| *** shakayumi has quit IRC | 00:31 | |
| *** john3213 has joined #openstack-swift | 00:34 | |
| *** john3213 has left #openstack-swift | 00:40 | |
| *** sungju has joined #openstack-swift | 01:04 | |
| *** saschpe has quit IRC | 01:05 | |
| *** saschpe has joined #openstack-swift | 01:12 | |
| *** nosnos has joined #openstack-swift | 01:41 | |
| *** matsuhashi has joined #openstack-swift | 02:17 | |
| *** kenhui1 has quit IRC | 02:55 | |
| *** kenhui has joined #openstack-swift | 03:05 | |
| *** mrsnivvel has joined #openstack-swift | 03:15 | |
| *** kenhui1 has joined #openstack-swift | 03:18 | |
| *** kenhui1 has quit IRC | 03:19 | |
| *** kenhui has quit IRC | 03:21 | |
| *** kenhui has joined #openstack-swift | 03:21 | |
| *** kenhui has quit IRC | 03:22 | |
| *** praveenkumar has joined #openstack-swift | 03:29 | |
| *** praveenkumar has joined #openstack-swift | 03:30 | |
| *** lnxnut has joined #openstack-swift | 03:30 | |
| *** h6w has quit IRC | 03:31 | |
| *** nosnos has quit IRC | 03:51 | |
| *** Diddi has quit IRC | 03:57 | |
| *** lnxnut has quit IRC | 03:57 | |
| *** Diddi has joined #openstack-swift | 03:59 | |
| *** haomaiwa_ has joined #openstack-swift | 04:09 | |
| *** ppai has joined #openstack-swift | 04:10 | |
| *** psharma has joined #openstack-swift | 04:18 | |
| *** madhuri has joined #openstack-swift | 04:25 | |
| madhuri | Hi, is there any way to meter an account usage in swift? Does Ceilometer provide such support? | 04:25 |
|---|---|---|
| *** wer has quit IRC | 04:26 | |
| *** Manish_ has joined #openstack-swift | 04:27 | |
| *** wer has joined #openstack-swift | 04:27 | |
| *** nosnos has joined #openstack-swift | 04:32 | |
| *** nshaikh has joined #openstack-swift | 05:23 | |
| *** haomai___ has joined #openstack-swift | 05:30 | |
| *** haomaiwa_ has quit IRC | 05:33 | |
| *** Midnightmyth has joined #openstack-swift | 06:13 | |
| *** matsuhashi has quit IRC | 06:25 | |
| *** matsuhashi has joined #openstack-swift | 06:28 | |
| *** sungju has quit IRC | 07:30 | |
| *** foexle has joined #openstack-swift | 07:35 | |
| *** mmcardle has joined #openstack-swift | 07:47 | |
| *** nacim has joined #openstack-swift | 08:01 | |
| *** odyssey4me has joined #openstack-swift | 08:04 | |
| *** Midnightmyth has quit IRC | 08:12 | |
| *** kenhui has joined #openstack-swift | 08:13 | |
| *** kenhui has quit IRC | 08:20 | |
| *** mlipchuk has joined #openstack-swift | 08:32 | |
| *** nshaikh has quit IRC | 08:42 | |
| *** praveenkumar has quit IRC | 08:48 | |
| *** praveenkumar has joined #openstack-swift | 09:06 | |
| *** Manish_ has quit IRC | 09:15 | |
| *** otoolee has joined #openstack-swift | 09:17 | |
| *** Midnightmyth has joined #openstack-swift | 09:29 | |
| *** mlipchuk has quit IRC | 09:47 | |
| *** matsuhashi has quit IRC | 09:50 | |
| fbo_away_away | madhuri, hi maybe have a look at https://github.com/enovance/swift-account-stats | 09:54 |
| madhuri | thanks fbo_away_away | 09:59 |
| madhuri | Isn't Ceilometer provide such support? | 10:00 |
| fbo_away_away | madhuri, you will find what Ceilometer fetch regarding Swift here : http://docs.openstack.org/developer/ceilometer/measurements.html#object-storage-swift | 10:04 |
| madhuri | fbo_away_away, I have read this document, but was looking for API or any interface that can fetch metering values? | 10:08 |
| fbo_away_away | madhuri, AFAIK the ceilometer REST API provide that | 10:10 |
| *** mlipchuk has joined #openstack-swift | 10:10 | |
| *** praveenkumar has quit IRC | 10:11 | |
| madhuri | fbo_away_away:I have one mory query. | 10:12 |
| madhuri | What mechanism is used for swift to collect data using Ceilometer? Is it polling, push or some other mechanism? | 10:13 |
| fbo_away_away | madhuri, according to that http://docs.openstack.org/developer/ceilometer/architecture.html it is polling for Swift at least | 10:15 |
| madhuri | I have looked for any documentation for Ceilometer usage with swift but couldn't find one. Is there any related article or documentation? | 10:18 |
| fbo_away_away | madhuri, don't know but feel free to ask on #openstack-ceilometer | 10:23 |
| madhuri | ok thanks fbo_away_away | 10:23 |
| *** mlipchuk has quit IRC | 10:26 | |
| madhuri | fbo_away_away: Why is the ring data stored gzip format? | 10:26 |
| *** praveenkumar has joined #openstack-swift | 10:27 | |
| madhuri | fbo_away_away: Can't the swift-account-stats be used with tempauth for testing purpose? | 10:28 |
| madhuri | fbo_away_away: ^^^ | 10:31 |
| fbo_away_away | madhuri, I guess it is possible, at least I don't see right now why it could'nt be | 10:33 |
| madhuri | Is this throwing error http://pastebin.com/9zHufYXM | 10:36 |
| *** Midnightmyth has quit IRC | 10:39 | |
| *** jamie_h has joined #openstack-swift | 10:43 | |
| madhuri | fbo_away_away: ^^^ | 10:47 |
| *** mlipchuk has joined #openstack-swift | 10:54 | |
| fbo_away_away | madhuri, oh yes by default it uses keystone auth - then you need to modify a bit to test with auth v1 | 10:58 |
| fbo_away_away | patches are welcome :) | 10:58 |
| madhuri | Thank you fbo_away_away | 10:58 |
| madhuri | Surely i would love to | 10:58 |
| *** Midnightmyth has joined #openstack-swift | 10:58 | |
| *** ppai has quit IRC | 11:01 | |
| *** fbo_away_away is now known as fbo_away_away_aw | 11:08 | |
| madhuri | fbo_away_away_aw: there? | 11:10 |
| *** lnxnut has joined #openstack-swift | 11:13 | |
| *** ppai has joined #openstack-swift | 11:13 | |
| *** nosnos has quit IRC | 11:21 | |
| *** lnxnut has quit IRC | 11:24 | |
| *** nshaikh has joined #openstack-swift | 11:33 | |
| *** mmcardle has quit IRC | 11:46 | |
| *** dmsimard has joined #openstack-swift | 11:48 | |
| *** pconstantine has quit IRC | 11:49 | |
| *** mmcardle has joined #openstack-swift | 11:53 | |
| *** fbo_away_away_aw has quit IRC | 11:55 | |
| *** fbo_away has joined #openstack-swift | 11:56 | |
| *** mlipchuk has quit IRC | 12:18 | |
| *** miqui_ has joined #openstack-swift | 12:21 | |
| *** krtaylor has joined #openstack-swift | 12:22 | |
| *** ppai has quit IRC | 12:22 | |
| *** early has quit IRC | 12:29 | |
| *** early has joined #openstack-swift | 12:32 | |
| *** acoles is now known as acoles_away | 12:35 | |
| *** acoles_away is now known as acoles | 12:35 | |
| *** haomai___ has quit IRC | 12:46 | |
| *** haomaiwang has joined #openstack-swift | 12:46 | |
| *** fbo_away is now known as fbo_away_away | 13:01 | |
| *** mlipchuk has joined #openstack-swift | 13:02 | |
| *** kenhui has joined #openstack-swift | 13:09 | |
| *** mlipchuk has quit IRC | 13:12 | |
| *** kenhui has quit IRC | 13:19 | |
| *** mlipchuk has joined #openstack-swift | 13:29 | |
| *** mkollaro has joined #openstack-swift | 13:40 | |
| mkollaro | s there some documentation about what happens when a swift object server goes down? | 13:43 |
| mkollaro | afaik the replication starts only after some timeout, since otherwise it would cause a I/O spike | 13:43 |
| mkollaro | but I can't find any docs about this | 13:43 |
| glange | when an object server goes down, requests to it start to fail | 13:44 |
| *** psharma has quit IRC | 13:45 | |
| glange | requests will eventually start being handed off to other servers | 13:45 |
| cschwede_ | Manish: Ceilometer has it’s own middleware for Ceilometer that is added to the proxy server pipeline: https://github.com/openstack/ceilometer/blob/master/ceilometer/objectstore/swift_middleware.py | 13:45 |
| glange | but replication will still try to send data to the failed server | 13:45 |
| glange | you have to remove the server from the ring to stop it from being replicated to | 13:46 |
| mkollaro | glange: well, if a disk fails, the replication process will create extra replicas on the handoff nodes, so there is some replica_count amount of them after a while...but I heard that if a whole server fails, this process doesn't start immediately, because it would cause a big I/O spike | 13:48 |
| *** mmcardle has quit IRC | 13:48 | |
| *** mmcardle has joined #openstack-swift | 13:49 | |
| mkollaro | glange: are there any docs about this? | 13:50 |
| glange | mkollaro: I'm not sure it works the way you describe | 13:53 |
| mkollaro | I'm not sure either | 13:53 |
| glange | the object servers don't know another object server is down, they only know their requests to it fail | 13:53 |
| *** bill_az has joined #openstack-swift | 13:53 | |
| *** mlipchuk has quit IRC | 13:54 | |
| mkollaro | I think notmyname described it like this to me | 13:54 |
| notmyname | mkollaro: if a whole object server fails (ie other servers can't connect to the IP address), the existing data won't be pushed to a handoff. if just a drive fails, the existing data will be pushed to a handoff automatically. in both cases, new data will go to a handoff | 13:56 |
| mkollaro | notmyname: awesome | 13:56 |
| mkollaro | notmyname: are there any docs about this? | 13:56 |
| glange | but there is not a time delay for that to happen? | 13:56 |
| glange | if an object server goes down | 13:57 |
| *** annegent^ has joined #openstack-swift | 13:57 | |
| notmyname | glange: like the error limited state? | 13:57 |
| glange | ok, there is not a different time delay? :) | 13:57 |
| notmyname | glange: I think it's the same. so swift will retry communication with the failed node every so often until it's marked as out of the cluster. | 13:58 |
| mkollaro | notmyname: and the replicas won't get pushed to a handoff node even if that server stays down for some long time? | 13:59 |
| miqui_ | notmyname: thanks for your awsome feedback during summit, 1/2+ way through the book, cant wait to start contributing... | 14:00 |
| notmyname | mkollaro: if it's a ConnectionTimeout (ie can't connect) then no. if it could connect but gets a 507 response (ie failed drive), then it will use the handoff. of course, the code would be what confirms this ;-) | 14:00 |
| mkollaro | notmyname: ok, thanks :) | 14:01 |
| notmyname | mkollaro: and where I thought there might be docs on this, there isn't. so the answer to your question is one of the worst in open-source: "the docs are the code". ugh. | 14:01 |
| mkollaro | notmyname: classic :D | 14:01 |
| notmyname | mkollaro: it's mentioned in the 4th paragraph on http://docs.openstack.org/developer/swift/overview_replication.html but it's not explicit | 14:03 |
| mkollaro | notmyname: oh, thanks | 14:04 |
| * notmyname goes to get breakfast | 14:06 | |
| notmyname | I'll be online later | 14:06 |
| *** mattoliverau has quit IRC | 14:09 | |
| *** byeager has joined #openstack-swift | 14:09 | |
| *** mlipchuk has joined #openstack-swift | 14:13 | |
| openstackgerrit | A change was merged to openstack/swift: Fixed several Python3 syntax errors https://review.openstack.org/93923 | 14:14 |
| *** lnxnut has joined #openstack-swift | 14:14 | |
| openstackgerrit | Alex Gaynor proposed a change to openstack/swift: Converted several files to have Python3 friendly syntax https://review.openstack.org/94197 | 14:21 |
| *** mattoliverau has joined #openstack-swift | 14:22 | |
| *** mattoliverau has quit IRC | 14:26 | |
| *** byeager has quit IRC | 14:27 | |
| *** byeager has joined #openstack-swift | 14:31 | |
| *** mattoliverau has joined #openstack-swift | 14:38 | |
| *** tdasilva has joined #openstack-swift | 14:39 | |
| *** scotticus has joined #openstack-swift | 14:43 | |
| openstackgerrit | Constantine Peresypkin proposed a change to openstack/swift: account to account copy implementation https://review.openstack.org/72157 | 14:44 |
| *** lnxnut is now known as hipster | 14:45 | |
| *** Midnightmyth has quit IRC | 14:45 | |
| openstackgerrit | Chuck Thier proposed a change to openstack/swift: Add targeted config loading to swift-init https://review.openstack.org/92933 | 14:46 |
| creiht | added docs to that patch -^ | 14:46 |
| creiht | should be good to go now | 14:46 |
| *** mwstorer has joined #openstack-swift | 14:48 | |
| *** mattoliverau has quit IRC | 14:48 | |
| *** pconstantine has joined #openstack-swift | 14:50 | |
| *** mattoliverau has joined #openstack-swift | 14:58 | |
| *** fbo_away_away is now known as fbo | 14:59 | |
| *** mattoliverau has quit IRC | 15:02 | |
| *** madhuri has quit IRC | 15:06 | |
| *** hipster has quit IRC | 15:08 | |
| *** kenhui has joined #openstack-swift | 15:08 | |
| *** mattoliverau has joined #openstack-swift | 15:10 | |
| *** mlipchuk has quit IRC | 15:14 | |
| *** mattoliverau has quit IRC | 15:17 | |
| *** mlipchuk has joined #openstack-swift | 15:18 | |
| notmyname | good morning world | 15:18 |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Documentation https://review.openstack.org/85824 | 15:21 |
| peluse_ | good morning! | 15:21 |
| *** mlipchuk has quit IRC | 15:22 | |
| peluse_ | notmyname: I believe the docs are ready to merge to feature/EC based on numerous reviews and formatting work done by acoles. clayg can add reconciler doc as a separate patch when ready... | 15:22 |
| notmyname | awesome | 15:23 |
| notmyname | I'm going through this weekend's email and then I'll start with the post-summit todo list | 15:23 |
| notmyname | looks like chmouel has been working on getting the specs repo set up (thanks!) | 15:23 |
| creiht | cool... where is that going to live? | 15:24 |
| notmyname | peluse_: I'm going to cc you on a response to Kota about EC and the hackathon | 15:24 |
| notmyname | creiht: new top-level repo. under the "object storage" program. ie sibling to swift, python-swiftclient, and swift-bench | 15:24 |
| creiht | cool | 15:24 |
| *** PVince81 has joined #openstack-swift | 15:25 | |
| notmyname | and, importantly, same group of reviewers | 15:25 |
| PVince81 | hello | 15:25 |
| PVince81 | I'm struggling to find any example or docs about how to detect changes with SWIFT | 15:25 |
| PVince81 | the container/bucket doesn't seem to have a timestamp/mtime | 15:25 |
| peluse_ | notmyname: cool, will keep an eye out. back in a few hrs... | 15:26 |
| notmyname | PVince81: it's not something built in to swift today. if I were writing it, I'd make a simple middleware that puts every PUT/POST/DELETE request metadata into a queue and then have a reader service indexing those things | 15:26 |
| notmyname | (this is one of the many reasons I'd really like to see progress on marconi...) | 15:27 |
| cschwede_ | notmyname: what about an option to sort container listings by timestamp? something like ?format=json&order=mtime | 15:28 |
| cschwede_ | notmyname: though i like the idea for a marconi middleware | 15:29 |
| *** mlipchuk has joined #openstack-swift | 15:29 | |
| notmyname | cschwede_: that would require new container DB indexes. could be done, but it's not free, and we'd have to consider migration paths (either indexing existing containers [ouch!] or having a new type of container [double ouch!]) | 15:30 |
| *** mattoliverau has joined #openstack-swift | 15:30 | |
| cschwede_ | notmyname: yes, you’re right. i was playing with this on my flight back ;) | 15:30 |
| notmyname | cool | 15:30 |
| *** mkerrin has joined #openstack-swift | 15:30 | |
| notmyname | cschwede_: I'd much rather see a "reverse order" listing feature. we had someone work on that a while back and get about 90% done, but then he stopped working on it and it was dropped | 15:31 |
| *** byeager has quit IRC | 15:31 | |
| cschwede_ | notmyname: well, adding a order=mtime could include something like ?order=name_desc or similar. I mean both options require very similar modifications | 15:33 |
| *** byeager has joined #openstack-swift | 15:33 | |
| *** mlipchuk has quit IRC | 15:33 | |
| notmyname | cschwede_: not really, since we already have an index on the name. essentially it's adding an "ORDER BY DESC" to the generated query. | 15:34 |
| notmyname | cschwede_: the only "hard" part is getting the paging right when paging in reverse order | 15:34 |
| cschwede_ | notmyname: yes, but then we need to update also the way we handle the paging | 15:34 |
| cschwede_ | notmyname: exactly | 15:34 |
| *** mlipchuk has joined #openstack-swift | 15:35 | |
| notmyname | cschwede_: IMO most of that is just making a decision on it and documenting it. eg what do marker and end_marker mean. open or closed ranges? etc | 15:36 |
| cschwede_ | notmyname: i think i’ll wrap my head around this and try to find a nice solution. currently i’m jetlagged, don’t take me too serious ;) | 15:36 |
| notmyname | cschwede_: heh. this is the first morning since ATL that I've felt close to human. :-) | 15:36 |
| PVince81 | notmyname: thanks. I was hoping that object creation/edition would simply change the bucket's timestamp | 15:37 |
| notmyname | cschwede_: is chmouel still around? | 15:37 |
| PVince81 | notmyname: but currently the X-Timestamp seems to be the bucket creation time, not modified time | 15:37 |
| cschwede_ | notmyname: i hope to get into that state asap ;) | 15:37 |
| *** igor_ has joined #openstack-swift | 15:37 | |
| notmyname | PVince81: right. although the listing of objects does have the last-modified timestamp (or is it date-created?), so you may be able to use that. just note that the listings aren't always immediately up to date after the object operation (eventual consistency, etc) | 15:38 |
| cschwede_ | notmyname: not sure if chmouel is around at the moment, but i already saw him online today | 15:39 |
| PVince81 | notmyname: but it means I'd need to request the whole list every time, if sorting doesn't work? | 15:39 |
| notmyname | cschwede_: ya, me too. looks like he was doing the swift-specs thing, but I also though he was going to work on the py3 flake test | 15:39 |
| notmyname | PVince81: yes. see above about swift not handling that use case too well without additional middleware ;-) | 15:40 |
| PVince81 | notmyname: yeah, saw that... thanks for the info | 15:40 |
| cschwede_ | notmyname: i think Alex_Gaynor worked on that? https://review.openstack.org/#/c/93987/ | 15:40 |
| notmyname | ah, nice. thanks | 15:41 |
| cschwede_ | PVince81: does it matter how you’re objects are named? | 15:41 |
| PVince81 | cschwede_: unfortunately yes. People want to be able to upload files to their container without going through our code | 15:42 |
| PVince81 | cschwede_: it's about files... | 15:42 |
| * notmyname is now going though summit etherpads to look for TODO items... | 15:42 | |
| *** mlipchuk has quit IRC | 15:42 | |
| PVince81 | cschwede_: so we need to detect these changes to sync up our local cache | 15:42 |
| notmyname | PVince81: makes total sense. (and I'd echo what your users are saying that uploading directly to swift is the way to go) | 15:45 |
| notmyname | PVince81: this sounds very similar to the feature that some swift deployers have added to do metadata searching | 15:45 |
| notmyname | PVince81: and in that case they have added some middleware to send a message over a queue to the indexer service every time an object is written | 15:46 |
| notmyname | PVince81: briancline may be able to give more info on what softlayer has done for this (although I don't see anything on softlayers github page that would help here) | 15:48 |
| PVince81 | notmyname: I see, it makes sense | 15:48 |
| chmouel | notmyname: yes working on that py3 as well, | 15:48 |
| chmouel | notmyname: we need to add the tox target first (ie: Alex_Gaynor's review) | 15:49 |
| notmyname | PVince81: if you do go down the route of writing some middleware, I think you should be able to find a lot of help in here on how to make sure it works well | 15:49 |
| PVince81 | actually I'm not familiar at all with the whole SWIFT stack, was just trying to fix PHP code that uses the REST API | 15:49 |
| notmyname | chmouel: thanks! | 15:49 |
| PVince81 | so I don't think I'm going to start writing middlewares :-S | 15:50 |
| PVince81 | anyway, thanks for the input | 15:50 |
| chmouel | i would have point to cschwede_ online youtube presentation at the summit about howto write swift middleware but it's not online yet :( | 15:50 |
| cschwede_ | PVince81: any Python developer around your place? | 15:50 |
| notmyname | PVince81: ah, ok. we're (mostly) nice people. and python is (mostly) nice to code in. jump on in! :-) | 15:50 |
| PVince81 | I love Python, yes, and know it as well | 15:50 |
| cschwede_ | chmouel: thanks, wanted to do the same, i’m also wondering why it’s not online yet | 15:50 |
| chmouel | cschwede_: video post processing? | 15:51 |
| notmyname | cschwede_: chmouel: adding the explosion effects? | 15:51 |
| cschwede_ | chmouel: the talks before and after mine in the same room are already online (since a few days) | 15:51 |
| chmouel | or the camera man didn't press the record button? | 15:51 |
| cschwede_ | PVince81: it’s not that hard to write a middleware, have a look here: https://github.com/enovance/swift-middleware-sample - it’s a step by step guide | 15:53 |
| PVince81 | I don't even know how to install the server itself | 15:53 |
| * PVince81 noob with SWIFT | 15:53 | |
| PVince81 | and I was hoping I wouldn't need to have to go that deep :-) | 15:54 |
| notmyname | chmouel: Alex_Gaynor: how do I test the py3pep8 patch? https://review.openstack.org/#/c/93987/5/tox.ini | 15:55 |
| cschwede_ | PVince81: ok - but if you are going to play with it on one day, just ask here if you need help :) | 15:55 |
| chmouel | notmyname: you need py34 (brew install python3 should do it) | 15:55 |
| PVince81 | sounds like a plan. I have other stuff on my plate at the moment | 15:55 |
| PVince81 | but will keep the middleware solution in mind | 15:55 |
| PVince81 | thanks a lot, guys! | 15:55 |
| notmyname | chmouel: just get py3 and then `tox -e py3pep8`? | 15:56 |
| *** Midnightmyth has joined #openstack-swift | 15:56 | |
| chmouel | notmyname: yah | 15:56 |
| *** mattoliverau has quit IRC | 15:57 | |
| *** byeager has quit IRC | 15:59 | |
| *** kenhui has quit IRC | 16:00 | |
| notmyname | chmouel: does it have to be py3.4 or is py3.3 ok? | 16:00 |
| chmouel | notmyname: py33 should be ok but we were thinking python34 should be easier for osx folks to install | 16:00 |
| notmyname | why would people be running tests on os x? (only half kidding there) | 16:01 |
| chmouel | notmyname: flake8 is only syntax check :) | 16:02 |
| notmyname | I know :-) | 16:02 |
| * cschwede_ is waiting for the first publicy announced Swift cluster on top of OSX - http://www.macminivault.com/datacenter/ | 16:02 | |
| *** mlipchuk has joined #openstack-swift | 16:05 | |
| *** mattoliverau has joined #openstack-swift | 16:07 | |
| *** byeager has joined #openstack-swift | 16:13 | |
| *** mmcardle1 has joined #openstack-swift | 16:13 | |
| *** mmcardle has quit IRC | 16:13 | |
| *** byeager has quit IRC | 16:14 | |
| PVince81 | is it possible to quickly install the openstack object store in a VM ? | 16:18 |
| PVince81 | without much fiddling with settings? | 16:18 |
| *** hipster has joined #openstack-swift | 16:18 | |
| PVince81 | most docs seem to include a lot of notes and different services | 16:18 |
| PVince81 | which probably takes time to setup | 16:18 |
| notmyname | PVince81: all the default settings work well (so no fiddling). although designed for a dev environment, https://github.com/swiftstack/vagrant-swift-all-in-one is the fastest way to get started | 16:18 |
| PVince81 | notmyname: thanks! | 16:19 |
| PVince81 | five steps, nice! ;-) | 16:19 |
| notmyname | PVince81: I just give links. clayg and alpha_ori actually wrote it :-) | 16:19 |
| PVince81 | I tried installing ceph in the past... got lost in the process | 16:19 |
| *** byeager has joined #openstack-swift | 16:26 | |
| *** nshaikh has quit IRC | 16:27 | |
| *** foexle has quit IRC | 16:45 | |
| *** igor_ has quit IRC | 16:45 | |
| *** igor_ has joined #openstack-swift | 16:45 | |
| *** igor_ has quit IRC | 16:50 | |
| PVince81 | notmyname: what region should I use when connecting to that all-in-one ? | 16:58 |
| *** gyee has joined #openstack-swift | 17:01 | |
| *** haomaiwang has quit IRC | 17:02 | |
| *** mmcardle1 has quit IRC | 17:02 | |
| *** Tyger has joined #openstack-swift | 17:04 | |
| *** PVince81 has quit IRC | 17:05 | |
| notmyname | chmouel: Alex_Gaynor: wow. just ran the py3pep8 job on my saio. there's a _lot_ of errors | 17:09 |
| Alex_Gaynor | notmyname: yup! It's cool though, they're mostly very easy to fix :-) | 17:09 |
| *** acoles is now known as acoles_away | 17:11 | |
| notmyname | Alex_Gaynor: chmouel: FWIW I ran it under py32 because that's what I get on precise with `apt-get install python3` | 17:15 |
| Alex_Gaynor | that should be fine | 17:15 |
| Alex_Gaynor | There's not much (any?) variability in what it shows between different py3s | 17:15 |
| notmyname | chmouel: I added a +2. if it's ok with you, can you go ahead and merge it? | 17:15 |
| Alex_Gaynor | notmyname: since you're around, can you leave a comment on the config change so that swiftclient cna start gating on py3? | 17:16 |
| *** igor_ has joined #openstack-swift | 17:16 | |
| Alex_Gaynor | ()https://review.openstack.org/#/c/92988/ | 17:16 |
| notmyname | Alex_Gaynor: ah, ok. looking | 17:17 |
| Alex_Gaynor | thanks | 17:17 |
| *** mattoliverau has quit IRC | 17:21 | |
| *** igor_ has quit IRC | 17:21 | |
| notmyname | Alex_Gaynor: hmm..fails with py32. gotta go get py33 now | 17:22 |
| Alex_Gaynor | notmyname: For swiftclient itself? Yeah that's py33 | 17:23 |
| Alex_Gaynor | (also works on 34 is that's easier) | 17:23 |
| notmyname | actually, it seems to fail in simplejson (?!) | 17:23 |
| notmyname | ESCAPE = re.compile(u'[\\x00-\\x1f\\\\"\\b\\f\\n\\r\\t\u2028\u2029]') | 17:24 |
| notmyname | Alex_Gaynor: why is that invalide py32 syntax? | 17:24 |
| Alex_Gaynor | the u'' prefix | 17:24 |
| notmyname | ah | 17:24 |
| Alex_Gaynor | it was removed from 3.0-3.2 and brought back in 3.3 | 17:24 |
| notmyname | gotcha | 17:24 |
| notmyname | because "everything's unicode" right? | 17:24 |
| Alex_Gaynor | right | 17:24 |
| notmyname | and then they realized the world didn't start with pep3000, therefore add the u prefix back. :-) | 17:25 |
| Alex_Gaynor | pretty much | 17:27 |
| *** mattoliverau has joined #openstack-swift | 17:34 | |
| notmyname | yay deadsnakes ppa! | 17:36 |
| *** mattoliverau has quit IRC | 17:38 | |
| *** zaitcev has joined #openstack-swift | 17:43 | |
| *** ChanServ sets mode: +v zaitcev | 17:43 | |
| *** mlipchuk has quit IRC | 17:45 | |
| notmyname | Alex_Gaynor: https://gist.github.com/notmyname/8ce23e527d7a44352590 <-- what I get on swiftclient | 17:49 |
| *** mattoliverau has joined #openstack-swift | 17:50 | |
| Alex_Gaynor | notmyname: Any chance you have an old .tox or something? Try blowing it away and seeing it that fixes it | 17:50 |
| notmyname | Alex_Gaynor: rm -rf .tox ? | 17:51 |
| Alex_Gaynor | yeah | 17:51 |
| openstackgerrit | Christian Berendt proposed a change to openstack/python-swiftclient: replace string format arguments with function parameters https://review.openstack.org/94241 | 17:52 |
| *** shri has joined #openstack-swift | 17:54 | |
| notmyname | Alex_Gaynor: same | 17:54 |
| Alex_Gaynor | notmyname: Not sure what to make of it, it works fine for me locally, and I think for chmouel and also for the CI | 17:55 |
| notmyname | Alex_Gaynor: what about the .testrepository/ dir? what's that for? | 17:55 |
| Alex_Gaynor | notmyname: testr uses it for... stuff, probably you can blow it away too | 17:55 |
| notmyname | Alex_Gaynor: any options to give me more insight on what's actually failing? | 17:55 |
| Alex_Gaynor | I don't really know anything abotu testr, sorry | 17:55 |
| notmyname | oh, that worked | 17:56 |
| notmyname | killing .testrepository dir | 17:56 |
| Alex_Gaynor | files you can randomly delete are cool | 17:57 |
| creiht | lol | 17:57 |
| notmyname | heh | 17:57 |
| peluse_ | cores: fyi fairly straigtforward bug fix (well, filled as a bug but not really) aging @ 4 weeks... availeble for review at https://review.openstack.org/#/c/88633/ | 17:57 |
| notmyname | ok, added my +1 to making python-swiftclient py3 checks voting | 17:57 |
| *** mmcardle has joined #openstack-swift | 18:02 | |
| *** tdasilva has quit IRC | 18:04 | |
| notmyname | peluse_: one minor nit there, and a more general question. I'd like comments from the public service providers before merging it, though (briancline, acoles_away, dfg) | 18:05 |
| notmyname | oh, and cschwede_ ^ | 18:05 |
| creiht | notmyname: yeah I left a comment | 18:06 |
| notmyname | thanks | 18:06 |
| creiht | I'm not fond of having the blacklist/whitelist public | 18:06 |
| *** mkollaro has quit IRC | 18:07 | |
| *** mmcardle has quit IRC | 18:07 | |
| *** Tyger has quit IRC | 18:07 | |
| dfg | added a comment too- just added couple more keys to what creiht said | 18:10 |
| *** mkollaro has joined #openstack-swift | 18:11 | |
| *** Midnightmyth has quit IRC | 18:13 | |
| *** hipster has quit IRC | 18:15 | |
| *** igor_ has joined #openstack-swift | 18:16 | |
| *** igor_ has quit IRC | 18:21 | |
| notmyname | zaitcev: last week you were looking for Kota's slides on the replication improvements he was proposing. looks like he sent a link to the -dev mailing list today | 18:21 |
| zaitcev | notmyname: thanks | 18:22 |
| *** mmcardle has joined #openstack-swift | 18:22 | |
| notmyname | zaitcev: http://www.slideshare.net/tsuyuzaki/open-stacks-ummitjuno | 18:22 |
| zaitcev | notmyname: I see someone added them to Etherpad | 18:24 |
| notmyname | ah ok | 18:24 |
| notmyname | creiht: wait, wut? we're finally merging profiling middleware? cool! | 18:37 |
| *** kenhui has joined #openstack-swift | 18:39 | |
| *** Midnightmyth has joined #openstack-swift | 18:46 | |
| *** hipster has joined #openstack-swift | 18:47 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Ratelimit parameters to /info https://review.openstack.org/88633 | 18:47 |
| peluse_ | ^ comments addressed. Thanks guys! | 18:48 |
| *** hipster has quit IRC | 18:51 | |
| *** mattoliverau has quit IRC | 19:04 | |
| *** lpabon has joined #openstack-swift | 19:07 | |
| cschwede_ | peluse_: hmm, not sure about this. while i think that most operators are fine with it, others might need to tweak it. and yet others might be ok with more info exposed… | 19:11 |
| cschwede_ | peluse_: … what do you think about making the returned info configurable? | 19:11 |
| cschwede_ | creiht dfg notmyname ^^ | 19:12 |
| *** mattoliverau has joined #openstack-swift | 19:12 | |
| notmyname | cschwede_: no | 19:13 |
| notmyname | ;-) | 19:13 |
| cschwede_ | no? | 19:13 |
| notmyname | cschwede_: less config options is better | 19:13 |
| notmyname | cschwede_: I'd rather see a subset that is on (as currently proposed, I think). and then maybe the full set as an admin response | 19:14 |
| notmyname | maybe | 19:14 |
| cschwede_ | notmyname: agreed, less is better. and +1 for exposing the other settings as admin response. | 19:16 |
| *** igor_ has joined #openstack-swift | 19:17 | |
| *** hipster has joined #openstack-swift | 19:18 | |
| *** byeager has quit IRC | 19:19 | |
| *** ophuk has joined #openstack-swift | 19:19 | |
| *** kenhui has quit IRC | 19:20 | |
| *** igor_ has quit IRC | 19:22 | |
| *** hipster has quit IRC | 19:23 | |
| *** mmcardle has quit IRC | 19:40 | |
| *** hipster has joined #openstack-swift | 19:47 | |
| creiht | notmyname: :) | 19:51 |
| *** hipster has quit IRC | 19:52 | |
| *** patchbot has quit IRC | 19:52 | |
| *** patchbot` has joined #openstack-swift | 19:52 | |
| *** patchbot` is now known as patchbot | 19:52 | |
| *** sandywalsh_ has joined #openstack-swift | 19:53 | |
| creiht | the only other input that I have on the ratelimit stuff is that it is kinda raw | 19:53 |
| *** mwstorer has quit IRC | 19:54 | |
| *** zul has quit IRC | 19:54 | |
| *** minnear_ has quit IRC | 19:54 | |
| *** sandywalsh has quit IRC | 19:54 | |
| *** pberis has quit IRC | 19:54 | |
| creiht | should we make it something that is more easily consumed? | 19:54 |
| creiht | I'm really fine either way | 19:54 |
| creiht | just curious what others thought | 19:54 |
| *** j_king has quit IRC | 19:54 | |
| *** mlanner has quit IRC | 19:55 | |
| *** scotticus has quit IRC | 19:55 | |
| *** gpocentek has quit IRC | 19:55 | |
| *** swills has quit IRC | 19:55 | |
| *** j_king has joined #openstack-swift | 19:55 | |
| ophuk | Any idea as to why I would get a insufficient space error when running swift stat? | 19:56 |
| *** gpocentek has joined #openstack-swift | 19:56 | |
| *** mlanner has joined #openstack-swift | 19:56 | |
| *** scotticus has joined #openstack-swift | 19:56 | |
| *** zackmdavis has quit IRC | 19:57 | |
| *** pandemicsyn has quit IRC | 19:57 | |
| *** zul has joined #openstack-swift | 19:57 | |
| *** physcx has quit IRC | 19:58 | |
| *** swills has joined #openstack-swift | 20:00 | |
| *** swills has quit IRC | 20:00 | |
| *** swills has joined #openstack-swift | 20:00 | |
| *** minnear has joined #openstack-swift | 20:01 | |
| *** pandemicsyn has joined #openstack-swift | 20:01 | |
| *** zackmdavis has joined #openstack-swift | 20:03 | |
| *** foexle has joined #openstack-swift | 20:03 | |
| *** lpabon has quit IRC | 20:05 | |
| *** byeager has joined #openstack-swift | 20:06 | |
| *** erlon has joined #openstack-swift | 20:08 | |
| *** byeager has quit IRC | 20:09 | |
| *** mattoliverau has quit IRC | 20:10 | |
| *** mmcardle has joined #openstack-swift | 20:10 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Add keystoneV2.0 compatability to swift-dispersion-populate/report https://review.openstack.org/93611 | 20:11 |
| *** physcx has joined #openstack-swift | 20:13 | |
| *** elambert has joined #openstack-swift | 20:14 | |
| *** mmcardle has quit IRC | 20:15 | |
| *** pberis has joined #openstack-swift | 20:18 | |
| *** igor_ has joined #openstack-swift | 20:18 | |
| *** kenhui has joined #openstack-swift | 20:22 | |
| *** mattoliverau has joined #openstack-swift | 20:22 | |
| *** igor_ has quit IRC | 20:23 | |
| *** foexle has quit IRC | 20:29 | |
| *** elambert has quit IRC | 20:32 | |
| *** elambert has joined #openstack-swift | 20:34 | |
| *** kenhui has quit IRC | 20:34 | |
| *** mmcardle has joined #openstack-swift | 20:36 | |
| *** mmcardle1 has joined #openstack-swift | 20:41 | |
| *** mmcardle has quit IRC | 20:41 | |
| *** otoolee has quit IRC | 20:42 | |
| *** mmcardle1 has quit IRC | 20:45 | |
| *** hipster has joined #openstack-swift | 20:47 | |
| *** hipster has quit IRC | 20:52 | |
| *** mkollaro has quit IRC | 21:08 | |
| *** igor_ has joined #openstack-swift | 21:19 | |
| notmyname | meeting agenda updated for wednesday's meeting. should be a full hour, i think | 21:20 |
| notmyname | https://wiki.openstack.org/wiki/Meetings/Swift | 21:20 |
| openstackgerrit | A change was merged to openstack/swift: Added a tox job for running flake8 under Python 3 https://review.openstack.org/93987 | 21:22 |
| *** igor_ has quit IRC | 21:23 | |
| notmyname | and like that we've got py3 syntax checking | 21:28 |
| portante | nice! | 21:30 |
| portante | looking at the python27 check failure for swift profile middleware ... | 21:32 |
| *** arahal has joined #openstack-swift | 21:33 | |
| portante | looks like a possible race in test_sleeper ... | 21:35 |
| *** mmcardle has joined #openstack-swift | 21:41 | |
| *** mmcardle has quit IRC | 21:46 | |
| *** hipster has joined #openstack-swift | 21:47 | |
| *** hipster has quit IRC | 21:52 | |
| notmyname | portante: I think my favorite test pattern is mocking out time. I know it's not functionally different than mocking out any other callable, but just saying that "for this test I'm going to slow time down and go backwards" is just mind-bending :-) | 21:53 |
| *** kenhui has joined #openstack-swift | 21:54 | |
| portante | ;) | 21:56 |
| portante | filed a bug for this, if nobody else picks this up, will probably fix this test tomorrow. | 21:56 |
| notmyname | thanks | 21:56 |
| portante | welcome | 21:57 |
| notmyname | looks like another bug is about to be filed too (from #openstack). swift doesn't like mount points that have an underscore | 21:57 |
| portante | I'll just post -1s on all reviews until this is fixed | 21:57 |
| notmyname | lol | 21:57 |
| notmyname | passive aggressive much? ;-) | 21:57 |
| portante | really, underscores? | 21:57 |
| portante | who is passive? | 21:57 |
| portante | action jackson is the name | 21:58 |
| notmyname | ophuk: oh hai! you're in here too! | 21:58 |
| *** igor_ has joined #openstack-swift | 21:58 | |
| notmyname | ya, I think underscores in mount points is a pretty low priority thing, but still, sounds like a parse error on our part | 21:58 |
| notmyname | and ophuk was hit by it through no fault of his own | 21:58 |
| notmyname | well, except for the using underscores part | 21:59 |
| ophuk | notmyname: yup:) This channel is usually more helpful than the other one if people are in here | 22:01 |
| *** mattoliverau has quit IRC | 22:02 | |
| notmyname | ophuk: I'm in both (as are a few others). I don't want #openstack to be a desert, but this one is more active (and actively monitored) for swift-related things | 22:02 |
| *** igor_ has quit IRC | 22:03 | |
| ophuk | notmyname: yeah that's what I figured. Usually I tend to ask my question initially in the other one since this one is supposed to be development but when nobody answers or if nobody has been answering like last friday I just post in both:) | 22:03 |
| *** Midnightmyth has quit IRC | 22:03 | |
| *** elambert has quit IRC | 22:03 | |
| ophuk | notmyname: bug report filed | 22:08 |
| *** foexle has joined #openstack-swift | 22:10 | |
| *** mattoliverau has joined #openstack-swift | 22:18 | |
| notmyname | ophuk: thanks for posting it. ya, last week many of us were at the openstack summit in atlanta (or recovering from it) | 22:27 |
| peluse_ | notmyname: do you think I should go aheand and ammend the ratelimit /info patch to include those other items I just yanked for admin or just leave it w/those things removed (as it is now)? | 22:27 |
| notmyname | peluse_: I wouldn't worry about the admin /info for now | 22:31 |
| notmyname | peluse_: keeps this patch smaller, and I haven't heard a clamoring for it | 22:31 |
| *** arahal has quit IRC | 22:34 | |
| notmyname | sdague: because all reviewers, as currently listed, are seen as equal. the problem that needs to be solved is giving new and non-core contributors a contact point for question and someone who is shepherding the patch through the process of submission->master | 22:39 |
| *** elambert has joined #openstack-swift | 22:39 | |
| notmyname | wrong channel | 22:42 |
| *** hipster has joined #openstack-swift | 22:42 | |
| *** mmcardle has joined #openstack-swift | 22:43 | |
| *** hipster_ has joined #openstack-swift | 22:44 | |
| *** ophuk has quit IRC | 22:46 | |
| *** hipster has quit IRC | 22:47 | |
| *** hipster has joined #openstack-swift | 22:47 | |
| *** mmcardle has quit IRC | 22:48 | |
| *** hipster_ has quit IRC | 22:49 | |
| *** r-daneel has joined #openstack-swift | 22:49 | |
| peluse_ | notmyname: gracias... and I was wondering about that last comment :) | 22:50 |
| notmyname | peluse_: and I was wondering why he wasn't responding ;-) | 22:51 |
| *** hipster has quit IRC | 22:52 | |
| *** jamie_h has quit IRC | 22:53 | |
| *** sungju has joined #openstack-swift | 22:53 | |
| *** igor_ has joined #openstack-swift | 22:59 | |
| *** r-daneel has quit IRC | 23:03 | |
| *** igor_ has quit IRC | 23:03 | |
| *** r-daneel has joined #openstack-swift | 23:04 | |
| *** r-daneel has quit IRC | 23:07 | |
| *** sungju has quit IRC | 23:07 | |
| *** sungju has joined #openstack-swift | 23:07 | |
| *** matsuhashi has joined #openstack-swift | 23:27 | |
| openstackgerrit | A change was merged to openstack/swift: Add profiling middleware in Swift https://review.openstack.org/53270 | 23:30 |
| *** kenhui has quit IRC | 23:33 | |
| *** kenhui has joined #openstack-swift | 23:34 | |
| *** kenhui1 has joined #openstack-swift | 23:35 | |
| *** kenhui has quit IRC | 23:39 | |
| *** mmcardle has joined #openstack-swift | 23:44 | |
| *** kenhui has joined #openstack-swift | 23:46 | |
| *** elambert has quit IRC | 23:46 | |
| *** hipster has joined #openstack-swift | 23:47 | |
| *** kenhui1 has quit IRC | 23:48 | |
| *** occupant has quit IRC | 23:52 | |
| *** mmcardle has quit IRC | 23:52 | |
| *** hipster has quit IRC | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!