| notmyname | mattoliverau: Tristan Goode <tristan@aptira.com> | 00:01 |
|---|---|---|
| *** dmorita has joined #openstack-swift | 00:21 | |
| *** zhiyan_ is now known as zhiyan | 00:27 | |
| *** matsuhashi has joined #openstack-swift | 00:27 | |
| *** nosnos has joined #openstack-swift | 00:32 | |
| notmyname | I'm packing up for the night. talk to ya later | 00:39 |
| zaitcev | Good luck with your surgery. | 00:41 |
| mattoliverau | notmyname: night, good luck tomorrow! | 00:42 |
| *** shri has quit IRC | 00:44 | |
| *** dmsimard is now known as dmsimard_away | 01:08 | |
| *** haomaiwang has quit IRC | 01:09 | |
| *** kashyapk has joined #openstack-swift | 01:15 | |
| *** kashyapk has quit IRC | 01:20 | |
| *** gyee has quit IRC | 01:28 | |
| *** nsquare has quit IRC | 02:23 | |
| *** bkopilov_ has quit IRC | 02:48 | |
| *** zhiyan is now known as zhiyan_ | 03:30 | |
| *** zhiyan_ is now known as zhiyan | 03:41 | |
| *** sombrafam has joined #openstack-swift | 03:42 | |
| *** nosnos has quit IRC | 03:44 | |
| *** madhuri has joined #openstack-swift | 03:46 | |
| *** matsuhashi has quit IRC | 04:11 | |
| *** haomaiwa_ has joined #openstack-swift | 04:18 | |
| *** haomaiw__ has joined #openstack-swift | 04:20 | |
| *** haoma____ has joined #openstack-swift | 04:21 | |
| *** haomaiwa_ has quit IRC | 04:21 | |
| *** haomaiw__ has quit IRC | 04:23 | |
| *** haomaiwang has joined #openstack-swift | 04:24 | |
| *** haoma____ has quit IRC | 04:27 | |
| *** bkopilov_ has joined #openstack-swift | 04:32 | |
| *** matsuhashi has joined #openstack-swift | 04:48 | |
| *** nosnos has joined #openstack-swift | 04:48 | |
| *** kashyapk has joined #openstack-swift | 04:55 | |
| *** psharma has joined #openstack-swift | 04:57 | |
| *** haomaiwang has quit IRC | 05:02 | |
| *** haomaiwa_ has joined #openstack-swift | 05:03 | |
| *** gvernik has joined #openstack-swift | 05:03 | |
| *** gvernik has quit IRC | 05:04 | |
| *** Longgeek_ has joined #openstack-swift | 05:21 | |
| *** ppai has joined #openstack-swift | 05:25 | |
| *** deep_ has joined #openstack-swift | 05:34 | |
| deep_ | Hello, I have configured swift with keystone(Active Directory), while creating container i am getting error in syslog "Jun 11 22:52:36 node1 proxy-server: Unexpected response from keystone service: {u'error': {u'message': u'Could not find role, swift.', u'code': 404, u'title': u'Not Found'}} Jun 11 22:52:36 node1 proxy-server: Authorization failed for token 9ece2e7637704563fc61eff8d53e0508 Jun 11 22:52:36 node1 proxy-server: | 05:34 |
| deep_ | any idea what is missing ? | 05:34 |
| *** zaitcev has quit IRC | 05:52 | |
| mattoliverau | deep_: I'm no expert, so don't know if I can help, but the error seems to state that keystone couldn't find the role 'swift'. Does it exist in keystone? | 05:59 |
| *** haomaiwa_ has quit IRC | 06:00 | |
| *** sombrafam has quit IRC | 06:19 | |
| hugokuo | hole | 06:20 |
| hugokuo | hola | 06:20 |
| *** pberis has quit IRC | 06:26 | |
| *** pberis has joined #openstack-swift | 06:27 | |
| *** mmcardle has joined #openstack-swift | 06:43 | |
| *** jamiehannaford has joined #openstack-swift | 06:46 | |
| *** pconstantine has joined #openstack-swift | 06:55 | |
| mattoliverau | hugokuo: hola! | 06:57 |
| *** pconstantine has quit IRC | 07:01 | |
| *** kashyapk has quit IRC | 07:02 | |
| *** pconstantine has joined #openstack-swift | 07:02 | |
| *** kashyapk has joined #openstack-swift | 07:02 | |
| *** kashyapk has quit IRC | 07:07 | |
| *** pconstantine has quit IRC | 07:07 | |
| *** pconstantine has joined #openstack-swift | 07:08 | |
| *** haomaiwa_ has joined #openstack-swift | 07:13 | |
| *** 5EXABFP2R has joined #openstack-swift | 07:21 | |
| *** haomaiwa_ has quit IRC | 07:25 | |
| mattoliverau | Well I'm calling it a day, night all. | 07:33 |
| *** kashyapk has joined #openstack-swift | 07:36 | |
| *** pconstantine has quit IRC | 07:39 | |
| *** pconstantine has joined #openstack-swift | 07:39 | |
| *** mlipchuk has joined #openstack-swift | 07:43 | |
| openstackgerrit | Takashi Kajinami proposed a change to openstack/swift: Make container PUT and DELETE to allow asynchronous updating to account https://review.openstack.org/87882 | 07:45 |
| openstackgerrit | Yuan Zhou proposed a change to openstack/python-swiftclient: Allow to specify storage policy when uploading objects https://review.openstack.org/73920 | 07:47 |
| *** StevenK has quit IRC | 07:50 | |
| *** StevenK has joined #openstack-swift | 07:51 | |
| *** pconstantine has quit IRC | 08:02 | |
| *** fbo_away is now known as fbo | 08:14 | |
| openstackgerrit | Takashi Kajinami proposed a change to openstack/swift: Ensure update of the container by object-udater https://review.openstack.org/99598 | 08:18 |
| openstackgerrit | Takashi Kajinami proposed a change to openstack/swift: Ensure update of the container by object-updater https://review.openstack.org/99598 | 08:20 |
| *** haomaiwang has joined #openstack-swift | 08:22 | |
| *** 5EXABFP2R has quit IRC | 08:23 | |
| *** foexle has joined #openstack-swift | 08:35 | |
| *** kashyapk has quit IRC | 09:20 | |
| *** kashyapk has joined #openstack-swift | 09:21 | |
| *** zhiyan is now known as zhiyan_ | 09:23 | |
| *** kashyapk has quit IRC | 09:25 | |
| *** kashyapk has joined #openstack-swift | 09:42 | |
| *** dmorita has quit IRC | 09:46 | |
| *** kashyapk has quit IRC | 10:13 | |
| *** kashyapk has joined #openstack-swift | 10:14 | |
| *** kashyapk has quit IRC | 10:18 | |
| *** matsuhashi has quit IRC | 10:20 | |
| *** matsuhashi has joined #openstack-swift | 10:20 | |
| *** NM has joined #openstack-swift | 10:24 | |
| *** flowouffff has joined #openstack-swift | 10:25 | |
| *** matsuhashi has quit IRC | 10:25 | |
| *** haomaiwang has quit IRC | 10:25 | |
| *** haomaiwang has joined #openstack-swift | 10:26 | |
| deep_ | mattoliverau : I added the swift role under OU=Roles now error changed to "Jun 12 03:18:42 node1 proxy-server: Authorization failed for token e2e6e8b09916ac22dd714e3dbc155902 Jun 12 03:18:42 node1 proxy-server: Invalid user token - deferring reject downstream" | 10:34 |
| *** NM has quit IRC | 10:41 | |
| openstackgerrit | Alistair Coles proposed a change to openstack/python-swiftclient: Fix py33 gate issue with setuptools https://review.openstack.org/99624 | 10:47 |
| *** ppai has quit IRC | 10:51 | |
| *** mkollaro has joined #openstack-swift | 10:52 | |
| *** haomaiwang has quit IRC | 10:56 | |
| *** NM has joined #openstack-swift | 10:59 | |
| *** NM has quit IRC | 10:59 | |
| *** kashyapk has joined #openstack-swift | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:03 | |
| *** nosnos has quit IRC | 11:13 | |
| *** kashyapk has quit IRC | 11:23 | |
| *** kashyapk has joined #openstack-swift | 11:23 | |
| *** dmsimard_away is now known as dmsimard | 11:23 | |
| *** shakayumi has joined #openstack-swift | 11:24 | |
| *** ppai has joined #openstack-swift | 11:34 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/python-swiftclient: Mask token in debug output https://review.openstack.org/99632 | 11:38 |
| *** jamiehan_ has joined #openstack-swift | 11:39 | |
| *** jamiehannaford has quit IRC | 11:41 | |
| *** jamiehan_ is now known as jamiehannaford | 11:45 | |
| *** shakayumi has quit IRC | 11:50 | |
| *** mmcardle has quit IRC | 11:51 | |
| *** nsquare has joined #openstack-swift | 11:52 | |
| *** mlipchuk has left #openstack-swift | 11:56 | |
| *** mlipchuk has joined #openstack-swift | 11:59 | |
| *** tdasilva has joined #openstack-swift | 12:01 | |
| *** jamiehannaford has quit IRC | 12:14 | |
| *** jamiehannaford has joined #openstack-swift | 12:15 | |
| *** kashyapk has quit IRC | 12:20 | |
| *** kashyapk has joined #openstack-swift | 12:21 | |
| *** deep_ has quit IRC | 12:23 | |
| *** jamiehan_ has joined #openstack-swift | 12:23 | |
| *** jamiehannaford has quit IRC | 12:24 | |
| *** lpabon has joined #openstack-swift | 12:28 | |
| *** kashyapk has quit IRC | 12:29 | |
| *** ppai has quit IRC | 12:33 | |
| *** mmcardle has joined #openstack-swift | 12:38 | |
| *** psharma has quit IRC | 12:41 | |
| *** chandan_kumar has joined #openstack-swift | 12:41 | |
| *** chandan_kumar has quit IRC | 12:44 | |
| *** omame has quit IRC | 12:46 | |
| *** openstackgerrit has quit IRC | 12:46 | |
| *** openstackgerrit has joined #openstack-swift | 12:47 | |
| *** NM has joined #openstack-swift | 12:47 | |
| *** CaioBrentano has joined #openstack-swift | 12:51 | |
| *** chandankumar has joined #openstack-swift | 12:52 | |
| *** bkopilov_ has quit IRC | 12:56 | |
| *** haomaiwang has quit IRC | 12:57 | |
| *** haomaiwa_ has joined #openstack-swift | 12:58 | |
| *** haomaiw__ has joined #openstack-swift | 12:59 | |
| *** haomaiwa_ has quit IRC | 13:02 | |
| *** jamiehannaford has joined #openstack-swift | 13:03 | |
| *** jamiehan_ has quit IRC | 13:06 | |
| chmouel | I don't think that benchmark is fair http://blog.zhaw.ch/icclab/evaluating-the-performance-of-ceph-and-swift-for-object-storage-on-small-clusters/ (but obv I and most of us here are biased) | 13:15 |
| *** mkollaro has quit IRC | 13:22 | |
| tdasilva | what is swift 2.0.3? | 13:22 |
| chmouel | i am not sure maybe he talks about the client | 13:23 |
| *** kashyapk has joined #openstack-swift | 13:23 | |
| tdasilva | ok | 13:24 |
| *** jergerber has joined #openstack-swift | 13:32 | |
| *** mkollaro has joined #openstack-swift | 13:40 | |
| portante | chmouel: it seems like comparing the librados layer which has different semantics than the Swift API layer does not make much sense | 13:40 |
| chmouel | portante: yeah that's what i commented in that blog post | 13:41 |
| portante | you can't do large objects with librados, if I understand correctly, the client has to handle all that | 13:41 |
| portante | chmouel: yes, thanks for pointing that out | 13:41 |
| portante | I personally believe this kind of comparisons are healthy ways to get different technology strengths and weaknesses exposed, so that folks know where to position them, and improve them | 13:42 |
| *** sileht has quit IRC | 13:43 | |
| chmouel | portante: indeed, this is what glance is doing https://git.openstack.org/cgit/openstack/glance/tree/glance/store/rbd.py#n347 | 13:44 |
| *** haomaiw__ has quit IRC | 13:48 | |
| *** haomaiwa_ has joined #openstack-swift | 13:56 | |
| *** Longgeek_ has quit IRC | 13:57 | |
| *** sileht has joined #openstack-swift | 14:02 | |
| portante | chmouel: not sure I understand what they are doing there, don't know the rados code well enough | 14:09 |
| portante | the paper above talks about swift 2.0.3, does anybody know what that might mean? | 14:09 |
| chmouel | i think it's the swiftclient | 14:10 |
| portante | then we don't even know what version of swift they are comparing against? | 14:10 |
| portante | and it brings up the apparent bias towards ceph in how they consider the system | 14:11 |
| portante | chmouel: and this appears to show ceph is good for small files on a small cluster | 14:13 |
| chmouel | portante: i don't think the benchmarker knows very well swift | 14:16 |
| portante | agreed | 14:16 |
| chmouel | portante: yeah but either way if writing to the ods (object server in ceph) straight from the client, it would be better for small files, right? | 14:17 |
| *** haomaiwa_ has quit IRC | 14:22 | |
| *** haomaiwa_ has joined #openstack-swift | 14:23 | |
| *** haomaiwa_ has quit IRC | 14:29 | |
| *** haomaiwang has joined #openstack-swift | 14:30 | |
| *** bkopilov_ has joined #openstack-swift | 14:34 | |
| *** kashyapk_ has joined #openstack-swift | 14:39 | |
| *** kashyapk has quit IRC | 14:39 | |
| *** mkerrin has quit IRC | 14:44 | |
| *** kashyapk has joined #openstack-swift | 14:47 | |
| *** kashyapk_ has quit IRC | 14:47 | |
| *** byeager has joined #openstack-swift | 14:51 | |
| *** elmiko has joined #openstack-swift | 14:59 | |
| elmiko | hey all, is it possible to revoke, or force early expiration of, a TempURL? | 15:00 |
| *** kashyapk has quit IRC | 15:01 | |
| *** kevinc_ has joined #openstack-swift | 15:05 | |
| ahale | only by changing the key on the account, but that revokes all created with that key | 15:05 |
| elmiko | ahale: ok, thanks | 15:06 |
| elmiko | sounds like not a good option lol | 15:06 |
| *** foexle has quit IRC | 15:17 | |
| *** byeager has quit IRC | 15:28 | |
| *** byeager has joined #openstack-swift | 15:28 | |
| *** gyee has joined #openstack-swift | 15:30 | |
| creiht | chmouel, portante: heh yeah, I'm working on a response this morning | 15:30 |
| creiht | portante: and I agree that it is a good data point, I just want to make sure they are comparing apples to apples :) | 15:32 |
| creiht | well it is a data point, perhaps not a *good* data point :) | 15:32 |
| *** byeager has quit IRC | 15:33 | |
| creiht | I'm also not a fan of how he mis-characterizes how some of the things work in swift | 15:34 |
| *** jamiehannaford has quit IRC | 15:36 | |
| *** jamiehannaford has joined #openstack-swift | 15:37 | |
| *** jamiehannaford has quit IRC | 15:38 | |
| *** jamiehannaford has joined #openstack-swift | 15:40 | |
| creiht | he also optimizes the ceph workload a bit by puting the journal on a separate disk | 15:40 |
| creiht | which is also going to specifically boost small write speeds | 15:41 |
| creiht | It also doesn't make sense to me why the 1 container 4kb reads performed differently than the 20 container 4kb reads | 15:45 |
| creiht | I'm also curious how a crush look up could be more efficient than a ring lookup | 15:49 |
| creiht | I wonder how much he is seeing is http overhead | 15:50 |
| glange | elmiko: you could move or delete the file? | 15:51 |
| elmiko | glange: thanks, that's another option | 15:52 |
| *** sombrafam has joined #openstack-swift | 15:52 | |
| creiht | elmiko: you can change the secret key | 15:52 |
| creiht | for the account | 15:53 |
| glange | creiht: ahale said that :) | 15:53 |
| elmiko | creiht: thanks | 15:53 |
| creiht | ahh | 15:53 |
| creiht | hehe | 15:53 |
| creiht | sorry | 15:53 |
| elmiko | i was hoping there was a call or something to just revoke a TempURL | 15:53 |
| *** CaioBrentano has quit IRC | 15:54 | |
| elmiko | i'm investigating a use case where we might not know the exact expiration time needed, TempURL was suggested as something to look into. i'm trying to understand all the factors that might affect what we want to do. | 15:54 |
| creiht | ahh, yeah usually tempurls are generated with very short expiration times | 15:55 |
| elmiko | that's what it's looking like, but i want to make sure i've done the due dilligence :) | 15:55 |
| creiht | elmiko: can you describe the use case a bit more, and maybe we can give you a better way? | 15:55 |
| creiht | certainly :) | 15:55 |
| elmiko | well the main problem has to do with a security issue we are working on. we need access to swift objects and we want to become more secure about how we handle credentials. TempURL was suggested as a possible workaround. | 15:56 |
| creiht | ahh | 15:56 |
| *** pconstantine has joined #openstack-swift | 15:57 | |
| elmiko | so, i'm trying to workup a matrix of all our options. | 15:57 |
| *** NM has quit IRC | 15:57 | |
| *** byeager has joined #openstack-swift | 15:58 | |
| creiht | elmiko: well be sure to let us know any ways that we can help | 16:00 |
| elmiko | creiht: definitely, i will most likely have more questions :) | 16:01 |
| portante | creiht: agreed, it does not appear that they really worked towards an apples-to-apples comparison, but are trying to justify a position they hold | 16:03 |
| *** byeager has quit IRC | 16:04 | |
| creiht | yeah | 16:04 |
| portante | there are those that embrace the challenge of performance analysis meeting a level of open and objectivity (as much as possible) and those that don't | 16:04 |
| portante | openness | 16:04 |
| *** byeager has joined #openstack-swift | 16:04 | |
| creiht | well it is hard to be totally objective always | 16:04 |
| portante | yes, that is the challenge | 16:04 |
| creiht | I'm beginning to wonder if it is worth rebuting, or just leave as is | 16:08 |
| portante | your points above might be good as questions to get clarified | 16:09 |
| *** pconstantine has quit IRC | 16:10 | |
| creiht | I wonder if something was misconfigured | 16:12 |
| creiht | there is no way the same write workload for 1 container would be faster than for 20 containers | 16:12 |
| creiht | or why reads would be any different for either | 16:12 |
| creiht | heh | 16:14 |
| creiht | and the one point where swift out-performs, he calls swift more greedy | 16:14 |
| *** elmiko is now known as _elmiko | 16:17 | |
| *** kevinc_ has quit IRC | 16:35 | |
| *** openstackgerrit has quit IRC | 16:38 | |
| *** byeager has quit IRC | 16:40 | |
| *** byeager has joined #openstack-swift | 16:41 | |
| *** byeager has quit IRC | 16:45 | |
| *** nsquare has quit IRC | 16:47 | |
| *** byeager has joined #openstack-swift | 16:49 | |
| *** mkerrin has joined #openstack-swift | 16:53 | |
| *** kevinc_ has joined #openstack-swift | 16:54 | |
| *** kashyapk has joined #openstack-swift | 16:59 | |
| *** byeager has quit IRC | 17:01 | |
| *** shri has joined #openstack-swift | 17:02 | |
| *** zaitcev has joined #openstack-swift | 17:12 | |
| *** ChanServ sets mode: +v zaitcev | 17:12 | |
| *** _elmiko is now known as elmiko | 17:14 | |
| *** kevinc_ has quit IRC | 17:17 | |
| *** haomaiwang has quit IRC | 17:22 | |
| *** gyee has quit IRC | 17:23 | |
| *** kevinc_ has joined #openstack-swift | 17:24 | |
| *** nsquare has joined #openstack-swift | 17:30 | |
| *** mmcardle has quit IRC | 17:41 | |
| *** mkollaro has quit IRC | 17:47 | |
| *** ujjain has quit IRC | 17:48 | |
| *** kashyapk has quit IRC | 17:48 | |
| *** kashyapk_ has joined #openstack-swift | 17:48 | |
| goodes | good morning all | 17:50 |
| *** kashyapk_ has quit IRC | 17:53 | |
| *** mkollaro has joined #openstack-swift | 17:54 | |
| *** shri1 has joined #openstack-swift | 18:01 | |
| *** shri1 has quit IRC | 18:01 | |
| *** shri has quit IRC | 18:03 | |
| *** zul has quit IRC | 18:03 | |
| *** zul has joined #openstack-swift | 18:03 | |
| *** nsquare has quit IRC | 18:08 | |
| *** nsquare has joined #openstack-swift | 18:11 | |
| *** openstackgerrit has joined #openstack-swift | 18:12 | |
| *** tdasilva has left #openstack-swift | 18:18 | |
| openstackgerrit | Alex Gaynor proposed a change to openstack/swift: Changed several files to be Python3 friendly https://review.openstack.org/94197 | 18:19 |
| *** byeager has joined #openstack-swift | 18:20 | |
| *** mmcardle has joined #openstack-swift | 18:22 | |
| openstackgerrit | A change was merged to openstack/swift: Container sync no longer sending swift_bytes value https://review.openstack.org/94465 | 18:31 |
| *** shri has joined #openstack-swift | 18:36 | |
| openstackgerrit | A change was merged to openstack/swift: replace string format arguments with function parameters https://review.openstack.org/94331 | 18:52 |
| *** judd7 has joined #openstack-swift | 18:59 | |
| *** sombrafam has quit IRC | 19:02 | |
| *** ahale has quit IRC | 19:02 | |
| *** zz_wasmum has quit IRC | 19:02 | |
| *** ahale has joined #openstack-swift | 19:03 | |
| *** sombrafam has joined #openstack-swift | 19:03 | |
| *** zz_wasmum has joined #openstack-swift | 19:07 | |
| *** nsquare has quit IRC | 19:08 | |
| *** mlipchuk1 has joined #openstack-swift | 19:10 | |
| *** mlipchuk has quit IRC | 19:12 | |
| *** mmcardle has quit IRC | 19:29 | |
| *** mlipchuk1 has quit IRC | 19:39 | |
| *** byeager has quit IRC | 19:40 | |
| *** byeager has joined #openstack-swift | 19:41 | |
| portante | creiht: got read this in more detail myself | 19:43 |
| *** byeager has quit IRC | 19:45 | |
| creiht | portante: did you see my reply? | 19:45 |
| peluse_ | creiht: I did, it was very good | 19:46 |
| creiht | cool, thx | 19:46 |
| creiht | there certainly seems to be something wrong with his config | 19:46 |
| *** IRTermite1 has joined #openstack-swift | 19:47 | |
| portante | creiht: looking now ... | 19:50 |
| creiht | if you look at his other blog posts, they were about ceph, so he certainly is swaying that way a bit | 19:51 |
| portante | yes | 19:51 |
| peluse_ | creiht: you should post to his blog as well (same thing you put in your email) | 19:52 |
| creiht | peluse_: ahh good point | 19:52 |
| glange | it's on now | 19:52 |
| creiht | lol | 19:52 |
| portante | what mailing list is this on? | 19:52 |
| portante | I was looking at the replies on the blog site | 19:53 |
| peluse_ | portante: openstack@lists.openstack.org | 19:54 |
| portante | k thx | 19:54 |
| *** byeager has joined #openstack-swift | 19:55 | |
| creiht | left a comment on the blog as well | 20:01 |
| creiht | but it is awaiting moderation | 20:01 |
| portante | creiht: great, nice response | 20:05 |
| creiht | thx | 20:05 |
| *** shri1 has joined #openstack-swift | 20:11 | |
| peluse_ | anyone from SwiftStack hear how the shoulder-bolting is going/went? | 20:13 |
| *** shri has quit IRC | 20:13 | |
| creiht | peluse_: his wife tweeted that the doc came out to report it went well about 2 hours ago | 20:14 |
| torgomatic | peluse_: all I know, I learned from Twitter https://twitter.com/notmyname/status/477149951219097600 | 20:14 |
| peluse_ | cool, thanks! | 20:14 |
| *** shri1 has quit IRC | 20:15 | |
| peluse_ | creiht: your reply post was 'accepted' and is part of the blog now :) | 20:16 |
| creiht | cool | 20:17 |
| openstackgerrit | A change was merged to openstack/swift: xLO bug with auth tokens expiring during download. https://review.openstack.org/92165 | 20:19 |
| openstackgerrit | A change was merged to openstack/swift: Fix file uploads > 2 GiB in formpost middleware https://review.openstack.org/97828 | 20:23 |
| *** pandemicsyn has quit IRC | 20:28 | |
| *** pandemicsyn has joined #openstack-swift | 20:29 | |
| zaitcev | https://twitter.com/notmyname/status/477184608002273281 "I am wolverine" | 20:34 |
| *** openstackgerrit has quit IRC | 20:35 | |
| *** openstackgerrit has joined #openstack-swift | 20:35 | |
| *** byeager has quit IRC | 20:36 | |
| *** nsquare has joined #openstack-swift | 20:36 | |
| *** byeager has joined #openstack-swift | 20:36 | |
| *** lpabon has quit IRC | 20:39 | |
| creiht | heh... don't let the metal plates get to his head :) | 20:39 |
| glange | do they use standard anesthesia in California or is it more "natural" and "plant-like"? | 20:40 |
| *** gyee has joined #openstack-swift | 20:42 | |
| torgomatic | that's just like, your opinion, man | 20:44 |
| creiht | hah | 20:45 |
| *** jamiehannaford has quit IRC | 20:53 | |
| *** byeager has quit IRC | 20:54 | |
| portante | clayg: so the state of the probetests with 99315, should I expect them to pass OOTB? | 20:58 |
| portante | on one of my SAIOs, 4 errors and 2 failures | 20:59 |
| portante | it was complaining about my python-swiftclient, so rechecking that I have my install correct | 21:02 |
| *** byeager has joined #openstack-swift | 21:02 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:08 |
| *** byeager has quit IRC | 21:08 | |
| torgomatic | yes, bot, please keep proposing the same damn -2'ed thing over and over again. :| | 21:09 |
| portante | if we did not just -2 it, but left it unanswered, would it then stop talking? | 21:13 |
| torgomatic | I don't think it's set up for people to not just automatically approve changes to their dependencies | 21:18 |
| portante | perhaps we could reprogram it ... ;) | 21:19 |
| *** fifieldt_ has quit IRC | 21:19 | |
| *** judd7 has quit IRC | 21:25 | |
| *** judd7 has joined #openstack-swift | 21:26 | |
| clayg | portante: eah probetests pass ootb on storage policies, run python swift/test/probe/common.py to do a quick config check | 21:27 |
| clayg | portante: except they don't pass with ssync, known but with ssync uncovered with probetests that use an slo manifest and hit replication -> https://bugs.launchpad.net/swift/+bug/1329093 | 21:27 |
| portante | clayg: "object ERROR (silver): /etc/swift/object-1.ring.gz has 2 replicas instead of 3" | 21:31 |
| portante | and the stack trace I am seeing in the object server is: | 21:32 |
| portante | https://gist.github.com/portante/5b63128e54b57d8974da | 21:32 |
| portante | I gotta run, online a bit later | 21:33 |
| *** fifieldt_ has joined #openstack-swift | 21:36 | |
| *** judd7 has quit IRC | 21:39 | |
| *** IRTermite has quit IRC | 21:39 | |
| *** judd7 has joined #openstack-swift | 21:40 | |
| clayg | portante: what "ssync docs"? | 21:41 |
| clayg | portante: I think both of the sp I have configured on my saio have 3 replicas - none of the current probetests really work with 2 replicas - hell lots of swift doesn't work with 2 replicas when theres failures... i guess pauls "setting up policies on a saio" suggest the second policy to be two replicas? | 21:45 |
| peluse_ | clayg: yes and the SAIO script for making the 2nd ring on the PS branch sets up 2 replicas | 21:46 |
| *** judd7 has quit IRC | 21:46 | |
| clayg | portante: that traceback looks like an earlier bug that popped up when gholt had peluse_ add the per-policy tmp and quarantine dirs - like the async update didn't know to use per-policy tmp dirs or something? but I thought that was fixed, so I'll try to look into that. | 21:46 |
| *** jamiehannaford has joined #openstack-swift | 21:46 | |
| clayg | peluse_: can you confirm if that setup works with probetests for you? | 21:47 |
| peluse_ | running now... | 21:47 |
| *** judd7 has joined #openstack-swift | 21:47 | |
| peluse_ | will swift to rsync first though... one min | 21:47 |
| clayg | funny muscle memory | 21:47 |
| peluse_ | heh, just noticed that I typed that :) | 21:48 |
| peluse_ | hmmm... 3 err 2 fail. will rerun with no 2nd policy and again with 2nd policy at 3x instead of 2x and see how it does... | 21:54 |
| clayg | peluse_: thanks | 21:54 |
| *** judd7 has quit IRC | 21:59 | |
| *** kashyapk has joined #openstack-swift | 22:03 | |
| *** judd7 has joined #openstack-swift | 22:03 | |
| *** dmsimard is now known as dmsimard_away | 22:05 | |
| *** judd7 has quit IRC | 22:07 | |
| *** kashyapk has quit IRC | 22:09 | |
| *** jergerber has quit IRC | 22:12 | |
| peluse_ | both at 3x: ......FFEES..... | 22:12 |
| peluse_ | 2nd at 2x or 3x: ......FFEES..E... | 22:12 |
| peluse_ | only pol (0): ......SSSSS..S... | 22:12 |
| peluse_ | clayg: (2nd one is 2x, 'not 2x or 3x') so clearly there is one that fails w/2x but still the other fail w/any 2nd policy | 22:13 |
| peluse_ | clayg: I can debug if you like but have to take of soon to teach people how to beat other people up. Have time tomorrow morn though | 22:14 |
| clayg | peluse_: well they all pass for me (and last I heard from creiht and gholt) for them as well? I've got torgomatic near by - I'll see if he has some failures we can look at. | 22:17 |
| *** elmiko has left #openstack-swift | 22:17 | |
| clayg | peluse_: I can also confirm that they fail for 2x policies (I expect they will) and then we'll have to figure out if it's feasable to have a probetests that will pass on 2x replica swift or if we should update the docs to recommend 3x for the second ring. | 22:17 |
| peluse_ | clayg: OK, if 3x though its docs as well as remakerings that need to be tweaked | 22:18 |
| clayg | peluse_: i'm currently still wading through doc comments, and I'd like to spend more time looking at acoles suggestions on the two vector time and writing more tests there | 22:18 |
| peluse_ | clayg: either way, I'll dig into my failures and see whats up | 22:18 |
| clayg | peluse_: noted | 22:18 |
| clayg | peluse_: np, thanks | 22:18 |
| peluse_ | clayg: let me know if you'd rather have me look at something else so you can focus on the 2d time deal | 22:19 |
| *** cutforth has quit IRC | 22:27 | |
| peluse_ | clayg: my failures were because I was missing /etc/swift/container-reconciler.conf. Pass now | 22:32 |
| peluse_ | well, pass with both rings at 3x. Believe one failure still there if 2nd ring is 2x... double checking now | 22:32 |
| peluse_ | peluse_: Yup, A_OK with both at 3x, One failure when 2nd ring is 2x (otherwise no issues) | 22:37 |
| peluse_ | talking to myself I guess... | 22:37 |
| mattoliverau | Morning all | 22:42 |
| peluse_ | good morning! | 22:42 |
| *** kevinc_ has quit IRC | 22:42 | |
| mattoliverau | peluse_: looks like your having a great conversation with your self.. sounds interesting tho :P | 22:49 |
| *** IRTermite1 has quit IRC | 22:59 | |
| *** kevinc_ has joined #openstack-swift | 23:05 | |
| clayg | peluse_: hehe sorry | 23:16 |
| clayg | peluse_: torgomatic just hit the same thing, and come to think of it creiht did his first round as well - I think maybe folks are setting up based on 96026 and doing the reconciler stuff which is added to the docs in 96038 "Add container-reconciler daemon" | 23:18 |
| clayg | next time i rebase I'll stop at 96038 and pull the doc changes up the chain abit | 23:18 |
| torgomatic | I'm running on patch 99315, which is the last one in the chain (I think) | 23:18 |
| clayg | Also considering moving docs to the end of the rebase change so people can commit over it w/o messing up the rebase line? | 23:18 |
| torgomatic | patchbot: patch 99315 | 23:19 |
| patchbot | torgomatic: https://review.openstack.org/#/c/99315/ | 23:19 |
| torgomatic | clayg: I have no problem with that. Write the code, then the docs, right? | 23:19 |
| *** jamiehannaford has quit IRC | 23:35 | |
| *** kevinc_ has quit IRC | 23:43 | |
| portante | clayg: I am running 99315 | 23:55 |
| portante | what I see in the code from the stack trace is that the async_update() call in delete at update has a hard-coded 0 for the policy indx | 23:56 |
| portante | and my guess is that some how /tmp is not created if only an object from the other storage policy is created | 23:57 |
| portante | but just guessing | 23:57 |
| portante | have to dig in more | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!