| Ryan_Lane | is there any update on Atlas? is the code ever going to be released? | 00:00 |
|---|---|---|
| Ryan_Lane | ah ha: https://github.com/rackspace/atlas-lb | 00:06 |
| *** Mandell has quit IRC | 00:10 | |
| *** agonella has quit IRC | 00:10 | |
| *** lloydde has joined #openstack-dev | 00:14 | |
| *** novas0x2a|laptop has quit IRC | 00:15 | |
| *** maplebed_ has joined #openstack-dev | 00:17 | |
| *** maplebed_ has joined #openstack-dev | 00:17 | |
| *** maplebed is now known as Guest77188 | 00:17 | |
| *** mnewby has quit IRC | 00:18 | |
| *** Guest77188 has quit IRC | 00:19 | |
| *** novas0x2a|laptop has joined #openstack-dev | 00:20 | |
| *** novas0x2a|laptop has quit IRC | 00:21 | |
| *** novas0x2a|laptop has joined #openstack-dev | 00:21 | |
| *** Mandell has joined #openstack-dev | 00:22 | |
| *** heckj has quit IRC | 00:22 | |
| *** Mandell has quit IRC | 00:23 | |
| *** maplebed_ has quit IRC | 00:24 | |
| *** maplebed__ has joined #openstack-dev | 00:25 | |
| *** maplebed__ has joined #openstack-dev | 00:25 | |
| *** maplebed__ is now known as maplebed | 00:27 | |
| *** anotherjesse is now known as anotherjesse_zz | 00:35 | |
| *** spiffxp has joined #openstack-dev | 00:37 | |
| *** pixelbeat has quit IRC | 00:39 | |
| *** andrewsmedina_ has joined #openstack-dev | 00:40 | |
| *** andrewsmedina has quit IRC | 00:43 | |
| *** andrewsmedina_ is now known as andrewsmedina | 00:43 | |
| *** maplebed has quit IRC | 00:45 | |
| *** martine has joined #openstack-dev | 00:49 | |
| *** tryggvil has quit IRC | 00:53 | |
| *** hub_cap has joined #openstack-dev | 00:55 | |
| *** bencherian has joined #openstack-dev | 00:57 | |
| *** anotherjesse_zz is now known as anotherjesse | 00:58 | |
| *** anotherjesse is now known as anotherjesse_zz | 01:01 | |
| *** thingee has quit IRC | 01:03 | |
| *** johnpostlethwait has quit IRC | 01:03 | |
| *** jog0 has left #openstack-dev | 01:10 | |
| *** roge has joined #openstack-dev | 01:14 | |
| *** vincentricci has left #openstack-dev | 01:16 | |
| *** PotHix has quit IRC | 01:34 | |
| *** rkukura has joined #openstack-dev | 01:43 | |
| *** johnpostlethwait has joined #openstack-dev | 01:44 | |
| *** eglynn has joined #openstack-dev | 01:44 | |
| *** bencherian has quit IRC | 01:45 | |
| *** eglynn__ has quit IRC | 01:47 | |
| *** bencherian has joined #openstack-dev | 01:49 | |
| *** shevek_ has joined #openstack-dev | 01:51 | |
| *** agonella has joined #openstack-dev | 01:52 | |
| *** ayoung has quit IRC | 01:55 | |
| *** mnewby has joined #openstack-dev | 01:56 | |
| *** mnewby has joined #openstack-dev | 01:56 | |
| *** roge has quit IRC | 01:58 | |
| *** shevek_ has quit IRC | 02:00 | |
| *** adjohn has quit IRC | 02:02 | |
| *** roge has joined #openstack-dev | 02:03 | |
| *** spiffxp has quit IRC | 02:03 | |
| *** novas0x2a|laptop has quit IRC | 02:04 | |
| *** dolphm has joined #openstack-dev | 02:05 | |
| *** letterj has left #openstack-dev | 02:15 | |
| *** jdurgin has quit IRC | 02:16 | |
| *** rods has quit IRC | 02:17 | |
| annegentle | hey I've got a devstack install, but can't find the glance ini and .conf files, where are they put? | 02:18 |
| annegentle | there's a bug to put them in /etc/glance https://bugs.launchpad.net/devstack/+bug/944169 but where are they now? | 02:19 |
| uvirtbot | Launchpad bug 944169 in devstack "Move service config files to /etc" [Medium,Triaged] | 02:19 |
| *** jakedahn is now known as jakedahn_zz | 02:22 | |
| *** agonella has quit IRC | 02:25 | |
| annegentle | nm, guess they are in /devstack/files and in /opt/stack/glance | 02:25 |
| *** cdub has quit IRC | 02:27 | |
| *** cdub has joined #openstack-dev | 02:27 | |
| *** Ryan_Lane has quit IRC | 02:31 | |
| mnewby | Is there anybody about that knows about nova's ec2 api? I'm trying to fix a bug and would appreciate guidance as to the best approach to take. | 02:33 |
| *** littleidea has quit IRC | 02:33 | |
| *** littleidea has joined #openstack-dev | 02:34 | |
| *** ncode has quit IRC | 02:48 | |
| *** adjohn has joined #openstack-dev | 02:58 | |
| *** andrewsmedina has quit IRC | 03:07 | |
| *** jakedahn_zz is now known as jakedahn | 03:08 | |
| *** andrewsmedina has joined #openstack-dev | 03:08 | |
| *** flaviamissi_ has joined #openstack-dev | 03:18 | |
| *** flaviamissi has quit IRC | 03:19 | |
| *** flaviamissi_ is now known as flaviamissi | 03:22 | |
| *** Mandell has joined #openstack-dev | 03:27 | |
| *** deshantm has quit IRC | 03:30 | |
| *** johnpostlethwait has quit IRC | 03:37 | |
| zaitcev | I have a Python question: does anyone know of a way to attach a debugger to Python that shows a current traceback, like gdb -p does for C? The program has several threads. | 03:42 |
| *** flaviamissi has quit IRC | 03:42 | |
| justinsb | zaitcev: I managed it using gdb to diagnose a crash, but it was very painful. What are you trying to do? | 03:43 |
| zaitcev | https://bugs.launchpad.net/swift/+bug/704015 | 03:44 |
| uvirtbot | Launchpad bug 704015 in swift "st and client.py's reads aren't checking for EOF condition" [Undecided,New] | 03:44 |
| zaitcev | Basically, st is looping somewhere, but if I interrupt it, KeyboardInterrupt ends in wrong thread and I see no traceback. | 03:45 |
| zaitcev | Or the right thread catches it. | 03:45 |
| justinsb | zaitcev: I put my notes here https://bugs.launchpad.net/nova/+bug/937554/comments/13 | 03:45 |
| uvirtbot | Launchpad bug 937554 in nova "Lots of problems with deleting a server immediately after create (dup-of: 934575)" [High,Fix committed] | 03:45 |
| uvirtbot | Launchpad bug 934575 in nova "notifier endless loops in is_primitive" [Medium,Fix released] | 03:45 |
| justinsb | zaitcev: I believe there are tricks with attaching the python debugger to the interrupt handler if it isn't a hard crash | 03:46 |
| zaitcev | mother of god | 03:46 |
| justinsb | zaitcev: http://stackoverflow.com/questions/132058/showing-the-stack-trace-from-a-running-python-application | 03:46 |
| *** dolphm has quit IRC | 03:46 | |
| justinsb | zaitcev: Search for dumpstacks on that page | 03:47 |
| justinsb | zaitcev: But I haven't tried this way.. | 03:47 |
| justinsb | zaitcev: But _much_ easier if it isn't a total crash | 03:48 |
| *** hub-cap has joined #openstack-dev | 03:52 | |
| *** hub_cap has quit IRC | 03:52 | |
| *** hub-cap is now known as hub_cap | 03:52 | |
| zaitcev | justinsb: Thanks a lot, I'm digesting the info now. | 03:52 |
| *** anotherjesse_zz is now known as anotherjesse | 03:58 | |
| anotherjesse | mnewby: still wondering about ec2? | 04:00 |
| mnewby | anotherjesse: yes | 04:00 |
| anotherjesse | mnewby: ask away | 04:01 |
| mnewby | bug 948286 | 04:01 |
| uvirtbot | Launchpad bug 948286 in nova "euca-describe-instances returns ImageNotFound on first run" [High,Triaged] https://launchpad.net/bugs/948286 | 04:01 |
| mnewby | I've reproduced the problem - euca-describe-images calls describe_images, which creates a uuid<->id mapping if it doesn't exist. | 04:02 |
| mnewby | euca-describe-instances presumes that the db is already populated, which results in an error if the mapping isn't present. | 04:02 |
| mnewby | So the question is - should the mapping be lazily-created by describe_instances, as it is for describe_images? | 04:02 |
| mnewby | Or should the mapping be created at instance creation instead, non-lazily? | 04:03 |
| anotherjesse | mnewby: the code is in cloud.py - nova/api/ec2/cloud.py | 04:03 |
| anotherjesse | let's go read :) | 04:03 |
| mnewby | have nova/image/s3.py open, too. | 04:03 |
| mnewby | in cloud.py, the relevant line is 1457 for describe_images - a call to s3.detail() | 04:05 |
| mnewby | for describe_instances, the relevant line is 1175 - a call to cloud._get_image_id(), which in turn calls s3.get_image_id() | 04:06 |
| mnewby | And the difference is pretty clear - cloud.describe_images -> s3.detail -> s3._translate_uuids_to_ids, which will either create or retrieve the uuid<->id mapping. | 04:07 |
| mnewby | cloud.describe_instances -> cloud._get_image_id -> s3.get_image_id, which only retrieves an existing mapping. | 04:08 |
| anotherjesse | mnewby: the best thing to do is create a test case for this - to ensure we don't regress | 04:09 |
| mnewby | anotherjesse: So in keeping with what a cloud.describe_images() call will do, cloud.describe_instances should probably be updated to create or retrieve a uuid<->id mapping rather than just retrieving it if it exists. | 04:09 |
| anotherjesse | mnewby: sounds correct | 04:09 |
| mnewby | anotherjesse: can do. I was just wondering whether the whole lazy loading thing should go away. It feels funny to be looking for something and always creating it if it doesn't exist - what is the mapping actually used for in that case? Am I missing something? | 04:10 |
| anotherjesse | mnewby: aws interopt used to be the primary api and we still have some issues from the transition to AWS being interopt ontop of an openstack api | 04:11 |
| mnewby | Ok. So it is correct for the uuid<->id to be created on the fly since it is more about interop than being authoritative? | 04:12 |
| *** martine has quit IRC | 04:12 | |
| anotherjesse | I would have to verify with vishy/others who have worked on that code more recently than I | 04:13 |
| anotherjesse | but it seems reasonable | 04:13 |
| mnewby | Ok. I'll go with that assumption and vishy et all can confirm the approach in review. | 04:13 |
| mnewby | danke! | 04:13 |
| anotherjesse | mnewby: yeppers | 04:15 |
| bcwaldon | anotherjesse/mnewby: can I help you guys with this? | 04:17 |
| bcwaldon | I wrote the code | 04:18 |
| anotherjesse | bcwaldon: here to save the day with his crossbow of truth | 04:18 |
| mnewby | bcwaldon: Certainly. You were the original author of much of the code in question. | 04:18 |
| mnewby | bcwaldon: are you up to speed or should I reiterate? | 04:18 |
| bcwaldon | I think I know what you're talking about | 04:19 |
| bcwaldon | the code works with the assumption that a user has called euca-describe-images | 04:19 |
| mnewby | I was thinking of exposing the nested find_or_create in nova.image.s3.S3ImageService, so that nova.api.ec2.get_image_id could use it. | 04:19 |
| mnewby | Ah. And if they haven't, as in the case of this bug? | 04:20 |
| bcwaldon | yes | 04:20 |
| bcwaldon | so what might make sense is to push the image uuid -> int id mapping into ec2utils | 04:20 |
| bcwaldon | so anything calling ec2utils.image_id_to_ec2_id gets this mapping for free | 04:21 |
| mnewby | Ah, ok. Let me take a look at ec2utils. | 04:21 |
| bcwaldon | I just pushed up a bugfix for the metadata service which had no idea about the s3_images table | 04:21 |
| bcwaldon | anotherjesse: I shot the crossbow, btw | 04:23 |
| anotherjesse | bcwaldon: I heard… wtf | 04:23 |
| bcwaldon | anotherjesse: used a hilighter instead of the arrows | 04:23 |
| bcwaldon | I had to test it | 04:23 |
| anotherjesse | in the office? | 04:23 |
| bcwaldon | maybe | 04:23 |
| anotherjesse | zomg | 04:24 |
| bcwaldon | it was anticlimactic | 04:24 |
| *** johnpostlethwait has joined #openstack-dev | 04:24 | |
| vishy | haha | 04:24 |
| vishy | I was just discussing this offline with waldon :) | 04:25 |
| mnewby | bcwaldon: So the (get/create/translate) methods in S3ImageService should be moved to ec2utils? | 04:25 |
| vishy | same solution i was considering | 04:25 |
| bcwaldon | vishy: curse you, mnewby didnt know I just stole your idea | 04:25 |
| mnewby | heh | 04:25 |
| vishy | bcwaldon: oh sorry for outing you :p | 04:26 |
| bcwaldon | how did three individual people start thinking about the same problem all at the same time? | 04:26 |
| bcwaldon | weird | 04:26 |
| *** johnpostlethwait has quit IRC | 04:38 | |
| mnewby | bcwaldon: another question? | 04:39 |
| bcwaldon | shoot | 04:39 |
| *** mattstep has quit IRC | 04:40 | |
| *** hattwick has quit IRC | 04:40 | |
| mnewby | bcwaldon:So I'm moving the get_image_id/etc methods from S3ImageService to ec2utils (and renaming to image_uuid_to_id/etc for consistency). | 04:40 |
| *** mattstep has joined #openstack-dev | 04:40 | |
| mnewby | You mentioned also translating from ec2 ids to uuids I think? | 04:41 |
| bcwaldon | so currently there's a glance id (a uuid) which is translated to integer ids internally before being presented as ec2 ids (hex) | 04:41 |
| mnewby | right | 04:41 |
| mnewby | So those integer ids probably don't need to be exposed? | 04:42 |
| bcwaldon | so I would like to get away from using 'uuid' and instead use 'glance' | 04:42 |
| bcwaldon | just in naming things | 04:42 |
| mnewby | Oh, ok. | 04:42 |
| bcwaldon | he integer ids never need to be exposed | 04:42 |
| bcwaldon | I'm not sure why we actually have them | 04:42 |
| mnewby | Ok. Makes sense. so ec2_id_to_glance_id? | 04:42 |
| bcwaldon | yes | 04:42 |
| mnewby | ad vice versa? | 04:42 |
| bcwaldon | yes | 04:42 |
| mnewby | Ok, can do. | 04:42 |
| bcwaldon | there might be a reason we need to have that interim integer id, I just dont remember off-hand | 04:43 |
| mnewby | Useful to have for internal use, I'm sure. But best to have direct translation between glance and ec2 ids. | 04:43 |
| mnewby | (for most uses) | 04:43 |
| bcwaldon | I'm not sure what that internal use is | 04:43 |
| mnewby | So I'll leave some internal (_ prefixed) methods to access the id. | 04:43 |
| bcwaldon | since this is all just for the ec2/s3 side of things | 04:44 |
| mnewby | So should the glance_id <-> db id translation methods disappear then? | 04:44 |
| bcwaldon | its probably safer to do this in two steps | 04:45 |
| bcwaldon | first just move code to ec2utils for the mapping | 04:45 |
| mnewby | Ok. | 04:45 |
| bcwaldon | but I am questioning how much value using an interim integer format gives us | 04:45 |
| bcwaldon | it is a simpler form of the hex id, which we get for free with the autoincrement column | 04:45 |
| mnewby | it can always be extracted from the ec2 id, too. | 04:46 |
| anotherjesse | anyone have bandwidth to fix https://bugs.launchpad.net/nova/+bug/953716 - hopefully simple | 04:56 |
| uvirtbot | Launchpad bug 953716 in nova "nova-manage floating delete fails" [Undecided,New] | 04:56 |
| *** zigo has joined #openstack-dev | 04:57 | |
| *** jakedahn is now known as jakedahn_zz | 05:04 | |
| *** hattwick has joined #openstack-dev | 05:08 | |
| *** roge has quit IRC | 05:08 | |
| *** hub_cap has quit IRC | 05:09 | |
| *** hattwick has quit IRC | 05:12 | |
| *** ches has quit IRC | 05:13 | |
| *** ches has joined #openstack-dev | 05:13 | |
| *** hattwick has joined #openstack-dev | 05:14 | |
| *** hattwick has quit IRC | 05:33 | |
| *** johnpostlethwait has joined #openstack-dev | 05:36 | |
| *** thingee has joined #openstack-dev | 05:39 | |
| *** thingee has quit IRC | 05:45 | |
| *** rgoodwin has quit IRC | 05:47 | |
| *** ohnoimdead has quit IRC | 05:48 | |
| *** ohnoimdead has joined #openstack-dev | 05:49 | |
| *** garyk has quit IRC | 05:50 | |
| *** anotherjesse is now known as anotherjesse_zz | 05:52 | |
| *** rgoodwin has joined #openstack-dev | 05:53 | |
| *** anotherjesse_zz is now known as anotherjesse | 06:05 | |
| *** anotherjesse is now known as anotherjesse_zz | 06:07 | |
| *** Mandell has quit IRC | 06:14 | |
| *** openstack has joined #openstack-dev | 06:36 | |
| *** shevek_ has joined #openstack-dev | 06:36 | |
| *** btorch has quit IRC | 06:36 | |
| *** cmagina_ has joined #openstack-dev | 06:37 | |
| *** johnpostlethwait has quit IRC | 06:37 | |
| *** JStoker_ has quit IRC | 06:37 | |
| *** adam_g has quit IRC | 06:37 | |
| *** rmk has quit IRC | 06:37 | |
| *** cmagina has quit IRC | 06:37 | |
| *** blamar__ has quit IRC | 06:37 | |
| *** sdague has quit IRC | 06:38 | |
| *** adam_g has joined #openstack-dev | 06:39 | |
| *** pknouff has quit IRC | 06:40 | |
| *** pknouff has joined #openstack-dev | 06:41 | |
| *** clayg_ has joined #openstack-dev | 06:43 | |
| *** littleidea has joined #openstack-dev | 06:44 | |
| *** bepernoot has joined #openstack-dev | 06:45 | |
| *** Ryan_Lane has joined #openstack-dev | 06:46 | |
| *** JStoker has joined #openstack-dev | 06:48 | |
| *** blamar__ has joined #openstack-dev | 06:50 | |
| vishy | tr3buchet: I responded to your comment on https://review.openstack.org/#change,5246 Let me know which you prefer | 06:53 |
| *** rmk has joined #openstack-dev | 06:54 | |
| *** JStoker_ has joined #openstack-dev | 06:59 | |
| *** shang has quit IRC | 07:00 | |
| *** littleidea has quit IRC | 07:01 | |
| *** dtroyer has quit IRC | 07:03 | |
| *** bepernoot has quit IRC | 07:04 | |
| *** shang has joined #openstack-dev | 07:07 | |
| *** johnpostlethwa-1 has quit IRC | 07:17 | |
| *** garyk has joined #openstack-dev | 07:21 | |
| *** anotherjesse is now known as anotherjesse_zz | 07:31 | |
| *** mnewby has quit IRC | 07:40 | |
| *** Remco_ has joined #openstack-dev | 07:41 | |
| *** bencherian has quit IRC | 07:46 | |
| *** Remco_ has quit IRC | 07:47 | |
| YorikSar | Are there any way to calm Jenkins down? It complained about unmerged dependency for 7 times now: https://review.openstack.org/4961 | 07:52 |
| *** hattwick has joined #openstack-dev | 07:59 | |
| *** bepernoot has joined #openstack-dev | 08:10 | |
| *** shevek_ has quit IRC | 08:10 | |
| *** reidrac has joined #openstack-dev | 08:12 | |
| *** Remco_ has joined #openstack-dev | 08:17 | |
| *** Remco_ has quit IRC | 08:26 | |
| *** dneary has joined #openstack-dev | 08:28 | |
| *** Ryan_Lane has quit IRC | 08:51 | |
| *** ghe_ has quit IRC | 09:03 | |
| *** ghe_ has joined #openstack-dev | 09:09 | |
| *** derekh has joined #openstack-dev | 09:10 | |
| *** Remco_ has joined #openstack-dev | 09:15 | |
| *** paulormg has joined #openstack-dev | 09:16 | |
| soren | ttx: Nova builds fixed again. | 09:16 |
| *** pixelbeat has joined #openstack-dev | 09:16 | |
| *** berendt has joined #openstack-dev | 09:23 | |
| *** markmc has joined #openstack-dev | 09:25 | |
| mikal | Do other people notice "qemu-img info" running quite a lot with essex's nova-compute, or is it just me? | 09:25 |
| mikal | I observe 4,000+ runs in 3 hours. | 09:26 |
| *** bepernoot has quit IRC | 09:28 | |
| *** hugokuo has joined #openstack-dev | 09:29 | |
| *** berendt has quit IRC | 09:40 | |
| *** sniperd_ has joined #openstack-dev | 09:41 | |
| *** berendt has joined #openstack-dev | 09:41 | |
| *** Kiall_ has joined #openstack-dev | 09:43 | |
| *** mancdaz1203 has joined #openstack-dev | 09:45 | |
| *** sdake__ has joined #openstack-dev | 09:47 | |
| *** markmc has quit IRC | 09:49 | |
| *** blamar__ has quit IRC | 09:49 | |
| *** JStoker_ has quit IRC | 09:49 | |
| *** sdake_ has quit IRC | 09:49 | |
| *** openstackjenkins has quit IRC | 09:49 | |
| *** nikhil__ has quit IRC | 09:49 | |
| *** rbasak has quit IRC | 09:49 | |
| *** sniperd has quit IRC | 09:49 | |
| *** zykes- has quit IRC | 09:49 | |
| *** Kiall has quit IRC | 09:49 | |
| *** mancdaz has quit IRC | 09:49 | |
| *** infernix has quit IRC | 09:49 | |
| *** _cerberus_ has quit IRC | 09:49 | |
| *** root____1 has quit IRC | 09:49 | |
| *** Kiall_ is now known as Kiall | 09:49 | |
| *** paulormg has quit IRC | 09:52 | |
| *** JStoker is now known as 31NAA1G4I | 09:55 | |
| *** markmc has joined #openstack-dev | 09:55 | |
| *** JStoker has joined #openstack-dev | 09:55 | |
| *** blamar__ has joined #openstack-dev | 09:55 | |
| *** openstackjenkins has joined #openstack-dev | 09:55 | |
| *** nikhil__ has joined #openstack-dev | 09:55 | |
| *** rbasak has joined #openstack-dev | 09:55 | |
| *** zykes- has joined #openstack-dev | 09:55 | |
| *** infernix has joined #openstack-dev | 09:55 | |
| *** _cerberus_ has joined #openstack-dev | 09:55 | |
| *** root____1 has joined #openstack-dev | 09:55 | |
| *** hubbard.freenode.net sets mode: +v _cerberus_ | 09:55 | |
| *** oneiroi has joined #openstack-dev | 10:54 | |
| *** flaviamissi has joined #openstack-dev | 10:57 | |
| *** shang has quit IRC | 10:59 | |
| *** flaviamissi has quit IRC | 11:09 | |
| *** ghe_ has quit IRC | 11:15 | |
| *** shang has joined #openstack-dev | 11:16 | |
| *** ghe_ has joined #openstack-dev | 11:20 | |
| *** dbuzz has joined #openstack-dev | 11:28 | |
| *** dbuzz has joined #openstack-dev | 11:28 | |
| *** dbuzz has quit IRC | 11:28 | |
| *** dbuzz has joined #openstack-dev | 11:28 | |
| *** oneiroi has quit IRC | 11:29 | |
| *** dbuzz is now known as oneiroi | 11:29 | |
| *** tryggvil_ has joined #openstack-dev | 11:37 | |
| *** Remco_ has quit IRC | 11:42 | |
| *** ghe_ has quit IRC | 11:43 | |
| *** dbuzz has joined #openstack-dev | 11:46 | |
| *** dbuzz has joined #openstack-dev | 11:46 | |
| *** lloydde has quit IRC | 11:46 | |
| *** oneiroi has quit IRC | 11:47 | |
| *** dbuzz has quit IRC | 11:47 | |
| *** markvoelker has joined #openstack-dev | 11:47 | |
| *** lloydde has joined #openstack-dev | 11:47 | |
| *** lts has joined #openstack-dev | 11:58 | |
| *** joesavak has joined #openstack-dev | 12:11 | |
| *** ghe_ has joined #openstack-dev | 12:17 | |
| *** tryggvil_ has quit IRC | 12:20 | |
| *** jsavak has joined #openstack-dev | 12:23 | |
| *** joesavak has quit IRC | 12:27 | |
| *** tryggvil has joined #openstack-dev | 12:28 | |
| *** sdague_ is now known as sdague | 12:39 | |
| *** martine has joined #openstack-dev | 12:42 | |
| *** dneary has quit IRC | 12:45 | |
| *** hugokuo has quit IRC | 12:50 | |
| *** kbringard has joined #openstack-dev | 12:52 | |
| *** andrewsmedina has quit IRC | 12:53 | |
| *** AlanClark has joined #openstack-dev | 12:58 | |
| *** bsza has joined #openstack-dev | 12:58 | |
| *** flaviamissi has joined #openstack-dev | 13:09 | |
| *** tryggvil has quit IRC | 13:16 | |
| flaviamissi | mourning guys, I work at globo.com and we made some packages (rpm) of the essex version and we want to share them, could you tell us how to proceed? | 13:18 |
| *** dprince has joined #openstack-dev | 13:20 | |
| *** tryggvil_ has joined #openstack-dev | 13:22 | |
| *** andrewsmedina has joined #openstack-dev | 13:22 | |
| *** rods has joined #openstack-dev | 13:26 | |
| *** ayoung has joined #openstack-dev | 13:29 | |
| *** davidkranz has joined #openstack-dev | 13:33 | |
| davidkranz | Is devstack/essex supposed to work on precise? | 13:33 |
| davidkranz | I get Note, selecting 'python' instead of 'python-wsgiref' Package python-argparse is a virtual package provided by: python2.7 2.7.3~rc1-1ubuntu2 python 2.7.2-9ubuntu2 You should explicitly select one to install. | 13:33 |
| ttx | davidkranz: I think it only supports 11.10 right now... but it sounds like a good idea to make it support Precise | 13:34 |
| ttx | flaviamissi: how are those different from the packages in Fedora ? | 13:34 |
| davidkranz | Here is the readable error http://pastebin.com/ZGif0Nex | 13:35 |
| ttx | flaviamissi: maybe join #openstack-packaging | 13:35 |
| davidkranz | Related to the recent thread on the openstack list, I am trying to upgrade to essex but am crashing and burning so thought to try devstack first. | 13:36 |
| flaviamissi | ttx: there are several adaptations to centos 6.2 | 13:36 |
| ttx | flaviamissi: I'm not really up-to-date with the various RPM distributions of OpenStack, but some people in #openstack-packaging should. | 13:37 |
| flaviamissi | ttx: ok, I'will talk to them | 13:37 |
| flaviamissi | ttx: thanks :) | 13:37 |
| *** Glacee has joined #openstack-dev | 13:38 | |
| davidkranz | I went through a lot of pain getting diablo to really work 6 months ago and am having a feeling of deja vu. Is it too soon for essex for non-nova-core type people? | 13:39 |
| davidkranz | Also, I saw today that essex will not be supported on oneiric but, three weeks from release, devstack does not work on precise. How could this be? | 13:41 |
| *** berendt has quit IRC | 13:44 | |
| *** crobinso has joined #openstack-dev | 13:44 | |
| ttx | Essex can run on oneiric. DevStack should be updated to work on Precise -- the reason why it's not done already is that Precise being under development, it's a bit of a moving target. As it stabilizes, I expect DevStack to start supporting it | 13:47 |
| *** garyk has quit IRC | 13:49 | |
| davidkranz | ttx: I understand what you mean about precise and can wait for that. But essex really does not run on oneiric with the libvirt hang bug and I was referring to the statement that | 13:51 |
| davidkranz | that bug would not be backported to oneiric. | 13:51 |
| *** littleidea has joined #openstack-dev | 13:52 | |
| davidkranz | There were also other libvirt issues and I thought there was a statement that you would need precise for essex. Is that not correct? | 13:52 |
| ttx | davidkranz: I think it's the other way around. Ubuntu 11.10 ships Diablo. Ubuntu 12.04 LTS will ship Essex. | 13:53 |
| ttx | You can do tricks to run Essex on 11.10 -- but that's a trick | 13:53 |
| ttx | not a production supported option. | 13:53 |
| ttx | unless Ubuntu backports Essex to 11.10 and supports that | 13:54 |
| ttx | but i don't expect them (us) to. | 13:54 |
| *** dtroyer has joined #openstack-dev | 13:55 | |
| *** jsavak has quit IRC | 13:56 | |
| ttx | bbl | 13:57 |
| davidkranz | ttx: That's what I thought. I was just not understanding why nova developers, most of whom I presume use devstack, are still using oneiric three weeks from precise/essex release!!! | 14:00 |
| *** pixelbeat_ has joined #openstack-dev | 14:01 | |
| *** berendt has joined #openstack-dev | 14:03 | |
| *** dneary has joined #openstack-dev | 14:07 | |
| *** dneary has quit IRC | 14:07 | |
| *** dneary has joined #openstack-dev | 14:07 | |
| *** Remco_ has joined #openstack-dev | 14:08 | |
| *** bepernoot has joined #openstack-dev | 14:17 | |
| *** dtroyer has quit IRC | 14:27 | |
| *** oneiroi has joined #openstack-dev | 14:29 | |
| *** russellb is now known as SureIllCodeThat | 14:38 | |
| *** SureIllCodeThat is now known as russellb | 14:38 | |
| dweimer | flaviamissi: I would be interested in taking a look at your CentOS Essex packages. I've been slowly working on modifying the fedora 17 packages to work with CentOS myself. | 14:39 |
| flaviamissi | dweimer: do you want the specs or everything? | 14:40 |
| flaviamissi | dweimer: to port to centos we needed to repack python-paste-deploy and python-web0b | 14:41 |
| dweimer | flaviamissi: Did you have to patch the code at all? There is a version of webob 1.0 in EPEL that I've been using. It requires a patch to some __init__.py files though. | 14:41 |
| flaviamissi | dweimer: no, we didn't patched the code, we're using the webob 1.1.1 | 14:43 |
| flaviamissi | dweimer: we're creating a git repository with the specs right now | 14:44 |
| flaviamissi | dweimer: as soon as we get it I send you the address | 14:44 |
| dweimer | flaviamissi: Excellent. Thank you. | 14:44 |
| flaviamissi | dweimer: here it is: https://github.com/timeredbull/openstack-essex-4-rpms | 14:47 |
| flaviamissi | dweimer: we haven't packaged quantum yet | 14:49 |
| *** jdg has joined #openstack-dev | 14:50 | |
| *** zzed has joined #openstack-dev | 14:51 | |
| dweimer | flaviamissi: I have some available test nodes to try these out today. Thanks again. If I end up with any patches do you take pull requests through github? | 14:53 |
| flaviamissi | dweimer: of course! it'd be welcome ;) | 14:54 |
| *** cp16net has joined #openstack-dev | 14:59 | |
| *** Remco_ has quit IRC | 14:59 | |
| *** msinhore has joined #openstack-dev | 15:07 | |
| *** hhoover has joined #openstack-dev | 15:08 | |
| *** deshantm has joined #openstack-dev | 15:12 | |
| *** lloydde has quit IRC | 15:13 | |
| *** hub_cap has joined #openstack-dev | 15:18 | |
| *** dolphm has joined #openstack-dev | 15:18 | |
| *** dubsquared has quit IRC | 15:31 | |
| *** belmoreira has joined #openstack-dev | 15:34 | |
| *** hhoover has left #openstack-dev | 15:35 | |
| *** deshantm_ has joined #openstack-dev | 15:36 | |
| *** ncode has joined #openstack-dev | 15:38 | |
| *** deshantm has quit IRC | 15:39 | |
| *** deshantm_ is now known as deshantm | 15:41 | |
| *** maploin has joined #openstack-dev | 15:46 | |
| *** maploin has quit IRC | 15:46 | |
| *** maploin has joined #openstack-dev | 15:46 | |
| Vek | so, 8 emails from jenkins just to tell us, "change could not be merged because of a dependency" is a little...absurd. | 15:55 |
| *** mikeyp has joined #openstack-dev | 15:55 | |
| Vek | any chance we could get that fixed up to just say it once, then notify us once the dependency is approved? | 15:55 |
| *** markmc has quit IRC | 15:57 | |
| *** hub-cap has joined #openstack-dev | 16:00 | |
| *** spiffxp has joined #openstack-dev | 16:01 | |
| *** hub_cap has quit IRC | 16:02 | |
| *** hub-cap is now known as hub_cap | 16:02 | |
| *** vincentricci has joined #openstack-dev | 16:02 | |
| *** paulormg has joined #openstack-dev | 16:02 | |
| *** cp16net has quit IRC | 16:03 | |
| *** bencherian has joined #openstack-dev | 16:03 | |
| *** hub_cap has quit IRC | 16:03 | |
| *** cp16net has joined #openstack-dev | 16:04 | |
| *** gyee has joined #openstack-dev | 16:04 | |
| *** ayoung has quit IRC | 16:06 | |
| *** cp16net has quit IRC | 16:06 | |
| *** heckj has joined #openstack-dev | 16:06 | |
| *** cp16net has joined #openstack-dev | 16:07 | |
| jdg | well this is annoying... see ya | 16:08 |
| *** jdg has quit IRC | 16:08 | |
| *** littleidea has quit IRC | 16:08 | |
| *** maploin has quit IRC | 16:08 | |
| *** jkoelker has quit IRC | 16:09 | |
| *** jdg has joined #openstack-dev | 16:10 | |
| *** jkoelker has joined #openstack-dev | 16:11 | |
| kbringard | wonder if they all got k-lined | 16:12 |
| kbringard | notmyname: still getting cannot send to channel | 16:13 |
| *** reidrac has quit IRC | 16:13 | |
| jdg | People are stupid and immature | 16:14 |
| kbringard | indeed | 16:14 |
| *** cmagina_ is now known as cmagina | 16:15 | |
| *** littleidea has joined #openstack-dev | 16:15 | |
| *** zzed has quit IRC | 16:15 | |
| *** lloydde has joined #openstack-dev | 16:15 | |
| jaypipes | notmyname: chance we could have you look into https://review.openstack.org/#change,4893 ? | 16:18 |
| jaypipes | notmyname: I have a dependent Glance bug related to that (https://bugs.launchpad.net/glance/+bug/944946) | 16:19 |
| uvirtbot | Launchpad bug 944946 in glance "Glance does not support auth server implementing 2.0 auth (ie: for Swift+Keystone)" [Medium,Confirmed] | 16:19 |
| jaypipes | notmyname: TIA! | 16:19 |
| YorikSar | johan_-_, Vek: So how about parenthesis vs backslash? | 16:19 |
| notmyname | ya, we've got a lot of partially reviewed stuff that needs to get a final review on and merged. I'd exepct to see a lot of merges this week as we get 1.4.8 for essex | 16:19 |
| *** jakedahn_zz is now known as jakedahn | 16:22 | |
| *** jakedahn is now known as jakedahn_zz | 16:23 | |
| jaypipes | notmyname: k, no worries thx :) | 16:23 |
| *** eglynn_ has joined #openstack-dev | 16:23 | |
| *** lloydde has quit IRC | 16:23 | |
| *** eglynn has quit IRC | 16:23 | |
| *** bepernoot has quit IRC | 16:27 | |
| smoser | http://paste.ubuntu.com/882017/ | 16:27 |
| smoser | is that known issue ? | 16:27 |
| smoser | ^ | 16:27 |
| mikeyp | Any KVMers willing to do doc reviews of https://review.openstack.org/#change,4985 and https://review.openstack.org/#change,5030 ? | 16:34 |
| mikeyp | As a bonus, you get to learn the new flags and behavior :-) | 16:34 |
| *** shevek_ has joined #openstack-dev | 16:35 | |
| *** dtroyer has joined #openstack-dev | 16:38 | |
| smoser | it seems that the 'glance' launcher script installed into /usr/local/bin is broken | 16:38 |
| smoser | at least if you have another glance library installed. | 16:38 |
| *** jdurgin has joined #openstack-dev | 16:39 | |
| russellb | mikeyp: done | 16:41 |
| *** gmi_ has joined #openstack-dev | 16:41 | |
| *** PotHix has joined #openstack-dev | 16:41 | |
| gmi_ | I have a question for nova-core devs | 16:42 |
| bcwaldon | gmi_: shoot | 16:44 |
| jaypipes | heckj: ping | 16:44 |
| gmi_ | I have a multi-host, multi-network VLAN mode setup installed from trunk and I created 100 networks using "nova-manage network create --multi_host=T --network_size=256 --num_networks=100 --bridge_interface=eth1 --fixed_range_v4=10.0.0.0/16 --label=private" and now when I an instance in the admin tenant, I see created 100 bridges, dnsmasq and all the networking needed to support them on my nova-compute node | 16:44 |
| jaypipes | heckj: getting a single failure in Keystone run_tests.sh on rfesh trunk... http://paste.openstack.org/show/9021/ | 16:44 |
| jaypipes | heckj: something known or should I file a bug? | 16:44 |
| heckj | jaypipes: yopu | 16:44 |
| jaypipes | heckj: ? :) | 16:45 |
| *** lloydde has joined #openstack-dev | 16:45 | |
| heckj | jaypipes: you're seeing that too? I see that on my mac, but not on linux - hadn't tracked it down yet. Thought I'd screwed up my local environment. | 16:45 |
| gmi_ | before, the dnsmasq and bridges would be created only as the instances from each project/tenant would be started on a nova-compute host | 16:45 |
| jaypipes | heckj: on Xubuntu 11.10 | 16:45 |
| heckj | (not enough coffee, can't type) | 16:45 |
| heckj | jaypipes: token cache expiration error? | 16:45 |
| jaypipes | heckj: looks like it... but not sure if I should report a bug? | 16:46 |
| heckj | jaypipes: please - I'm seeing it too, but I thought it was just me. It's working fine on the devstack, but I'm not sure where the issue actually is. | 16:46 |
| jaypipes | k, will do. | 16:46 |
| mikeyp | russelb: thanks | 16:46 |
| heckj | jaypipes: I've read the code, and it reads like it should work. walked through with PDB in the test and in those cases it didn't fail. Very odd. | 16:47 |
| *** sdake__ has quit IRC | 16:47 | |
| *** sdake has joined #openstack-dev | 16:48 | |
| jaypipes | heckj: looks like major hayden already reported: https://bugs.launchpad.net/keystone/+bug/954057 | 16:48 |
| uvirtbot | Launchpad bug 954057 in keystone "memcache test failing in python 2.6" [Undecided,Invalid] | 16:48 |
| gmi_ | check: http://paste.openstack.org/show/9023/ | 16:49 |
| jaypipes | heckj: failure is repeatable for me, FYI... not a random failure. | 16:49 |
| heckj | jaypipes: same here | 16:50 |
| jaypipes | k | 16:50 |
| heckj | jaypipes: reopened and assigned to myself. I'll do some digging on it later today | 16:50 |
| gmi_ | so my question is: is it normal to have dnsmasq running for networks/tenants that don't have running instances on a nova-compute host? | 16:51 |
| *** utlemming has quit IRC | 16:52 | |
| *** utlemming has joined #openstack-dev | 16:53 | |
| bcwaldon | gmi_: networking isnt my strong suit, looking for somebody else to help you | 16:54 |
| gmi_ | I have to leave for 30 min, I'll check again later | 16:54 |
| jaypipes | heckj: sounds good. | 16:54 |
| *** maplebed has joined #openstack-dev | 16:55 | |
| bcwaldon | gmi_: ok, sorry! | 16:55 |
| *** shevek_ has quit IRC | 16:56 | |
| jaypipes | notmyname: ping. Seeing errors in Glance test suite today popping up when connecting to Swift... seeing lots of these: "ClientException: Auth GET failed: http://auth.api.rackspacecloud.com:80/v1.0/ 301 Moved Permanently" Did something about authing with cloud files change recently? | 16:59 |
| notmyname | jaypipes: auth against RAX? | 17:00 |
| *** Mandell has joined #openstack-dev | 17:00 | |
| jaypipes | notmyname: yeah, we have a set of functional tests that use cloudf iles... | 17:00 |
| jaypipes | notmyname: lemme paste failure... one sec | 17:00 |
| notmyname | jaypipes: that's not a cloud files or swift error. that message is from RAX auth before it gets to anything I control | 17:00 |
| jaypipes | notmyname: hmm... ok. | 17:01 |
| notmyname | jaypipes: ah. you aren't using ssl | 17:01 |
| jaypipes | not sure what changed where this recently started breaking, but I'll dig into it | 17:01 |
| jaypipes | notmyname: yes, not using SSL afaik | 17:01 |
| notmyname | jaypipes: it requires it, so s/http/https/ and it should work | 17:02 |
| jaypipes | k, lemme see if that works... thx | 17:03 |
| *** mtaylor has quit IRC | 17:07 | |
| *** mtaylor has joined #openstack-dev | 17:08 | |
| jaypipes | notmyname: yup, that was it... thx | 17:08 |
| *** mtaylor has quit IRC | 17:08 | |
| *** mtaylor has joined #openstack-dev | 17:08 | |
| *** ChanServ sets mode: +v mtaylor | 17:08 | |
| *** thingee has joined #openstack-dev | 17:09 | |
| *** Ryan_Lane has joined #openstack-dev | 17:10 | |
| *** jog0 has joined #openstack-dev | 17:10 | |
| *** jog0 has quit IRC | 17:11 | |
| *** jog0 has joined #openstack-dev | 17:11 | |
| *** jdg has quit IRC | 17:12 | |
| *** jdg has joined #openstack-dev | 17:14 | |
| *** rods has quit IRC | 17:15 | |
| *** jdg has quit IRC | 17:22 | |
| * jaypipes re-nicks bcwaldon to Sir Glancelot | 17:23 | |
| dprince | bwaldon: whom wilt be thy queen? | 17:25 |
| bcwaldon | are you proposing? | 17:26 |
| jaypipes | lol | 17:26 |
| dprince | I'm a prince sir! | 17:26 |
| jaypipes | :) | 17:26 |
| *** rods has joined #openstack-dev | 17:27 | |
| jog0 | +1 | 17:28 |
| *** lloydde has quit IRC | 17:32 | |
| jaypipes | jeblair: ping | 17:34 |
| *** lloydde has joined #openstack-dev | 17:34 | |
| *** dubsquared has joined #openstack-dev | 17:37 | |
| jeblair | jaypipes: pong | 17:38 |
| jaypipes | jeblair: filed a bug on it: https://bugs.launchpad.net/openstack-ci/+bug/954258 | 17:38 |
| uvirtbot | Launchpad bug 954258 in openstack-ci "Glance functional tests for Swift and S3 are being skipped in gate jobs" [High,New] | 17:38 |
| jaypipes | jeblair: not sure when this changed, but I had provided mtaylor two config files to be injectetd into the build slaves that contained Cloud Files and amazon S3 credentials to use in the Swift and S3 functional tests,. | 17:38 |
| jaypipes | jeblair: hope it's just a case of reenabling those config file injections | 17:39 |
| *** mszilagyi has joined #openstack-dev | 17:39 | |
| *** dneary has quit IRC | 17:43 | |
| *** bepernoot has joined #openstack-dev | 17:45 | |
| *** mattstep has quit IRC | 17:46 | |
| *** mattstep has joined #openstack-dev | 17:46 | |
| *** novas0x2a|laptop has joined #openstack-dev | 17:48 | |
| *** bepernoot has quit IRC | 17:48 | |
| *** bepernoot has joined #openstack-dev | 17:50 | |
| gmi_ | bcwaldon: I'm back and alive :) | 17:50 |
| bcwaldon | gmi_: ok, conferred with sleepsonzzz that you will need to talk to vishy | 17:51 |
| gmi_ | thanks, should I send him an email? | 17:51 |
| *** derekh has quit IRC | 17:51 | |
| *** oneiroi has quit IRC | 17:55 | |
| *** andrewsmedina has quit IRC | 17:55 | |
| *** adjohn has joined #openstack-dev | 17:58 | |
| gmi_ | vishy: any idea if it is normal behaviour in the trunk code to have all the networks created on a nova-compute host even if only one tenant has instances there? | 17:59 |
| *** _adjohn has joined #openstack-dev | 18:00 | |
| vishy | gmi_: catching up | 18:00 |
| vishy | gmi_: it definitely did not used to work that way | 18:00 |
| gmi_ | ok, thanks | 18:01 |
| vishy | gmi_: so something is awry | 18:01 |
| gmi_ | I know, that's why I was surprised :( | 18:01 |
| *** wwkeyboard2 is now known as wwkeyboard | 18:01 | |
| gmi_ | Nova is installed from git with the last commit being fd80f1b2e1dc03b77fa51cfa27dedce1c593201d | 18:02 |
| *** adjohn has quit IRC | 18:02 | |
| *** _adjohn is now known as adjohn | 18:02 | |
| *** cp16net_ has joined #openstack-dev | 18:03 | |
| vishy | gmi_: interesting | 18:04 |
| vishy | gmi_: did you set a fixed ip for your hosts in the db? | 18:04 |
| gmi_ | no | 18:04 |
| vishy | gmi_: so they all got randomly assigned ips in each vlan? | 18:05 |
| gmi_ | I only ran two commands "nova-manage network create --multi_host=T --network_size=256 --num_networks=100 --bridge_interface=eth1 --fixed_range_v4=10.0.0.0/16 --label=private" | 18:05 |
| gmi_ | and "nova-manage floating create --ip_range=172.16.31.224/27" | 18:05 |
| *** tryggvil_ has quit IRC | 18:05 | |
| *** cp16net has quit IRC | 18:06 | |
| *** cp16net_ is now known as cp16net | 18:06 | |
| gmi_ | I have some problems with the compute node right now (instances stucked in networking status) but the dnsmasq and bridges are already created by nova | 18:06 |
| vishy | ok i think i may see the problem | 18:06 |
| vishy | gmi_: if you look at the bridge, does it have an ip on it? | 18:07 |
| vishy | for example, br100 br101 etc. | 18:07 |
| gmi_ | let me check | 18:07 |
| vishy | --bridge_interface= looks wrong | 18:07 |
| *** rackerhacker has joined #openstack-dev | 18:08 | |
| jeblair | jaypipes: swift is expected to fail currently? | 18:10 |
| gmi_ | each bridge has 10.0.x.3, so br101 has 10.0.1.3, br102 has 10.0.2.3 and so on | 18:10 |
| jeblair | jaypipes: specifically "test_copy_from_swift" ? | 18:10 |
| vishy | gmi_: do you have a second node with .4? | 18:11 |
| vishy | and si br100 bridged into vlan100 which is on eth1? | 18:11 |
| gmi_ | not yet, I could build one by tomorrow | 18:11 |
| vishy | gmi_: just curious | 18:11 |
| *** reed has joined #openstack-dev | 18:12 | |
| *** johnpostlethwait has joined #openstack-dev | 18:13 | |
| vishy | tr3buchet, _cerberus_ : made the change you requested https://review.openstack.org/#change,5246 | 18:13 |
| _cerberus_ | thanks dear | 18:13 |
| gmi_ | vishy:chek http://paste.openstack.org/show/9043/ | 18:14 |
| gmi_ | I remember that when I tried to run "nova-manage network create private $FIXED_RANGE 1 $FIXED_NETWORK_SIZE" without the --bridge_interface=eth1 it errored | 18:14 |
| vishy | tr3buchet, jkoelker, Vek: I can't remember which of you untangled fixed ip from floating ips but it looks like we missed one | 18:15 |
| vishy | gmi_: i was thinking you need to specify --vlan_interface | 18:15 |
| vishy | gmi_: but that may be just the config option vs the flag to pass | 18:16 |
| vishy | so i think that is fine | 18:16 |
| jeblair | jaypipes: the config files were in place, but the tests did not have the environment variables set pointing to them. I've set them, as you requested, but the run currently fails: https://jenkins.openstack.org/view/Gate/job/gate-glance-python27/83/console | 18:16 |
| jkoelker | vishy: whereat? | 18:17 |
| vishy | jkoelker: http://paste.openstack.org/show/9044/ | 18:17 |
| jeblair | jaypipes: let me know how you would like to proceed | 18:17 |
| vishy | jkoelker: I think we need an explicit join there now? or two queries? | 18:17 |
| vishy | gmi_: i think the above call is the source of your issue | 18:17 |
| vishy | jkoelker: In fact, I'm not sure how that query works at all | 18:17 |
| jkoelker | yea, let me look at this | 18:18 |
| vishy | jkoelker: awesome afk a few, let me know what you find. | 18:18 |
| jkoelker | the 2 queries would be best for the separation, but i think the join might be more the getr-done so it works for release | 18:18 |
| gmi_ | vishy; did I create the networks with the wrong flag, or has something to do with the code? | 18:22 |
| *** Ryan_Lane1 has joined #openstack-dev | 18:22 | |
| *** maplebed_ has joined #openstack-dev | 18:22 | |
| *** maplebed_ has joined #openstack-dev | 18:23 | |
| *** maplebed is now known as Guest78930 | 18:23 | |
| *** Ryan_Lane1 has quit IRC | 18:23 | |
| *** Ryan_Lane1 has joined #openstack-dev | 18:23 | |
| heckj | FYI: keystone meeting in #openstack-meeting | 18:24 |
| *** Ryan_Lane has quit IRC | 18:26 | |
| *** Guest78930 has quit IRC | 18:26 | |
| andrewbogott | are there release-based (essex, folsom, etc.) branches maintained in gerrit? Or is there just One True Branch? | 18:27 |
| *** mikeyp has left #openstack-dev | 18:28 | |
| *** phorgh has joined #openstack-dev | 18:30 | |
| russellb | andrewbogott: there is a stable/diablo branch. There isn't an essex branch yet. | 18:31 |
| andrewbogott | ok. And if I have code that I want in folsom but not in essex I should just sit on it for now, right? | 18:31 |
| andrewbogott | I guess that's what "there isn't an essex branch yet" means :) | 18:32 |
| *** cp16net has quit IRC | 18:35 | |
| *** cp16net has joined #openstack-dev | 18:38 | |
| *** Ryan_Lane1 is now known as Ryan_Lane | 18:45 | |
| *** mnewby has joined #openstack-dev | 18:51 | |
| *** jakedahn_zz is now known as jakedahn | 18:52 | |
| mnewby | mtaylor: Are you around? The gate-quantum-unittests job appears broken. | 18:52 |
| *** johnpur has joined #openstack-dev | 18:54 | |
| *** ChanServ sets mode: +v johnpur | 18:54 | |
| jeblair | mnewby: attempted fix in place, retriggering. | 18:56 |
| mnewby | jeblair: danke! | 18:58 |
| *** phorgh has quit IRC | 18:59 | |
| heckj | rackerhacker, chmouel: I'm trying with the timezone change and without chmouel's patch to see if that's exhibiting it. | 19:02 |
| rackerhacker | cool beans | 19:02 |
| jkoelker | vishy: http://paste.openstack.org/show/9057/ I think that will fix it | 19:06 |
| jkoelker | it ends up generating http://paste.openstack.org/show/9058/ which looks like it will return all the right networks | 19:07 |
| soren | heckj: For bug 952977, I presume you mean the keystone conf? | 19:12 |
| uvirtbot | Launchpad bug 952977 in keystone "Can't fetch EC2 credentials" [High,Confirmed] https://launchpad.net/bugs/952977 | 19:12 |
| *** gmi_ has quit IRC | 19:12 | |
| *** Guest81752 has joined #openstack-dev | 19:12 | |
| heckj | soren: yes please - what's your config, etc - just trying to get to a repro quickly, didn't want to guess it through | 19:13 |
| soren | heckj: It's whatever devstack spits out if you go with all the defaults. | 19:13 |
| soren | heckj: ...but I'll pastebin it for you, if that helps. | 19:13 |
| heckj | rackerhacker, chmouel: bug 954057 is definitely a timezone issue - it's been in there since the redux merge, we just didn't catch it earlier | 19:14 |
| uvirtbot | Launchpad bug 954057 in keystone "memcache test failing in python 2.6 when system not in UTC" [High,Confirmed] https://launchpad.net/bugs/954057 | 19:14 |
| heckj | soren: no, asserting current devstack is sufficient - gets me what I need. | 19:14 |
| soren | heckj: Alright. Cool. I'll rerun the whole thing to make sure it's still accurate. | 19:15 |
| soren | heckj: Oh, I'm on Ubuntu Precise, by the way. I guess that might be relevant. | 19:15 |
| soren | heckj: /me makes a note on the bug about that. | 19:15 |
| heckj | soren: thxusir! | 19:15 |
| soren | heckj: http://paste.ubuntu.com/882206/ | 19:16 |
| soren | heckj: That's the output from "pip freeze" | 19:16 |
| *** Guest81752 has quit IRC | 19:17 | |
| andrewbogott | ttx or dragondm: Is there away for an extension to add an additional tool to rootwrap? Or, for that matter, is there a way for a /user/ to do so, short of editing the rootwrap code? | 19:20 |
| jaypipes | jeblair: hey, sorry, stepped away... reading back. | 19:20 |
| *** pixelbeat has quit IRC | 19:20 | |
| jaypipes | jeblair: so... yes, Swift is expected to fail. I have a patch ready that fixes that. | 19:21 |
| jeblair | jaypipes: cool, you should be good to go then | 19:21 |
| chmouel | heckj: nice found (about the tz issue) | 19:22 |
| heckj | chmouel: yeah, just gotta fix it now :-) | 19:22 |
| chmouel | hey hey | 19:23 |
| jaypipes | jeblair: https://review.openstack.org/5283 has a fix. | 19:23 |
| *** littleidea has quit IRC | 19:29 | |
| dolphm | jaypipes: just updated https://review.openstack.org/#change,5278 -- i rebased and borked someone's refactoring in patch 1 | 19:30 |
| *** littleidea has joined #openstack-dev | 19:30 | |
| heckj | dolphm: just +2'd it | 19:31 |
| *** adjohn has quit IRC | 19:31 | |
| jaypipes | dolphm: yup, code looks great, thx | 19:31 |
| dolphm | heckj: jaypipes: thanks | 19:31 |
| dolphm | ttx: does "Incomplete" mean "bug is not clearly defined" or "bug is still not fixed"? or either.. | 19:35 |
| *** oneiroi has joined #openstack-dev | 19:36 | |
| *** oneiroi has joined #openstack-dev | 19:38 | |
| ttx | dolphm: "bug is not clearly defined" | 19:42 |
| ttx | dolphm: scroll to bottom of http://wiki.openstack.org/BugsLifecycle | 19:43 |
| ttx | andrewbogott: you need to edit the rootwrap code, at least to list another py file reference to load | 19:44 |
| andrewbogott | ttx: Yep! bug 954375 | 19:45 |
| uvirtbot | Launchpad bug 954375 in nova "rootwrap needs to be extensible" [Undecided,New] https://launchpad.net/bugs/954375 | 19:45 |
| heckj | adam_g - thanks for grabbing that bug! | 19:45 |
| smoser | anotherjesse_zz, https://bugs.launchpad.net/nova/+bug/954377 | 19:46 |
| uvirtbot | Launchpad bug 954377 in nova "ec2 metadata service regressed for block-device-mapping, kernel-id, ramdisk-id, ami-id" [Undecided,New] | 19:46 |
| *** bencherian has quit IRC | 19:47 | |
| *** danwent has joined #openstack-dev | 19:55 | |
| *** pmyers has quit IRC | 19:57 | |
| *** pmyers has joined #openstack-dev | 19:57 | |
| *** belmoreira has quit IRC | 19:59 | |
| *** belmoreira has joined #openstack-dev | 20:03 | |
| *** bhall has joined #openstack-dev | 20:08 | |
| *** bhall has quit IRC | 20:08 | |
| *** bhall has joined #openstack-dev | 20:08 | |
| adam_g | heckj: np, does a traceback ending in NotImplementedError instead of AttributeError satisfy your 'no traceback for errors' constraint? :) | 20:08 |
| *** mnewby has quit IRC | 20:09 | |
| *** mnewby has joined #openstack-dev | 20:10 | |
| *** shevek_ has joined #openstack-dev | 20:11 | |
| heckj | adam_g: minimally sufficient, but I'd prefer to not have a traceback at all :-) | 20:15 |
| * heckj wonders what a good HTTP response for "NotImplemented" is from a REST API | 20:15 | |
| *** roge has joined #openstack-dev | 20:15 | |
| bcwaldon | heckj: 501 | 20:18 |
| heckj | adam_g: can you have it barf up a 501 response to the API request instead of AttributeError | 20:19 |
| heckj | bcwaldon (thank you) | 20:19 |
| *** clayg_ has quit IRC | 20:22 | |
| *** clayg_ has joined #openstack-dev | 20:22 | |
| *** clayg has quit IRC | 20:23 | |
| *** clayg_ is now known as clayg | 20:23 | |
| * heckj needs a cheat sheet of those things | 20:23 | |
| *** clayg_ has joined #openstack-dev | 20:24 | |
| bcwaldon | heckj: http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html | 20:24 |
| *** roge has quit IRC | 20:25 | |
| * heckj pastes that on the wall | 20:25 | |
| *** johnpur has quit IRC | 20:26 | |
| *** johnpur has joined #openstack-dev | 20:29 | |
| *** ChanServ sets mode: +v johnpur | 20:29 | |
| statik | heckj: this one is really nice too http://wiki.basho.com/images/http-headers-status-v3.png | 20:30 |
| dubsquared | +1 ^ | 20:30 |
| *** adjohn has joined #openstack-dev | 20:32 | |
| *** vincentricci has quit IRC | 20:32 | |
| vishy | jkoelker: I think there are issues with subquerys on postgres, but maybe it was just when there is an update involved | 20:33 |
| vishy | jkoelker: yeah i think so, can you prop it? | 20:33 |
| jkoelker | vishy: yea, that i don't know, I know nothing about postgre | 20:34 |
| jkoelker | already did https://review.openstack.org/5282 | 20:34 |
| vishy | jkoelker: yes subquery is ok on non-others | 20:34 |
| *** vincentricci has joined #openstack-dev | 20:34 | |
| vishy | *non-updates | 20:35 |
| jkoelker | ah cools | 20:36 |
| jkoelker | looks like matt sent it in | 20:36 |
| *** belmoreira has quit IRC | 20:37 | |
| *** dtroyer has quit IRC | 20:38 | |
| vishy | smoser: https://bugs.launchpad.net/nova/+bug/954335 | 20:41 |
| uvirtbot | Launchpad bug 954335 in nova "registering new image from existing s3 location does not respect kernel/ramdisk/name" [Undecided,New] | 20:41 |
| vishy | smoser: i don't think we do anything with kernel/ramdisk params to euca-register | 20:41 |
| smoser | i thought that might be the case. | 20:41 |
| smoser | you read them from the manifest | 20:42 |
| smoser | ? | 20:42 |
| smoser | i didn't look | 20:42 |
| smoser | if thats the case, its a pretty easy fix. | 20:42 |
| *** dprince has quit IRC | 20:42 | |
| *** martine has quit IRC | 20:50 | |
| *** anotherjesse_zz is now known as anotherjesse | 20:52 | |
| *** oneiroi has quit IRC | 20:52 | |
| heckj | chmouel (if you're still awake), dolphm - would like a review on https://review.openstack.org/5290 | 20:56 |
| heckj | ^^ resolves the DST issue we're seeing in the unit tests | 20:56 |
| uvirtbot | heckj: Error: "^" is not a valid command. | 20:56 |
| dolphm | heckj: looking | 20:56 |
| chmouel | heckj: sure | 20:56 |
| *** dtroyer_zzz has joined #openstack-dev | 20:56 | |
| *** dtroyer_zzz is now known as dtroyer | 20:57 | |
| anotherjesse | smoser: going to add some tests for the things you found | 20:57 |
| anotherjesse | smoser: any chance you can compare to before the bug landed - since I think some of those pre-existed | 20:57 |
| smoser | anotherjesse, i'm sorry that i suck and i didn't test before. | 20:57 |
| anotherjesse | smoser: regarding https://bugs.launchpad.net/nova/+bug/954377 | 20:57 |
| uvirtbot | Launchpad bug 954377 in nova "ec2 metadata service regressed for block-device-mapping and empty kernel-id, ramdisk-id" [Undecided,New] | 20:57 |
| smoser | some of those definitely *did* exist before (the ami-000000) did i know that. | 20:57 |
| dolphm | heckj: awesome, i just saw this for the first time a few minutes ago | 20:57 |
| heckj | fucking DST… grumble grumble grumble | 20:58 |
| anotherjesse | smoser: the tests for the metadata service were minimal … so you found a lot of tests that need added | 20:58 |
| Vek | +1 | 20:58 |
| heckj | anotherjesse: +1 | 20:58 |
| smoser | anotherjesse, in actual test, the client that is good to test is: python -c 'import boto.utils; boto.utils.get_instance_metadata()' | 20:59 |
| Vek | \o | 20:59 |
| *** davlap has joined #openstack-dev | 21:00 | |
| dolphm | heckj: +2'd | 21:00 |
| smoser | you may be able to mock aroudn that, i'm not sure. | 21:00 |
| heckj | dolphm: thanks! | 21:00 |
| dolphm | heckj: easy review https://review.openstack.org/#change,5289 | 21:01 |
| anotherjesse | smoser: is there a way to point it to a different metadata url? for testing I run outside of an instance usually | 21:01 |
| *** jdg has joined #openstack-dev | 21:01 | |
| anotherjesse | if too hard I'll test from a deploy | 21:01 |
| dolphm | heckj: oh wait, i have a patch lol | 21:01 |
| smoser | boto.utils.get_instance_metadata() will take a url, but i tihnk you'd have to add the x-forwarded-for header... not sure how you'd do that. | 21:01 |
| anotherjesse | smoser: I hardcode in the service ;) | 21:01 |
| vishy | smoser: that is the case, i think we could support passing them as well | 21:02 |
| dolphm | heckj: there we go... https://review.openstack.org/#change,5289 | 21:02 |
| vishy | smoser: can you work around it by setting them in the manifest in cloud register? | 21:02 |
| vishy | smoser: or do we need to get that adddressed for essex? | 21:02 |
| smoser | vishy, well, the first time the do get set in the manifest | 21:02 |
| vishy | smoser: yeah which is why it works | 21:03 |
| smoser | this but is expcliitly when trying to register a second image using the same manifest | 21:03 |
| smoser | ie, i did this: | 21:03 |
| smoser | upload image, forgot to add ramdisk | 21:03 |
| smoser | register with ramdisk | 21:03 |
| smoser | didnt work | 21:03 |
| * anotherjesse broke essex, head hung in shame | 21:03 | |
| *** littleidea has quit IRC | 21:03 | |
| smoser | ie, trying to short cut the 'upload image' step after failure. | 21:03 |
| smoser | and forgot/didn't think about just run-instances --ramdisk | 21:04 |
| smoser | (which does work) | 21:04 |
| *** maplebed_ is now known as maplebed | 21:04 | |
| *** littleidea has joined #openstack-dev | 21:05 | |
| *** lts has quit IRC | 21:07 | |
| *** bencherian has joined #openstack-dev | 21:07 | |
| *** zigo has quit IRC | 21:08 | |
| *** cp16net has quit IRC | 21:09 | |
| chmouel | heckj: that can triaged as low #954030 by the way | 21:10 |
| *** bepernoot has quit IRC | 21:13 | |
| heckj | bug 954030 | 21:13 |
| uvirtbot | Launchpad bug 954030 in keystone "Container Sync and Keystone" [Undecided,New] https://launchpad.net/bugs/954030 | 21:13 |
| heckj | ah, easy link | 21:13 |
| anotherjesse | dolphm: easy to get a test for https://review.openstack.org/#change,5291 | 21:14 |
| heckj | chmouel: done, thanks! | 21:14 |
| dolphm | anotherjesse: true that | 21:14 |
| *** andrewsmedina has joined #openstack-dev | 21:15 | |
| rackerhacker | heckj: nice work on 954057... remind me to buy you a beer if we're in the same place sometime | 21:18 |
| heckj | rackerhacker: I'll be at the summit - the the beer is owed to mguis (one of the guys that works with me at Nebula) who suggested where to look for the issue | 21:21 |
| rackerhacker | two beers then! | 21:21 |
| rackerhacker | i'll be at the summit as well | 21:21 |
| rackerhacker | heckj: tests are all green over here on all the boxes i tried -- good stuff | 21:23 |
| heckj | thanks! | 21:23 |
| *** msinhore has quit IRC | 21:23 | |
| *** nikhil has quit IRC | 21:24 | |
| *** dolphm has quit IRC | 21:28 | |
| Daviey | adam_g: Hey.. Can you field some questions from anotherjesse ? | 21:28 |
| Daviey | anotherjesse: shoot! | 21:28 |
| anotherjesse | Daviey / adam_g: http://etherpad.openstack.org/cliauth-redux | 21:34 |
| Daviey | anotherjesse: Can it not just accept both formats, with the prior deprecated ? | 21:37 |
| anotherjesse | Daviey: definitely | 21:38 |
| anotherjesse | Daviey: although for most of the tools (excluding swift which just shipped this in 1.4.7), there hasn't been a release that supported the prior version) | 21:38 |
| Daviey | oh | 21:38 |
| adam_g | --os_{username, password, etc} seems like a definite improvement | 21:39 |
| anotherjesse | Daviey: this is the first time we have this … hence when working on keystone's client we realized it was confusing to have --username mean "authentication user" not user I'm working with | 21:39 |
| adam_g | i usually rely on the OS_ environment variables when using these tools, anyway | 21:40 |
| anotherjesse | adam_g: it would be a step back if we did this for swift, glance, … but ubuntu's version was different because swift 1.4.7 was different | 21:40 |
| anotherjesse | adam_g: yeah, me too | 21:40 |
| anotherjesse | we can work up the proposals to the projects in the next (dtroyer estimate here) | 21:41 |
| anotherjesse | but if the clock is out, it would be better to doc what we have | 21:41 |
| adam_g | anotherjesse: not sure i follow... swift 1.4.8 releases tomorrow? if so, we should have that in precise by the weekend | 21:42 |
| anotherjesse | adam_g: 1.4.8 is next week | 21:42 |
| anotherjesse | looking for date | 21:42 |
| anotherjesse | we would need to get it proposed by tomorrow to make 1.4.8 cutoff | 21:43 |
| adam_g | anotherjesse: 1.4.8 is the swift target for essex, though? | 21:43 |
| dtroyer | anotherjesse: day or two max | 21:43 |
| anotherjesse | adam_g: yes | 21:43 |
| anotherjesse | adam_g: in #openstack-meeting Daviey had mentioned 1.4.7 might be what LTS has | 21:43 |
| Daviey | anotherjesse: it's still a consideration, not set it stone. | 21:44 |
| anotherjesse | Daviey: theoretically you could backport the patch to prepend os_ | 21:44 |
| dtroyer | also, I'm not certain that the presence of --username was ever documented for swift, it's not in the help text | 21:45 |
| Daviey | anotherjesse: right, that should be trivial, right? | 21:45 |
| anotherjesse | dtroyer: I think the blueprint is the only place that we talked about it iirc | 21:46 |
| dtroyer | it still shows their original --user | 21:46 |
| anotherjesse | dtroyer: we should get buy in from glance/swift that they are good with the patches before doing any others | 21:47 |
| dtroyer | ok | 21:47 |
| anotherjesse | if ptl thinks it is a good idea, we can send the proposal to the email list | 21:47 |
| anotherjesse | and start work | 21:47 |
| *** 31NAA1G4I has quit IRC | 21:48 | |
| *** joesavak has joined #openstack-dev | 21:50 | |
| *** rods has quit IRC | 21:50 | |
| vishy | Daviey, smoser, adam_g: what are you guys doing for upload_bundle support ? | 21:51 |
| vishy | are you suggesting people use nova-objectstore? | 21:52 |
| anotherjesse | dtroyer: once notmyname and jaypipes & bcwaldon agree, we should update http://wiki.openstack.org/CLIAuth | 21:52 |
| anotherjesse | as well | 21:52 |
| heckj | I ran into an interesting deployment issue that touchs on CLIAuth as well | 21:54 |
| anotherjesse | heckj: what's that? | 21:54 |
| heckj | The default completion place for novaclient is set via an ENV VAR - with a default of ~/.… - but that's really akward for a dashboard instance running as www-data | 21:55 |
| heckj | especially if you're not using Apache (like, say, gunicorn+nginx) | 21:55 |
| heckj | It's (so far) only in novalcient for the bash completiion, which is really cool - but it messes up using the client directly. | 21:56 |
| heckj | (er, it can) | 21:56 |
| Daviey | vishy: do we have any other options? | 21:57 |
| Daviey | (considering we love the ec2 api etc.) | 21:57 |
| anotherjesse | Daviey: you can run swift with S3 middleware and do the image registry magic | 21:58 |
| vishy | well not quite | 21:58 |
| vishy | that is why i was asking | 21:58 |
| anotherjesse | darn you vishy | 21:58 |
| vishy | we are about 95% | 21:58 |
| chmouel | yeah i need to finish the last bit | 21:58 |
| vishy | chmouel: you are here! | 21:58 |
| anotherjesse | chmouel: is there a bug? | 21:58 |
| Daviey | anotherjesse: Hmm.. i thought it had to be proxied through nov-objstore ? | 21:59 |
| vishy | chmouel: I need to know if the patch i sent you for nova works | 21:59 |
| Daviey | vishy: wow, you are working on this | 21:59 |
| vishy | chmouel: if so I have to get it in to essex | 21:59 |
| chmouel | anotherjesse: i don't think i had one logged, I shall do that now | 21:59 |
| vishy | Daviey: yes, we've had chmouel on it for weeks now | 21:59 |
| anotherjesse | bad chmouel for no bug! | 21:59 |
| chmouel | vishy: give me until tomorrow i was travelling lately | 22:00 |
| Daviey | vishy: You sir, have made an Englishman very happy. | 22:00 |
| vishy | Daviey: trying to get s3 front end for swift working with objectstore | 22:00 |
| Daviey | anotherjesse: we have a bug about the security of objectstore. | 22:00 |
| vishy | Daviey: the issue is nova needs to be able to get all buckets | 22:00 |
| vishy | Daviey: but it talks to the objectstore over s3 | 22:00 |
| *** jdg has quit IRC | 22:00 | |
| vishy | Daviey: and we don't have a good way of telling swift which tenant the request is for | 22:01 |
| Daviey | vishy: right! I think smoser raised that bug. | 22:01 |
| vishy | Daviey: here is my current strategy: https://github.com/vishvananda/nova/commit/65f1f6a723b3c2e5714ef2adbc1e4fc12445ba62 | 22:03 |
| chmouel | Daviey, vishy: 954505 | 22:03 |
| chmouel | Daviey, vishy: https://bugs.launchpad.net/nova/+bug/954505 | 22:03 |
| uvirtbot | Launchpad bug 954505 in nova "Nova s3 should be able to talk to swift " [Undecided,New] | 22:03 |
| vishy | I'm just waiting chmouel to verify that he can make the middleware support it | 22:03 |
| anotherjesse | notmyname: do you have any comments on http://etherpad.openstack.org/cliauth-redux or if we get you a patch today will you be cool? | 22:03 |
| *** danwent has quit IRC | 22:03 | |
| anotherjesse | with the proposal | 22:03 |
| *** rods has joined #openstack-dev | 22:03 | |
| anotherjesse | jaypipes / bcwaldon: same question to you | 22:03 |
| *** danwent has joined #openstack-dev | 22:04 | |
| anotherjesse | heckj / vishy: as well, but I think you already agree given the pains of using keystoneclient | 22:04 |
| vishy | chmouel: i think my approach above with the accesskey is better than the url because it doesn't break other s3 implementations like nova-objectstore | 22:04 |
| notmyname | anotherjesse: looks fine (if I'm correct in seeing it's just a change in the arg names) | 22:04 |
| *** cp16net has joined #openstack-dev | 22:05 | |
| *** davidkranz has quit IRC | 22:05 | |
| notmyname | Daviey: I had to leave the meeting right after the swift time. I'd love to know your concerns with swift 1.4.8 | 22:05 |
| bcwaldon | anotherjesse: +2 | 22:05 |
| vishy | anotherjesse: I'm ok with it, although it hasn't been painful for me :) | 22:05 |
| *** johnpur has quit IRC | 22:05 | |
| notmyname | Daviey: (unfortunately, I do have to go again very soon. but I do want to know) | 22:06 |
| anotherjesse | notmyname: yes - just prepending os_ to keystone auth params | 22:06 |
| ttx | notmyname: I think the concern is that they are past their FeatureFreeze and 1.4.8 adds a significant feature | 22:07 |
| Daviey | notmyname: The release schedule has left us planning for 1.4.7. It's really sucky to be in limbo, when all of our processes were based on expecting one version. | 22:07 |
| notmyname | ah | 22:07 |
| Daviey | notmyname: I'm not saying it's a 'no go'... It's a hesitant position. | 22:07 |
| *** joesavak has quit IRC | 22:08 | |
| *** bsza1 has joined #openstack-dev | 22:08 | |
| notmyname | Daviey: 1.4.7 was originally scheduled for the same date 1.4.8 is now scheduled for. we just had to do a quick important release at rackspace, and we pushed it publicly to keep in sync | 22:08 |
| *** kbringard has quit IRC | 22:08 | |
| notmyname | Daviey: so there is no real change in schedules, just a number | 22:08 |
| heckj | anotherjesse: I'm totally good with it | 22:08 |
| Daviey | notmyname: Caring about Openstack on one head- wanting the latest greatness, and being a member of the Ubuntu Release Team means i also need to be conservative. | 22:09 |
| Daviey | in a word, suck. | 22:09 |
| notmyname | Daviey: I gotta run. will you be on later to talk more? | 22:09 |
| Daviey | notmyname: pushed it publicly? What does that mean? | 22:09 |
| Daviey | notmyname: no, it's late here already | 22:09 |
| Daviey | notmyname: Ping me tomorrow? | 22:10 |
| notmyname | Daviey: ah, ok. let's catch up tomororw | 22:10 |
| *** andrewsmedina has quit IRC | 22:10 | |
| Daviey | sure thing! | 22:10 |
| *** flaviamissi has quit IRC | 22:11 | |
| *** bsza has quit IRC | 22:11 | |
| justinsb | Is there an equivalent of S3's pre-signed requests for swift (without using the S3 API)? | 22:12 |
| chmouel | vishy: I would need to patch s3_token | 22:12 |
| chmouel | vishy: but that should not be too much of work | 22:13 |
| anotherjesse | justinsb: i don't believe so - chmouel know of anything? | 22:14 |
| justinsb | Thanks - I'll probably file a bug. It's a really useful feature.. | 22:15 |
| chmouel | justinsb: no there is none | 22:15 |
| justinsb | chmouel: Thanks for confirming! | 22:17 |
| chmouel | justinsb: sorry just making sure this is not that what you are looking for http://swift.openstack.org/misc.html#module-swift.common.middleware.tempurl | 22:18 |
| justinsb | chmouel: Oh cool ... that looks like exactly what I want (I'm building a Jenkins plugin, the idea is that a jar gets uploaded to Swift and then can be downloaded directly rather than via Jenkins). Thanks. | 22:20 |
| chmouel | justinsb: oh ok sorry i misunderstood yeah you should be able to use that with staticweb | 22:20 |
| chmouel | (in case you were planning to do it with swift+keystone then unfortunately that doesn't work just yet) | 22:21 |
| *** pixelbeat has joined #openstack-dev | 22:21 | |
| justinsb | chmouel: Ah. I was planning on using Keystone. Thanks for the warning. | 22:23 |
| *** j05h has quit IRC | 22:23 | |
| *** j05h has joined #openstack-dev | 22:28 | |
| *** j05h has quit IRC | 22:32 | |
| vishy | justinsb: what are pre-signed requests? | 22:34 |
| *** roge has joined #openstack-dev | 22:34 | |
| justinsb | vishy: It's a cool feature so e.g. Jenkins has the keys for a private Swift bucket, but can construct a URL so that a browser can download from that swift bucket directly, without the client discovering the password. | 22:36 |
| vishy | justinsb: gotcha | 22:37 |
| *** j05h has joined #openstack-dev | 22:37 | |
| justinsb | vishy: The companies that claim to have discovered a novel way to sell files online are rediscovering or repackaging this. | 22:37 |
| vishy | justinsb: tee hee | 22:38 |
| *** roge has quit IRC | 22:38 | |
| *** bsza1 has quit IRC | 22:42 | |
| *** cp16net has quit IRC | 22:42 | |
| *** villep_ has quit IRC | 22:42 | |
| *** villep has joined #openstack-dev | 22:42 | |
| *** AlanClark has quit IRC | 22:43 | |
| *** j05h has quit IRC | 22:43 | |
| *** j05h has joined #openstack-dev | 22:44 | |
| bcwaldon | annegentle: https://review.openstack.org/#change,5301 | 22:49 |
| *** dtroyer_ has joined #openstack-dev | 22:49 | |
| *** vricci has joined #openstack-dev | 22:49 | |
| *** dtroyer_ has quit IRC | 22:50 | |
| *** vincentricci has quit IRC | 22:51 | |
| *** vricci is now known as vincentricci | 22:51 | |
| *** statik has quit IRC | 22:53 | |
| *** ywu has joined #openstack-dev | 22:54 | |
| *** Glacee has quit IRC | 22:55 | |
| jog0 | https://bugs.launchpad.net/nova/+bug/954553 | 22:55 |
| uvirtbot | Launchpad bug 954553 in nova "tools/conf/generate_sample.sh doesn't handle code as default value" [Undecided,New] | 22:55 |
| *** dolphm has joined #openstack-dev | 22:56 | |
| *** statik has joined #openstack-dev | 22:59 | |
| *** statik has joined #openstack-dev | 22:59 | |
| *** lloydde has quit IRC | 23:06 | |
| *** dtroyer is now known as dtroyer_zzz | 23:08 | |
| *** devananda has joined #openstack-dev | 23:09 | |
| sleepsonzzz | jeblair: just fyi, looks like merge conflicts are not caught before integration tests for devstack: https://jenkins.openstack.org/job/gate-integration-tests-devstack-vm/2517/console | 23:10 |
| *** berendt has quit IRC | 23:15 | |
| *** anotherjesse is now known as anotherjesse_zz | 23:16 | |
| justinsb | I'm using FlatManager, and the /os-floating-ips gives a 500 error (get_floating_ips_by_project not defined on FlatManager). This stops horizon from launching an instance. I can launch an instance through the API. Is this a bug in Horizon, Nova or both? | 23:16 |
| *** anotherjesse_zz is now known as anotherjesse | 23:17 | |
| *** anotherjesse is now known as anotherjesse_zz | 23:18 | |
| jaypipes | anotherjesse_zz: my only thought would be to mention the -S --strategy option... | 23:24 |
| *** thingee has quit IRC | 23:31 | |
| *** ayoung has joined #openstack-dev | 23:31 | |
| *** agonella has joined #openstack-dev | 23:31 | |
| jeblair | sleepsonzzz: yeah, we have the -merge test to check that and report something sensible back to gerrit, but it's not straightforward to chain other tests to that (at the moment), so they still run anyway. | 23:32 |
| *** dtroyer_zzz is now known as dtroyer | 23:33 | |
| *** flaviamissi has joined #openstack-dev | 23:34 | |
| *** flaviamissi has quit IRC | 23:38 | |
| *** davlap has quit IRC | 23:42 | |
| *** anotherjesse_zz is now known as anotherjesse | 23:42 | |
| *** davlap has joined #openstack-dev | 23:42 | |
| *** hhoover has joined #openstack-dev | 23:43 | |
| *** davlap has quit IRC | 23:44 | |
| *** davlap has joined #openstack-dev | 23:44 | |
| *** lloydde has joined #openstack-dev | 23:45 | |
| *** davlap has quit IRC | 23:45 | |
| chmouel | vishy: with your patch on nova/s3.py and this patch on s3_token http://pastie.org/private/fwvhtcdqdw7eqw3a3og i got it to register | 23:52 |
| chmouel | vishy: assuming we have secret_key and access_key with a user in reselleradmin account in nova.conf | 23:52 |
| chmouel | vishy: I'll cleanup all of this by tomorrow and this should be ready to review when you guys wake up :) | 23:53 |
| *** hhoover has quit IRC | 23:54 | |
| *** dolphm has quit IRC | 23:54 | |
| *** danwent has quit IRC | 23:54 | |
| *** danwent has joined #openstack-dev | 23:55 | |
| *** jakedahn is now known as jakedahn_zz | 23:56 | |
| *** jog0 has left #openstack-dev | 23:57 | |
| *** jog0 has joined #openstack-dev | 23:57 | |
| *** lloydde has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
