf, apart from that, this setup is blank | 20:56 |
| *** clopez has quit IRC | 20:56 | |
| *** bruges has quit IRC | 20:57 | |
| dolphm | Madkiss: with an admin token, follow the "Authenticating with a Token" section | 20:57 |
| Madkiss | w00t | 20:57 |
| Madkiss | dolphm: thanks a lot. sorry, it's been a long day ... :) | 20:57 |
| dolphm | Madkiss: there's a change on it's way that will make those error messages you're seeing much more helpful https://review.openstack.org/#change,5344 | 20:57 |
| *** dieterdemeyer has left #openstack | 20:57 | |
| Madkiss | root@netbook:~# ./keystone_data.sh | 20:58 |
| Madkiss | root@netbook:~# echo $? | 20:58 |
| Madkiss | 0 | 20:58 |
| Madkiss | you just made me the happiest man on earth. :) | 20:58 |
| uvirtbot | New bug: #961621 in horizon "Automatically name the file according to its original name with the extension" [Undecided,New] https://launchpad.net/bugs/961621 | 20:58 |
| dolphm | Madkiss: =) anytime | 20:58 |
| *** jdg has joined #openstack | 20:58 | |
| Madkiss | dolphm: do you have a document indicating how to connect glance to this? | 20:59 |
| dolphm | Madkiss: i think that's in the glance docs somewhere | 20:59 |
| Madkiss | let's see | 20:59 |
| Madkiss | ah yes | 21:00 |
| dolphm | Madkiss: this may be helpful, but i've heard its out of date, and doesn't cover glance much http://keystone.openstack.org/configuringservices.html | 21:00 |
| dolphm | Madkiss: we'll be hitting docs hard after keystone makes rc1 | 21:00 |
| Madkiss | I wonder whether keystone is flawlessly integrated into all OpenStack services now | 21:00 |
| dolphm | Madkiss: in theory :) | 21:00 |
| Madkiss | last time I looked one couldn't connect swift to it | 21:00 |
| *** njoy has joined #openstack | 21:00 | |
| Madkiss | OTOH, I plan to use RADOS/Ceph in this anyway | 21:01 |
| dolphm | Madkiss: are you looking at this? http://glance.openstack.org/authentication.html | 21:01 |
| Madkiss | dolphm: yes, I think that's gonna help :) | 21:01 |
| dolphm | Madkiss: open lots of bugs if you run into issues :) | 21:01 |
| Madkiss | i'll take that as an appea | 21:01 |
| Madkiss | l | 21:01 |
| *** primeministerp has quit IRC | 21:02 | |
| *** anotherjesse_zz is now known as anotherjesse | 21:04 | |
| *** mrmartin has quit IRC | 21:05 | |
| *** njoy has quit IRC | 21:06 | |
| *** anotherjesse is now known as anotherjesse_zz | 21:07 | |
| *** littleidea has joined #openstack | 21:07 | |
| *** anotherjesse_zz is now known as anotherjesse | 21:08 | |
| *** lts has quit IRC | 21:10 | |
| *** bepernoot has joined #openstack | 21:10 | |
| uvirtbot | New bug: #961637 in swift "Swift-Dispersion-Report KeyError in threading" [Undecided,New] https://launchpad.net/bugs/961637 | 21:10 |
| *** kayyagari has quit IRC | 21:11 | |
| *** andrewsmedina has quit IRC | 21:11 | |
| *** andrewsmedina has joined #openstack | 21:13 | |
| *** marrusl_ has joined #openstack | 21:14 | |
| uvirtbot | New bug: #961640 in swift "Swift-Bench Error3 on Ubuntu Precise" [Undecided,New] https://launchpad.net/bugs/961640 | 21:16 |
| *** bepernoot has quit IRC | 21:17 | |
| Madkiss | ahu. | 21:18 |
| Madkiss | The request returned 503 Service Unavilable. This generally occurs on service overload or other transient outage.. | 21:18 |
| *** jdg has quit IRC | 21:19 | |
| *** stuntmachine has quit IRC | 21:19 | |
| *** jdg has joined #openstack | 21:19 | |
| *** jastr has quit IRC | 21:20 | |
| *** Oneiroi is now known as Oneiroi|Zzzz | 21:20 | |
| *** msavy has quit IRC | 21:21 | |
| *** ldlework has quit IRC | 21:21 | |
| Madkiss | CRITICAL [keystone.middleware.auth_token] Unable to obtain admin token: invalid json response | 21:22 |
| Madkiss | ah, great | 21:22 |
| Madkiss | dolphm: help!! ;) | 21:22 |
| manlymatt83 | What's the link to the countdown of RC1 for each project again? | 21:23 |
| *** melmoth has quit IRC | 21:24 | |
| dolphm | manlymatt83: it moved to http://wiki.openstack.org/releasestatus/rc.html | 21:24 |
| manlymatt83 | thanks | 21:25 |
| *** swill has joined #openstack | 21:25 | |
| manlymatt83 | dolphm: Do you know if there are Ubuntu packages for RC1? | 21:25 |
| manlymatt83 | (Nova) | 21:26 |
| swill | does anyone know why accessing swift with curl works perfectly, but using the swift command returns [Errno 111] ECONNREFUSED? | 21:26 |
| dolphm | manlymatt83: i believe they're still in the works, but i'm sure someone else could give a better status | 21:26 |
| dolphm | swill: maybe keystone is not running? | 21:27 |
| swill | i am not using keystone. but then curl would not work either. | 21:27 |
| *** tmichael has joined #openstack | 21:27 | |
| swill | dolphm: i wrote my own auth middleware | 21:27 |
| dolphm | the CLI client may be trying to authenticate you against keystone | 21:27 |
| *** tmichael has quit IRC | 21:28 | |
| Madkiss | hm. I wonder whether I got my OS_environment setup right | 21:28 |
| swill | dolphm: wouldn't it use the auth configured in my proxy conf? | 21:28 |
| Madkiss | OS_TOKEN ought to be the service token as specified in keystone.conf, right? | 21:28 |
| *** tmichael has joined #openstack | 21:28 | |
| Madkiss | and OS_USERNAME is the username I used for the admin account, in this case "admin"? | 21:28 |
| manlymatt83 | dolphm: thanks :) | 21:28 |
| dolphm | swill: not something i've tested :/ but i would assume so | 21:28 |
| dolphm | Madkiss: yes | 21:29 |
| dolphm | Madkiss: if you're using OS_TOKEN and OS_ENDPOINT, then the following aren't required: OS_USERNAME, OS_PASSWORD, OS_AUTH_URL | 21:29 |
| Madkiss | OS_ENDPOINT? I don't even have that ;) | 21:29 |
| swill | my error: http://pastebin.com/FC89BXWa | 21:29 |
| notmyname | swill: just a guess--the auth step is returning something unexpected | 21:30 |
| notmyname | swill: perhaps http when you are expecting https (or vice versa) | 21:30 |
| swill | notmyname: ok, i will look into that. thanks. | 21:30 |
| Madkiss | dolphm: what about OS_TENANT_NAME? :) | 21:30 |
| *** oNeToWn has quit IRC | 21:31 | |
| swill | notmyname: i would expect curl to also fail then. | 21:31 |
| dolphm | swill: it looks like swift is trying to validate your token against a remote service (i.e. keystone) | 21:31 |
| swill | curl is working perfectly | 21:31 |
| swill | dolphm: yes, cloudstack | 21:31 |
| dolphm | Madkiss: only required if you not using OS_TOKEN & OS_ENDPOINT | 21:31 |
| Glacee | notmyname: are there known issues with greenlet/evenlet on precise? Having some errors with dispersion and swift-bench so far | 21:31 |
| notmyname | swill: except that with curl you have a 2 step manual process (one auth and one access) | 21:31 |
| dolphm | Madkiss: OS_ENDPOINT should be http://keystone_host:35357/v2.0/ | 21:31 |
| swill | notmyname: yes | 21:32 |
| Madkiss | okay | 21:32 |
| Madkiss | dolphm: and it will assume that the username is "admin"? | 21:32 |
| notmyname | Glacee: I don't know. I haven't used precise. however the bug reports are somewhat concerning | 21:32 |
| dolphm | Madkiss: you don't have a username yet, so no need to set OS_USERNAME (despite the misleading error message) | 21:32 |
| Glacee | notmyname: hehe.. I hope I do not have to reinstall on lucid lynx :( | 21:33 |
| dolphm | Madkiss: in fact, if you update your keystoneclient, that error will now be much different :) | 21:33 |
| dolphm | Madkiss: as of a few minutes ago | 21:33 |
| Madkiss | ah | 21:33 |
| Madkiss | Okay. Let's stick with the OS_AUTH_URL-stuff for now. I set these variables correctly, and now I get "Response from Keystone does not contain a Glance endpoint." | 21:34 |
| *** bbcmicrocomputer has quit IRC | 21:34 | |
| Madkiss | which appears to be true, as keystone_data.sh from devsteck doesn't define one | 21:34 |
| Madkiss | methnks | 21:34 |
| dolphm | Madkiss: you *must* use OS_TOKEN and OS_ENDPOINT to create a real admin user before you can use OS_USERNAME, OS_PASSWORD, OS_TENANT_* and OS_AUTH_URL | 21:34 |
| *** jastr has joined #openstack | 21:34 | |
| Madkiss | dolphm: in keystone, you mean? | 21:35 |
| *** mattstep has quit IRC | 21:35 | |
| dolphm | Madkiss: for keystoneclient, yes | 21:35 |
| *** mattstep has joined #openstack | 21:35 | |
| Madkiss | okay, so I am one step behnd apparently. I ran keystone_data.sh and it returned 0 | 21:36 |
| dolphm | Madkiss: ah, i'm one step behind you then.. | 21:37 |
| dolphm | Madkiss: it creates an admin user & admin tenant for you, so you don't have to use OS_TOKEN and OS_ENDPOINT | 21:37 |
| Madkiss | I seem to recall that earlier versions of keystone_data.sh actually added endpoints for glance and the like to keystone | 21:37 |
| dolphm | Madkiss: by default then... | 21:37 |
| Madkiss | dolphm: thanks for the heads up, but it looks like I am missing an endpoint in keystone for Glance | 21:38 |
| dolphm | Madkiss: you're right; endpoints are managed differently now! | 21:38 |
| Madkiss | oh wonderful | 21:38 |
| dolphm | Madkiss: look in your keystone.conf... | 21:38 |
| dolphm | Madkiss: it's easier :) | 21:38 |
| Madkiss | apparently 140% of this have changed since december | 21:38 |
| dolphm | Madkiss: that's accurate | 21:39 |
| dolphm | Madkiss: look for the [catalog] section in keystone.conf | 21:39 |
| Madkiss | my keystone.conf doesn't include a reference to endpoints ;) | 21:39 |
| Madkiss | okay | 21:39 |
| dolphm | Madkiss: i assume it reads TemplatedCatalog backend? | 21:39 |
| Madkiss | no | 21:39 |
| Madkiss | driver = keystone.catalog.backends.sql.Catalog | 21:39 |
| dolphm | oh boo | 21:39 |
| dolphm | Madkiss: is this devstack? lol | 21:39 |
| *** lloydde has joined #openstack | 21:39 | |
| Madkiss | it's ubuntu 12.04 default packages | 21:40 |
| Madkiss | according to the package name, it's exactly 5 days old. | 21:40 |
| lloydde | where does python-novaclient live today? | 21:41 |
| Madkiss | okay | 21:41 |
| Madkiss | so I need TemplatedCatalog | 21:41 |
| dolphm | hmm, i thought we decided that driver was undocumented for essex, but anyway... | 21:41 |
| *** GheRivero_ has quit IRC | 21:41 | |
| dolphm | Madkiss: yes, i'd recommend TemplatedCatalog, and asking ubuntu wtf | 21:41 |
| Madkiss | And TemplatedCatalog is documented ... nowhere? :) | 21:41 |
| *** andrewsmedina has quit IRC | 21:42 | |
| dolphm | Madkiss: sql.Catalog is undocumented for essex, TemplatedCatalog is documented pretty well | 21:42 |
| dolphm | Madkiss: http://keystone.openstack.org/configuration.html#migrating-your-service-catalog-from-legacy-versions-of-keystone | 21:42 |
| lloydde | launchpad seems to say https://github.com/openstack/python-novaclient | 21:42 |
| Madkiss | dolphm: I don't _have_ keystone legacy data ;) | 21:42 |
| lloydde | but the documentation is stale and other github repos are too | 21:42 |
| dolphm | Madkiss: looking for docs for fresh installs, if not, i'll walk you through then write some.. | 21:43 |
| Madkiss | Okay. I guess the file I am looking for is default_catalog.templates | 21:43 |
| dolphm | Madkiss: yep | 21:44 |
| Madkiss | and now I need to get this into Keystone somehow I guess | 21:44 |
| dolphm | Madkiss: your catalog section should look something liek this http://paste.openstack.org/raw/11762/ | 21:44 |
| Madkiss | via TemplateCatalog | 21:44 |
| Madkiss | no. | 21:45 |
| dolphm | Madkiss: copy paste that then :) | 21:45 |
| Madkiss | http://paste.debian.net/160568/ | 21:45 |
| Madkiss | oh, wait | 21:45 |
| Madkiss | you mean the section | 21:45 |
| Madkiss | sorry | 21:45 |
| dolphm | Madkiss: yeah, my bad, in keystone.conf | 21:46 |
| dolphm | the example catalog template you posted will work though | 21:46 |
| dolphm | Madkiss: assuming you're deploying everything to localhost, of course | 21:46 |
| Madkiss | cool, now I'm back to the original error message again | 21:47 |
| Madkiss | which is | 21:47 |
| Madkiss | "Unable to obtain admin token: invalid json response" | 21:47 |
| Madkiss | greatgodalmighty :) | 21:47 |
| *** medieval1 has quit IRC | 21:47 | |
| lloydde | anyone know how python-novaclient is versioned? | 21:47 |
| *** dcramer_ has joined #openstack | 21:48 | |
| *** dwcramer has quit IRC | 21:49 | |
| Madkiss | dolphm: any idea what could be wrong there? | 21:51 |
| *** lborda has quit IRC | 21:51 | |
| uvirtbot | New bug: #961672 in openstack-ci "Tackle LodgeIt Spam" [Critical,New] https://launchpad.net/bugs/961672 | 21:51 |
| Madkiss | The request returned 503 Service Unavilable. This generally occurs on service overload or other transient outage. | 21:51 |
| *** nOStahl has joined #openstack | 21:51 | |
| *** bcwaldon has joined #openstack | 21:51 | |
| dolphm | Madkiss: so, i'm supposed to be somewhere in 9 minutes :P ... 503 is not normal though! | 21:52 |
| Madkiss | okay | 21:52 |
| *** dspano has quit IRC | 21:52 | |
| Madkiss | I guess i'll just call it a day then | 21:52 |
| dolphm | Madkiss: we can pick this up tomorrow, or i'll be back in a couple hours | 21:53 |
| Madkiss | ok | 21:53 |
| Madkiss | no worries :) | 21:53 |
| dolphm | Madkiss: i think we'll need to dig into your logs though | 21:53 |
| *** jkyle has joined #openstack | 21:53 | |
| *** bcwaldon has quit IRC | 21:53 | |
| dolphm | Madkiss: not sure if fedora enabled keystone log files out of the box or not? | 21:53 |
| *** bcwaldon has joined #openstack | 21:53 | |
| Madkiss | this is ubuntu, and I enabvled them manually :) | 21:53 |
| jkyle | does openstack recommend using its own package repository for ubuntu installs or the packages in the standard repositories? Looking at the diablo manual, seems they used standard packages for nova and qemu* | 21:54 |
| *** medieval1 has joined #openstack | 21:54 | |
| dolphm | Madkiss: s/fedora/ubuntu | 21:55 |
| dolphm | Madkiss: bbl | 21:55 |
| *** dolphm has quit IRC | 21:55 | |
| Madkiss | bye | 21:56 |
| Madkiss | : | 21:56 |
| Madkiss | and thans | 21:56 |
| *** joearnold has quit IRC | 21:58 | |
| *** Ruetobas has quit IRC | 22:00 | |
| lloydde | vishy: who would I ask about how python-novaclient versioning works? how it corresponds to releases | 22:00 |
| lloydde | the release notes in the pypi package most recent list 2.5.8 (July 11, 2011) | 22:01 |
| swill | out of curiosity. is the reseller_prefix in the auth middleware required to be set to something other than an empty string? | 22:05 |
| *** warik has joined #openstack | 22:08 | |
| *** Ruetobas has joined #openstack | 22:11 | |
| *** flaviamissi has quit IRC | 22:11 | |
| *** swill has quit IRC | 22:12 | |
| *** dtroyer is now known as dtroyer_zzz | 22:14 | |
| *** joearnold has joined #openstack | 22:17 | |
| *** marrusl has quit IRC | 22:18 | |
| *** marrusl_ has quit IRC | 22:18 | |
| *** vitiho_ has joined #openstack | 22:20 | |
| *** vitiho has quit IRC | 22:20 | |
| *** natea has quit IRC | 22:20 | |
| *** natea has joined #openstack | 22:20 | |
| vishy | lloydde: there isn't an official versioning in place for the clients | 22:21 |
| lloydde | vishy: that adds spice to life. Will there be? | 22:21 |
| *** rods has quit IRC | 22:25 | |
| uvirtbot | New bug: #961697 in openstack-ci "tmpreaper is generating cronspam" [Undecided,New] https://launchpad.net/bugs/961697 | 22:26 |
| *** kbringard has quit IRC | 22:27 | |
| *** troytoman is now known as troytoman-away | 22:28 | |
| uvirtbot | New bug: #961706 in openstack-manuals "Difficult to figure out section context within PDF" [Undecided,New] https://launchpad.net/bugs/961706 | 22:31 |
| *** jce has joined #openstack | 22:33 | |
| *** po has quit IRC | 22:33 | |
| jce | is there a way to use NFS for VMs instead of LVM, etc? | 22:33 |
| cloudfly | i don't see why you couldn't just provide NFS mount points in your images | 22:35 |
| cloudfly | so when they fire up they access the NFS points they need | 22:35 |
| cloudfly | ala fstab | 22:35 |
| cloudfly | you might even be able to pass nfs path data dynamically via cloudinit | 22:37 |
| jce | but I can't have their images stored on nfs like I do with ESXi | 22:37 |
| cloudfly | well you could set up the image path that nova uses to be an NFS mount point | 22:37 |
| cloudfly | nova volume manager might still need a volume group | 22:37 |
| cloudfly | but i don't think the compute service will complain | 22:38 |
| cloudfly | that's addressing the compute nodes anyways | 22:39 |
| cloudfly | not sure about glance requirements | 22:39 |
| jce | ok... setting this all up is proving more difficult than anticipated. | 22:39 |
| cloudfly | it's still the early days. lots of stuff hasn't been sanded down yet. you can still cut yourself =P | 22:39 |
| *** mattstep has quit IRC | 22:39 | |
| *** mattstep has joined #openstack | 22:40 | |
| *** arBmind has quit IRC | 22:40 | |
| *** aliguori has quit IRC | 22:44 | |
| *** RichardRaseley has quit IRC | 22:45 | |
| *** lloydde has quit IRC | 22:45 | |
| *** justfine has quit IRC | 22:48 | |
| *** rnorwood has quit IRC | 22:49 | |
| *** mattray has quit IRC | 22:50 | |
| *** AlanClark has quit IRC | 22:50 | |
| *** nOStahl has quit IRC | 22:50 | |
| *** dcramer_ has quit IRC | 22:50 | |
| uvirtbot | New bug: #961732 in openstack-manuals "Online documentation have to click through each search result to get context" [Undecided,New] https://launchpad.net/bugs/961732 | 22:51 |
| *** Glacee has quit IRC | 22:51 | |
| *** rnirmal has quit IRC | 22:52 | |
| *** heckj has quit IRC | 22:53 | |
| *** ahasenack has quit IRC | 22:54 | |
| *** ryanpetrello has quit IRC | 23:00 | |
| *** heckj has joined #openstack | 23:00 | |
| *** zzed has quit IRC | 23:00 | |
| *** sandywalsh has quit IRC | 23:00 | |
| *** heckj has quit IRC | 23:02 | |
| *** bhall has quit IRC | 23:02 | |
| *** marant has quit IRC | 23:04 | |
| *** llang629 has joined #openstack | 23:08 | |
| *** llang629 has left #openstack | 23:08 | |
| *** marant has joined #openstack | 23:11 | |
| *** zomGreg has quit IRC | 23:12 | |
| *** alrs has quit IRC | 23:15 | |
| *** alrs has joined #openstack | 23:15 | |
| *** sandywalsh has joined #openstack | 23:16 | |
| *** zomGreg has joined #openstack | 23:18 | |
| *** alrs has quit IRC | 23:20 | |
| *** chrisfer1 has joined #openstack | 23:21 | |
| uvirtbot | New bug: #961761 in horizon "If nova-network is down, Access and Security panel hangs forever" [High,Confirmed] https://launchpad.net/bugs/961761 | 23:21 |
| uvirtbot | New bug: #961762 in nova "Rollback create_disk not catching StorageError" [Medium,In progress] https://launchpad.net/bugs/961762 | 23:21 |
| *** wonk has quit IRC | 23:22 | |
| *** chrisfer has quit IRC | 23:22 | |
| *** wonk has joined #openstack | 23:22 | |
| *** mdomsch has joined #openstack | 23:25 | |
| *** alrs has joined #openstack | 23:26 | |
| *** dubsquared has joined #openstack | 23:27 | |
| anticw | notmyname: about? | 23:28 |
| anticw | notmyname: swift + keystone ... the auth process people are using here is different to the 'header based' method in the past to get tokens | 23:33 |
| anticw | which basically makes it pointless ... because existing clients and sw breaks | 23:33 |
| anticw | ... or perhaps WeAreDoingItWrong and the old method still works somehow? | 23:33 |
| *** clauden___ has joined #openstack | 23:34 | |
| *** clauden has quit IRC | 23:34 | |
| *** clauden___ is now known as clauden | 23:34 | |
| *** lloydde has joined #openstack | 23:37 | |
| *** qazwsx has quit IRC | 23:40 | |
| *** flaviamissi has joined #openstack | 23:40 | |
| *** jkyle has quit IRC | 23:41 | |
| *** dolphm has joined #openstack | 23:42 | |
| *** nati has joined #openstack | 23:43 | |
| *** sjhon has joined #openstack | 23:44 | |
| sjhon | hi. Have anyone used the Rackspace mobile client (ISO/Android) with Swift ? | 23:46 |
| *** jdg has quit IRC | 23:49 | |
| *** lloydde has quit IRC | 23:51 | |
| *** deshantm has joined #openstack | 23:55 | |
| uvirtbot | New bug: #961784 in nova "python-novaclient package does not include revision history" [Undecided,New] https://launchpad.net/bugs/961784 | 23:56 |
| *** alrs has quit IRC | 23:57 | |
| *** wilmoore has quit IRC | 23:58 | |
| *** perestre1ka has joined #openstack | 23:59 | |
| *** perestrelka has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!