| *** _bluev has quit IRC | 00:14 | |
| *** RockKuo has joined #openstack-swift | 00:17 | |
| peluse | bobf: let me (or anyone here) know if you need some help - great to see someone trying to kick the tires | 00:23 |
|---|---|---|
| *** matsuhashi has joined #openstack-swift | 00:26 | |
| *** shri has quit IRC | 00:40 | |
| *** mlanner has quit IRC | 00:50 | |
| *** piousbox has quit IRC | 00:51 | |
| *** mlanner has joined #openstack-swift | 00:51 | |
| zackmdavis | ls | 00:51 |
| *** dmorita has joined #openstack-swift | 00:59 | |
| *** saschpe has quit IRC | 01:04 | |
| *** saschpe has joined #openstack-swift | 01:06 | |
| *** bobf has quit IRC | 01:13 | |
| *** corrigac has joined #openstack-swift | 01:18 | |
| *** SpamapS_ has joined #openstack-swift | 01:21 | |
| *** taras___ has joined #openstack-swift | 01:21 | |
| *** StevenK_ has joined #openstack-swift | 01:21 | |
| *** alpha_ori has quit IRC | 01:22 | |
| *** SpamapS has quit IRC | 01:22 | |
| *** taras_ has quit IRC | 01:22 | |
| *** cschwede has quit IRC | 01:22 | |
| *** alpha_ori_ has joined #openstack-swift | 01:22 | |
| *** StevenK has quit IRC | 01:22 | |
| *** alpha_ori_ is now known as alpha_ori | 01:23 | |
| *** mordred has quit IRC | 01:23 | |
| *** bsdkurt has quit IRC | 01:23 | |
| *** akscram has quit IRC | 01:23 | |
| *** ccorrigan has quit IRC | 01:23 | |
| *** wkelly has quit IRC | 01:23 | |
| *** amandap has quit IRC | 01:23 | |
| *** sileht has quit IRC | 01:24 | |
| *** amandap_ has joined #openstack-swift | 01:24 | |
| *** wklely has joined #openstack-swift | 01:24 | |
| *** mordred has joined #openstack-swift | 01:24 | |
| *** StevenK_ is now known as StevenK | 01:24 | |
| *** cschwede has joined #openstack-swift | 01:24 | |
| *** bsdkurt has joined #openstack-swift | 01:24 | |
| *** sileht has joined #openstack-swift | 01:24 | |
| *** akscram has joined #openstack-swift | 01:25 | |
| *** _bluev has joined #openstack-swift | 01:25 | |
| *** _bluev has quit IRC | 01:29 | |
| *** gadb has joined #openstack-swift | 01:30 | |
| *** csd has quit IRC | 01:39 | |
| *** sudorandom has quit IRC | 01:46 | |
| *** nosnos has joined #openstack-swift | 01:48 | |
| *** sudorandom has joined #openstack-swift | 01:48 | |
| *** bill_az_ has quit IRC | 01:54 | |
| *** saschpe has quit IRC | 02:00 | |
| *** saschpe has joined #openstack-swift | 02:03 | |
| *** changbl has joined #openstack-swift | 02:07 | |
| *** Gue______ has joined #openstack-swift | 02:11 | |
| *** lollipop has joined #openstack-swift | 02:11 | |
| lollipop | I have created an image from glance with swift successfuly, how I can list the image file from swift client? | 02:13 |
| lollipop | "swift list" can list other files but the image file | 02:13 |
| *** byeager has joined #openstack-swift | 02:21 | |
| openstackgerrit | Yuan Zhou proposed a change to openstack/swift: Update swift-container-info to be storage policy aware https://review.openstack.org/86304 | 02:39 |
| *** byeager has quit IRC | 02:49 | |
| *** zaitcev has quit IRC | 03:00 | |
| *** Gue______ has quit IRC | 03:04 | |
| *** praveenkumar has joined #openstack-swift | 03:07 | |
| *** praveenkumar has quit IRC | 03:10 | |
| *** praveenkumar has joined #openstack-swift | 03:11 | |
| *** zackf has joined #openstack-swift | 03:17 | |
| *** zaitcev has joined #openstack-swift | 03:19 | |
| *** ChanServ sets mode: +v zaitcev | 03:19 | |
| *** zackf has quit IRC | 03:20 | |
| *** matsuhashi has quit IRC | 03:22 | |
| *** zackf has joined #openstack-swift | 03:24 | |
| *** zackf has quit IRC | 03:30 | |
| *** matsuhashi has joined #openstack-swift | 03:32 | |
| *** mmcardle has joined #openstack-swift | 03:39 | |
| *** haomaiwang has joined #openstack-swift | 03:46 | |
| *** mmcardle has quit IRC | 03:49 | |
| *** haomaiw__ has quit IRC | 03:49 | |
| *** wklely is now known as wkelly | 03:59 | |
| *** nosnos has quit IRC | 04:09 | |
| *** ppai has joined #openstack-swift | 04:11 | |
| *** chandan_kumar_ has joined #openstack-swift | 04:14 | |
| *** cheri has joined #openstack-swift | 04:21 | |
| *** chandan_kumar_ has quit IRC | 04:21 | |
| *** zhiyan_ is now known as zhiyan | 04:22 | |
| *** cheri has quit IRC | 04:31 | |
| *** ppai has quit IRC | 04:34 | |
| *** chandan_kumar_ has joined #openstack-swift | 04:35 | |
| *** cheri has joined #openstack-swift | 04:43 | |
| *** gyee has quit IRC | 04:47 | |
| *** nosnos has joined #openstack-swift | 04:47 | |
| *** ppai has joined #openstack-swift | 04:51 | |
| *** bill_az has joined #openstack-swift | 04:52 | |
| *** mmcardle has joined #openstack-swift | 04:53 | |
| *** mmcardle has quit IRC | 05:03 | |
| *** ppai has quit IRC | 05:21 | |
| *** Longgeek_ has joined #openstack-swift | 05:23 | |
| *** Longgeek_ has quit IRC | 05:24 | |
| *** haomaiwang has quit IRC | 05:28 | |
| *** chandan_kumar_ has quit IRC | 05:31 | |
| *** bill_az has quit IRC | 05:33 | |
| *** cheri has quit IRC | 05:40 | |
| *** haomaiwa_ has joined #openstack-swift | 05:50 | |
| *** cheri has joined #openstack-swift | 05:53 | |
| *** ppai has joined #openstack-swift | 05:54 | |
| *** zaitcev has quit IRC | 06:10 | |
| *** Longgeek_ has joined #openstack-swift | 06:15 | |
| *** Longgeek_ has quit IRC | 06:15 | |
| *** SpamapS_ is now known as SpamapS | 06:20 | |
| *** SpamapS has joined #openstack-swift | 06:20 | |
| *** psharma has joined #openstack-swift | 06:23 | |
| openstackgerrit | A change was merged to openstack/swift: Update swift-container-info to be storage policy aware https://review.openstack.org/86304 | 06:25 |
| *** scohen has joined #openstack-swift | 06:36 | |
| *** chandan_kumar has joined #openstack-swift | 06:38 | |
| *** scohen has quit IRC | 06:38 | |
| *** scohen has joined #openstack-swift | 06:40 | |
| *** nshaikh has joined #openstack-swift | 07:14 | |
| *** mmcardle has joined #openstack-swift | 07:19 | |
| *** zigo has quit IRC | 07:29 | |
| hugokuo | Anyone ever user cname_lookup middleware ? Where is the proper position to insert the middleware in the pipline ? | 07:30 |
| lollipop | can we run "swift list" with local auth, without keystone? | 07:46 |
| lollipop | I mean to check the swift status passing by keystone auth | 07:47 |
| *** _bluev has joined #openstack-swift | 07:49 | |
| *** _bluev has left #openstack-swift | 07:50 | |
| *** ashish_ has joined #openstack-swift | 07:50 | |
| ashish_ | -+ | 07:50 |
| *** jamie_h has joined #openstack-swift | 07:56 | |
| ashish_ | I have a swift all in one installation on a single node.I am using a loopback device to create 4 nodes,I am running short of disk space.Could anyone tell me how to increse the storage capacity of my nodes. | 08:11 |
| *** ashish_ has quit IRC | 08:11 | |
| *** ashish_ has joined #openstack-swift | 08:12 | |
| ashish_ | I have a swift all in one installation on a single node.I am using a loopback device to create 4 nodes,I am running short of disk space.Could anyone tell me how to increse the storage capacity of my nodes. | 08:12 |
| ashish_ | Plese help. | 08:12 |
| *** tanee-away is now known as tanee | 08:13 | |
| *** tanee is now known as tanee-away | 08:14 | |
| *** tanee-away is now known as tanee | 08:14 | |
| *** tanee has quit IRC | 08:14 | |
| *** tanee-away has joined #openstack-swift | 08:15 | |
| *** tanee-away is now known as tanee | 08:16 | |
| mmcardle | does any one know what change the following refers to - "Prevent static large objects from containing other static large objects" in https://github.com/openstack/swift/blob/master/CHANGELOG#L432 | 08:17 |
| *** haomaiwa_ has quit IRC | 08:18 | |
| *** haomaiwang has joined #openstack-swift | 08:19 | |
| *** matsuhashi has quit IRC | 08:21 | |
| *** haomaiw__ has joined #openstack-swift | 08:21 | |
| *** matsuhashi has joined #openstack-swift | 08:21 | |
| *** zigo has joined #openstack-swift | 08:23 | |
| *** haomaiwang has quit IRC | 08:24 | |
| *** matsuhashi has quit IRC | 08:25 | |
| *** cheri has quit IRC | 08:25 | |
| *** cheri has joined #openstack-swift | 08:26 | |
| *** d89 has joined #openstack-swift | 08:35 | |
| *** matsuhashi has joined #openstack-swift | 08:37 | |
| *** mmcardle has quit IRC | 08:38 | |
| *** mmcardle has joined #openstack-swift | 08:41 | |
| *** lollipop has quit IRC | 08:47 | |
| *** mmcardle has quit IRC | 09:05 | |
| *** mmcardle has joined #openstack-swift | 09:06 | |
| *** ashish_ has quit IRC | 09:06 | |
| *** gadb has quit IRC | 09:20 | |
| *** matsuhashi has quit IRC | 09:30 | |
| *** foexle has joined #openstack-swift | 09:31 | |
| *** matsuhashi has joined #openstack-swift | 09:34 | |
| *** zhiyan is now known as zhiyan_ | 09:39 | |
| openstackgerrit | Christian Schwede proposed a change to openstack/swift: RADOS (CEPH) DiskFile Backend https://review.openstack.org/60215 | 09:42 |
| *** mkollaro has joined #openstack-swift | 09:51 | |
| *** Longgeek_ has joined #openstack-swift | 09:58 | |
| *** mmcardle has quit IRC | 09:59 | |
| *** mmcardle has joined #openstack-swift | 10:02 | |
| *** Longgeek_ has quit IRC | 10:02 | |
| openstackgerrit | Alistair Coles proposed a change to openstack/swift: Restrict keystone cross-tenant ACLs to IDs https://review.openstack.org/86430 | 10:11 |
| *** matsuhashi has quit IRC | 10:20 | |
| *** Longgeek_ has joined #openstack-swift | 10:22 | |
| *** mmcardle has quit IRC | 10:24 | |
| *** matsuhashi has joined #openstack-swift | 10:25 | |
| *** d89 has quit IRC | 10:27 | |
| *** mmcardle has joined #openstack-swift | 10:28 | |
| *** Longgeek_ has quit IRC | 10:35 | |
| *** d89 has joined #openstack-swift | 10:36 | |
| *** zhiyan_ is now known as zhiyan | 10:41 | |
| *** mmcardle has quit IRC | 10:41 | |
| *** mmcardle has joined #openstack-swift | 10:54 | |
| *** nosnos has quit IRC | 10:59 | |
| *** lpabon has joined #openstack-swift | 10:59 | |
| *** nosnos has joined #openstack-swift | 10:59 | |
| *** lpabon has quit IRC | 10:59 | |
| *** cheri has quit IRC | 11:01 | |
| *** mmcardle has quit IRC | 11:02 | |
| *** nosnos has quit IRC | 11:03 | |
| *** mmcardle has joined #openstack-swift | 11:04 | |
| *** ppai has quit IRC | 11:04 | |
| *** Trixboxer has joined #openstack-swift | 11:06 | |
| *** matsuhashi has quit IRC | 11:14 | |
| *** acoles is now known as acoles_away | 11:15 | |
| *** RockKuo has quit IRC | 11:15 | |
| *** mmcardle has quit IRC | 11:17 | |
| *** cheri has joined #openstack-swift | 11:18 | |
| *** mmcardle has joined #openstack-swift | 11:19 | |
| *** ppai has joined #openstack-swift | 11:21 | |
| *** matsuhashi has joined #openstack-swift | 11:22 | |
| *** zanc has joined #openstack-swift | 11:24 | |
| *** mkerrin has quit IRC | 11:25 | |
| *** mmcardle has quit IRC | 11:29 | |
| *** matsuhashi has quit IRC | 11:31 | |
| *** mkerrin has joined #openstack-swift | 11:32 | |
| *** matsuhas_ has joined #openstack-swift | 11:33 | |
| *** dmorita has quit IRC | 11:35 | |
| *** bill_az has joined #openstack-swift | 11:40 | |
| *** peterbus_ has quit IRC | 11:53 | |
| *** praveenkumar has quit IRC | 11:54 | |
| *** peterbusque has joined #openstack-swift | 11:54 | |
| *** matsuhas_ has quit IRC | 11:57 | |
| *** mkollaro has quit IRC | 12:01 | |
| *** mkollaro1 has joined #openstack-swift | 12:01 | |
| *** nshaikh has quit IRC | 12:03 | |
| *** JuanManuelOlle has joined #openstack-swift | 12:07 | |
| *** mmcardle has joined #openstack-swift | 12:09 | |
| *** cheri has quit IRC | 12:10 | |
| *** jamie_h has quit IRC | 12:13 | |
| *** mmcardle has quit IRC | 12:20 | |
| *** praveenkumar has joined #openstack-swift | 12:30 | |
| *** ppai has quit IRC | 12:33 | |
| *** mkollaro has joined #openstack-swift | 12:36 | |
| *** jamie_h has joined #openstack-swift | 12:37 | |
| *** mkollaro1 has quit IRC | 12:39 | |
| *** joeljwright has joined #openstack-swift | 12:47 | |
| joeljwright | In order to resolve the check-grenade-dsvm test fail with the patch here https://review.openstack.org/#/c/85453/ it seems I need to propose a change to openstack/requirements to be included in the stable/havana branch, but it's not clear to me how I should go about this | 12:53 |
| joeljwright | can someone help me out with the procedure to propose a change to a stable branch... | 12:54 |
| *** nacim has joined #openstack-swift | 12:56 | |
| *** judd7 has joined #openstack-swift | 12:57 | |
| openstackgerrit | Christian Schwede proposed a change to openstack/swift: RADOS (CEPH) DiskFile Backend https://review.openstack.org/60215 | 12:59 |
| *** tdasilva has joined #openstack-swift | 13:05 | |
| *** nshaikh has joined #openstack-swift | 13:12 | |
| *** JuanManuelOlle has quit IRC | 13:18 | |
| *** nacim has quit IRC | 13:20 | |
| openstackgerrit | Yuan Zhou proposed a change to openstack/swift: Clean up swift-object-info/swift-get-nodes bin files https://review.openstack.org/86593 | 13:29 |
| *** saschpe has quit IRC | 13:32 | |
| *** saschpe has joined #openstack-swift | 13:37 | |
| *** judd7_ has joined #openstack-swift | 13:43 | |
| *** judd7 has quit IRC | 13:45 | |
| *** saju_m has joined #openstack-swift | 13:50 | |
| *** zackf has joined #openstack-swift | 13:55 | |
| *** corrigac has quit IRC | 14:02 | |
| *** G________ has joined #openstack-swift | 14:02 | |
| *** praveenkumar has quit IRC | 14:05 | |
| *** corrigac1 has joined #openstack-swift | 14:05 | |
| *** joeljwright has quit IRC | 14:06 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Python 3: fix tests on HTTP headers https://review.openstack.org/84711 | 14:08 |
| *** mmcardle has joined #openstack-swift | 14:10 | |
| *** psharma has quit IRC | 14:15 | |
| *** praveenkumar has joined #openstack-swift | 14:32 | |
| *** G________ has quit IRC | 14:32 | |
| *** chandan_kumar has quit IRC | 14:47 | |
| *** nshaikh has quit IRC | 14:49 | |
| *** mmcardle has quit IRC | 14:50 | |
| *** mmcardle has joined #openstack-swift | 14:51 | |
| *** tanee is now known as tanee-away | 14:58 | |
| *** Guest_ has joined #openstack-swift | 15:00 | |
| *** saju_m has quit IRC | 15:05 | |
| *** tanee-away is now known as tanee | 15:05 | |
| *** piyush1 has joined #openstack-swift | 15:05 | |
| *** foexle has quit IRC | 15:12 | |
| *** IRTermite|away has joined #openstack-swift | 15:12 | |
| *** saschpe has quit IRC | 15:14 | |
| *** saschpe has joined #openstack-swift | 15:23 | |
| *** amandap_ is now known as amandap | 15:27 | |
| *** _bluev has joined #openstack-swift | 15:33 | |
| *** mwstorer has joined #openstack-swift | 15:35 | |
| openstackgerrit | Donagh McCabe proposed a change to openstack/swift: Restrict keystone cross-tenant ACLs to IDs https://review.openstack.org/86430 | 15:39 |
| notmyname | good morning world | 15:40 |
| *** cheri has joined #openstack-swift | 15:42 | |
| *** mkollaro1 has joined #openstack-swift | 15:43 | |
| *** mkollaro has quit IRC | 15:43 | |
| *** IRTermite|away has quit IRC | 15:44 | |
| creiht | torgomatic: ping | 15:44 |
| creiht | https://bugs.launchpad.net/swift/+bug/1306027 | 15:44 |
| creiht | notmyname: -^ | 15:45 |
| notmyname | creiht: ack | 15:45 |
| creiht | notmyname: not really sure what to do about that | 15:47 |
| creiht | was hoping torgomatic would have some insight, if he isn't out | 15:47 |
| notmyname | creiht: ya. IIRC that patch was written because the devstack stuff was bombing on some manifests testing | 15:48 |
| notmyname | I don't remember exactly what the testing was. maybe not manifests | 15:48 |
| notmyname | creiht: in about 10 minutes I'm going to start a 2 hour run of meetings, so I won't be here in IRC | 15:50 |
| notmyname | creiht: is this somethign you are looking more at? in your opinion is this a release issue or just an FYI? | 15:50 |
| *** mkollaro1 has quit IRC | 15:50 | |
| * notmyname is assuming a release issue | 15:50 | |
| creiht | hehe | 15:50 |
| creiht | I dunno | 15:50 |
| creiht | notmyname: I just started looking at it this morning | 15:51 |
| notmyname | ok. thanks. let me know what you find | 15:51 |
| creiht | after pandemicsyn started noticing it in our dev cluster | 15:51 |
| clayg | creiht: the traceback on shutdown is the only problem - logging comes back up and starts working cleanly? | 15:53 |
| creiht | I'll check | 15:54 |
| creiht | thought might be a bit | 15:55 |
| creiht | Is it time to just stop supporting 2.6? :) | 15:55 |
| notmyname | we kinda did that a while back, tacitly. but with the "well, if it breaks 2.6 RAX will let us know and we'll fix it" | 15:57 |
| creiht | heh | 15:57 |
| clayg | ick, we still have the py2.6 unittets! | 15:57 |
| clayg | i don't really maintain my lucid saio anymore :\ | 15:58 |
| *** russellb has quit IRC | 15:59 | |
| *** russellb has joined #openstack-swift | 15:59 | |
| creiht | so doing a little experimentation it looks like everything still works | 16:01 |
| *** scohen has quit IRC | 16:02 | |
| creiht | just spews a lot of traceback info into the error logs | 16:02 |
| creiht | which is annoying | 16:02 |
| creiht | not sure if you want it to be a release blocker | 16:02 |
| *** redbo has quit IRC | 16:03 | |
| peluse | clayg: did we decide we *did* want to update SAOI for SP support (and just have the same davies on both rings)? I think its a good idea | 16:03 |
| peluse | SAIO... | 16:03 |
| *** redbo has joined #openstack-swift | 16:03 | |
| *** ChanServ sets mode: +v redbo | 16:03 | |
| creiht | there was another issue that we were having related to logs | 16:04 |
| creiht | I need to see if this was related, and will ask pandemicsyn when he gets in | 16:04 |
| clayg | peluse: do you need me to send you a new keyboard? | 16:05 |
| clayg | :P | 16:05 |
| peluse | bobf: sorry I missed your chat earlier... I'll be in and out today. Wrt your question though, remakerings has not yet been udpated to support multiple policies (this my question to clayg). You'd have to manually edit swift.conf and either manually run the same commands in remakerings (with -index after object) or just copy/rename the ring file that's created for you as the 2nd ring | 16:06 |
| clayg | creiht: yeah definately annoying, please do update if pandemicsyn thinks there maybe be other issues than the traceback... | 16:06 |
| peluse | clayg: I actually got one! Now I have to get used to it :) | 16:06 |
| clayg | lol! | 16:06 |
| *** mkollaro has joined #openstack-swift | 16:08 | |
| clayg | peluse: I think the storage_policies development guide should include instructions for modifying an existing saio to have storage policies setup, and the saio_development guide should link to that at the end, or maybe next to the remakerings and somewhere around the ref to swift.conf | 16:08 |
| peluse | clayg: or we could have 2 files built in SAIO that could be copied over to make it SP (swift.conf and remakerings) along with the instructions | 16:08 |
| peluse | clayg: or both really (instructions for modifying and making it easy to setup the first time) | 16:09 |
| creiht | clayg: so the thing that is kind of annoying is that the container-updater spams that every pass | 16:09 |
| * peluse will be back in a few... | 16:10 | |
| clayg | creiht: oh that's weird... | 16:10 |
| clayg | creiht: i wonder why? | 16:10 |
| creiht | dunno | 16:12 |
| creiht | just finding this out myself :) | 16:12 |
| creiht | I'm out for a bit, but will check back in later | 16:12 |
| creiht | btw, in our dev cluster, I see that error constantly running over and over again in the logs | 16:13 |
| creiht | for container-updater | 16:13 |
| creiht | likely because we don't have a huge number of containers in it | 16:13 |
| creiht | heh | 16:14 |
| creiht | jobs.log on one server is 1MB, jobs.error is 3G :) | 16:15 |
| *** cheri has quit IRC | 16:16 | |
| *** ashish_ has joined #openstack-swift | 16:17 | |
| mmcardle | does any one know what change the following refers to - "Prevent static large objects from containing other static large objects" in https://github.com/openstack/swift/blob/master/CHANGELOG#L432Â | 16:28 |
| *** gyee has joined #openstack-swift | 16:34 | |
| *** piousbox has joined #openstack-swift | 16:42 | |
| *** corrigac1 has left #openstack-swift | 16:42 | |
| *** fbo is now known as fbo_away | 16:46 | |
| *** saju_m has joined #openstack-swift | 16:48 | |
| clayg | creiht: I'm not really having any luck reproing | 16:48 |
| clayg | I mean I think just overriding close to catch and ignore the KeyError would fix it... | 16:49 |
| clayg | creiht: i'm on python2.6.9 tho :\ | 16:50 |
| *** tanee is now known as tanee-away | 16:51 | |
| *** ashish_ has quit IRC | 16:56 | |
| *** ashish_ has joined #openstack-swift | 16:57 | |
| *** ashish__ has joined #openstack-swift | 16:58 | |
| *** lpabon has joined #openstack-swift | 17:07 | |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Supress traceback in exitfunc on old versions of Python https://review.openstack.org/86654 | 17:08 |
| clayg | creiht: still trying to duplicate, bringing up a lucid box | 17:09 |
| *** csd has joined #openstack-swift | 17:17 | |
| ashish__ | I have a swift all in one installation on a single node.I am using a loopback device to create 4 nodes,I am running short of disk space.Could anyone tell me how to increase the storage capacity of my nodes. | 17:17 |
| ctennis | add more disks? | 17:19 |
| ashish__ | ctennis is there a way I can increase the space in the existing nodes. | 17:21 |
| ctennis | are you doing a loopback device on top of a file in a filesystem? I guess I don't understand how your device structure is setup. | 17:22 |
| ashish__ | ctennis I'm using a loopback device on top of the srv directory in ubuntu. | 17:23 |
| ctennis | right but a loopback device is connected to a file in the filesystem | 17:24 |
| ctennis | how did you create the device? | 17:24 |
| *** zaitcev has joined #openstack-swift | 17:25 | |
| *** ChanServ sets mode: +v zaitcev | 17:25 | |
| ashish__ | This is how I did it. http://paste.openstack.org/show/75489/ | 17:26 |
| notmyname | ashish__: just to make sure I understand this..you aren't using this as a production system right? | 17:27 |
| ashish__ | notmyname no I am using the swift all in one installation on a single node.Just for testing purpose | 17:28 |
| notmyname | ashish__: then in that case, delete the current loopback devices, change the "1GB" value in line 4 of your paste, and start over | 17:29 |
| ashish__ | notmyname but doing so makes me lose all my data.Doesnt it? | 17:29 |
| *** _bluev has quit IRC | 17:30 | |
| notmyname | yes. but you just said it's for testing purposes. if you've got data on there you can't lose, then it's not for testing purposes | 17:30 |
| zaitcev | portante: I'm sorry, but I'm not a fan of confusing "automation" when I am trying to pinpoint problems, which is what testing is for. If I set the variable, I want it to run functests. If I don't set the variable, I want it to do what it did before. And your patch blows that. I hope I am not asking for too much. | 17:33 |
| ashish__ | notmyname got it.How can I add a new node to the existing loopback device | 17:33 |
| notmyname | ashish__: you mean adding a new loopback device? | 17:34 |
| openstackgerrit | Donagh McCabe proposed a change to openstack/swift: Restrict keystone cross-tenant ACLs to IDs https://review.openstack.org/86430 | 17:34 |
| *** d89 has quit IRC | 17:35 | |
| ashish__ | notmyname yes I want to add a new loopback device and integrate it with swift | 17:35 |
| notmyname | ashish__: and just to be explicit, your setup is ok for swift dev work, but it isn't really a good POC to simulate a prod deployment, nor is it a good prod deployment at all | 17:35 |
| notmyname | ashish__: wouldn't that simply be changing the loop constraints in your pastebin? eg change 4 to 5 or 6? | 17:36 |
| ashish__ | notmyname I did not get your point. | 17:36 |
| *** mmcardle has quit IRC | 17:37 | |
| notmyname | ashish__: I'm not sure what you are trying to do. what are you hoping to gain by adding another loopback device? what is the problem you are solving? | 17:38 |
| ctennis | ashish__ add another loopback device like you have in another spot, or use truncate to extend the current one and use xfs to extend the mount size | 17:39 |
| ashish__ | notmyname my current loopback device is running short of space.So is there a solution | 17:40 |
| ashish__ | ctennis Thanks.But would that add more space to my existing nodes | 17:40 |
| ctennis | you need to use the ring builder to leverage the new space | 17:40 |
| ashish__ | ctennis could you tell me how its actually done | 17:41 |
| ctennis | http://docs.openstack.org/developer/swift/admin_guide.html | 17:41 |
| ctennis | mostly the same way you did it the first time | 17:41 |
| ashish__ | ctennis thanks But I still did not get what needs to done. | 17:46 |
| zaitcev | ashish_: once you have the device added and mounted under /src/node, do the "swift-ring-builder add" like you did before only with the new device. | 17:49 |
| *** mkollaro has quit IRC | 17:50 | |
| *** piyush1 has quit IRC | 17:53 | |
| ashish__ | zaitcev i am following the http://docs.openstack.org/developer/swift/development_saio.html tutorial but in this tut there's no mention of swift ring builder.Thanks | 17:54 |
| zaitcev | Good grief. That's a sudden blowback for those who thought that doc/saio/bin was a good idea. The users have NO CLUE what they fuck they are doing anymore. | 17:57 |
| ctennis | ashish__: it's happening in the remakerings script | 17:58 |
| ctennis | look in there, it will have the steps | 17:58 |
| zaitcev | commit ba5fe5f39e18fe8e1483a3b5851f044fe3358d0b | 17:58 |
| zaitcev | Author: Peter Portante <peter.portante@redhat.com> | 17:58 |
| ashish__ | ctennis Thanks | 17:59 |
| zaitcev | "Many of the large files are included in the tree and the script now leverages a checked out swift tree to provide those files so that users don't have to cut/paste text from the document. The contents of those files are still included in the document for reference." | 17:59 |
| zaitcev | But they are not included | 17:59 |
| clayg | lol @ zaitcev | 18:02 |
| ashish__ | zaitcev. Thanks got ur point.So in case I want to increase the capacity of the existing loopback device Would this work http://paste.openstack.org/show/75492/ | 18:03 |
| zaitcev | ashish_: NO | 18:03 |
| zaitcev | original remakerings only contains commands for stock loopbacks | 18:03 |
| ashish__ | zaitcev.Could you tell me how could that be done. | 18:03 |
| zaitcev | just edit the script, it should be obvious once you're in it | 18:04 |
| ctennis | ashish__: I think your best bet is to just start over, and make a bigger loopback to begin with | 18:05 |
| ashish__ | ctennis actually I wanted to learn how to scale up swift. | 18:07 |
| zaitcev | it can't be that hard. vi ~/bin/remakerings | 18:08 |
| ashish__ | zaitcev thanks what do I need to do after this | 18:10 |
| clayg | so i added a localrc option to make vagrant-swift-all-in-one's loopback partition size configurable... but I guess that's not really what ashish__ needs... | 18:13 |
| *** Guest_ has joined #openstack-swift | 18:14 | |
| *** ashish_ has quit IRC | 18:20 | |
| *** ashish__ has quit IRC | 18:20 | |
| *** ashish_ has joined #openstack-swift | 18:22 | |
| *** ashish_ has quit IRC | 18:22 | |
| *** ashish_ has joined #openstack-swift | 18:23 | |
| *** zul has quit IRC | 18:25 | |
| portante | zaitcev: what files are not included? | 18:25 |
| creiht | clayg: hey just got back sorry, any luck reproducing? | 18:26 |
| zaitcev | portante: the one we're not supposed to copy-paste anymore | 18:26 |
| zaitcev | portante: in ashish's case it's remakerings | 18:26 |
| *** zackf1 has joined #openstack-swift | 18:26 | |
| *** zackf has quit IRC | 18:26 | |
| clayg | creiht: still trying to get my swift intalled on lucid - I'm sort of not trying that hard, but I occasionally flip over to that window and punch a few buttons | 18:27 |
| creiht | haha | 18:27 |
| zaitcev | As I see the problem here, ashish_ managed to install SAIO without ever running swift-ring-builder or understanding what it does. As a result, he's completely powerless when he needs to add one device to the cluster. | 18:27 |
| clayg | creiht: it's like impossible to install swift on lucid now right? You have to hold your tounge really funny and make some sort of child sacrifice to the gods or reasonable python packages or something? | 18:28 |
| creiht | lol | 18:28 |
| zaitcev | portante: So, your commit vastly improved convenience of quick-setup of SAIO, but made it less of an education tool, evidently... | 18:28 |
| creiht | clayg: that's only true if you try installing it with pbr ;P | 18:28 |
| portante | zaitcev: remakerings appears in my tree | 18:29 |
| *** mkollaro has joined #openstack-swift | 18:29 | |
| zaitcev | portante: oh. You meant that they're included into tree, not into development_saio.rst | 18:30 |
| portante | they are in both | 18:30 |
| *** zackf1 has quit IRC | 18:30 | |
| *** zackf has joined #openstack-swift | 18:30 | |
| clayg | portante: no, not the bin scripts i don't think | 18:30 |
| portante | we use the doc include feature last I checked | 18:30 |
| portante | clayg: hmmm | 18:30 |
| clayg | the didn't get the inline ref feature | 18:30 |
| ashish_ | zaitcev I have run the swift-ring-builder before and do know what it does.But am still not getting how to add a device.Please help.i am still a third year student.Learnin Days. | 18:30 |
| zaitcev | grep -l swift-ring-builder hail/swift-tip/doc/source/*.rst | 18:31 |
| *** jergerber has joined #openstack-swift | 18:31 | |
| zaitcev | okay, perhaps it's an isolated case | 18:32 |
| clayg | zaitcev: don't the multi node docs hav something on this or the admin docs or somewhere? | 18:33 |
| zaitcev | clayg: multi has everything | 18:33 |
| ashish_ | zaitcev could you please tell me what should I do now. | 18:33 |
| clayg | ashish_: read this http://docs.openstack.org/developer/swift/admin_guide.html | 18:35 |
| zaitcev | ashish_: look at http://docs.openstack.org/developer/swift/howto_installmultinode.html, hopefuly it helps. It has swift-ring-builder xxx add | 18:35 |
| zaitcev | oh yeah, clayg is right, that one is even better | 18:35 |
| zaitcev | portante: so, what do you think about the snippet of code I posted to 66108? | 18:36 |
| zaitcev | portante: if that one is accepptable to you, it would be fine to me too | 18:36 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Keep status_changed_at up-to-date with status changes. https://review.openstack.org/86680 | 18:36 |
| ashish_ | clayg zaitcev thanks for your help.and thanks for being so patient. | 18:37 |
| zaitcev | portante: my test.conf has no constraints, and why would it? | 18:37 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Add includes of referenced SAIO bin scripts https://review.openstack.org/86681 | 18:37 |
| portante | zaitcev: let me look | 18:37 |
| *** zul has joined #openstack-swift | 18:39 | |
| *** ashish_ has quit IRC | 18:40 | |
| creiht | clayg: yeah so puting a try/except around the close code fixes the issue | 18:40 |
| *** piyush1 has joined #openstack-swift | 18:40 | |
| clayg | creiht: oh good? | 18:40 |
| creiht | yeah dunno? :) | 18:40 |
| creiht | it "fixes" the symptom | 18:40 |
| creiht | but not sure if it is really the right thing to do | 18:40 |
| clayg | i don't think it's acctually a leak | 18:41 |
| clayg | the fact that it happens in close makes me think that gc is just beating the exit handler to it | 18:41 |
| creiht | yeah | 18:41 |
| clayg | yeah I think it's a bug in python 2.6.5 | 18:41 |
| *** piyush2 has joined #openstack-swift | 18:42 | |
| creiht | heh | 18:42 |
| clayg | the 2.6.9 and 2.7 close fucntions do a if "thing" in _handlers: del _handlers["thing"] | 18:42 |
| clayg | my lucid install just does the del | 18:42 |
| portante | zaitcev: I am not sure that is any more correct, but if you'd like, switching the behavior of the existing functional tests to key off of /info first would probably clear this all up | 18:44 |
| *** piyush1 has quit IRC | 18:44 | |
| zaitcev | portante: Do you mean fetching constraints from /info? But why is it so critical anyway? | 18:45 |
| portante | and I don't think auth_host is enough | 18:45 |
| creiht | clayg: oh just noticed your patch | 18:45 |
| zaitcev | please tell me more | 18:45 |
| gholt | notmyname: wth is swift/__init__.py only copyright HP? | 18:45 |
| *** Guest_ has quit IRC | 18:45 | |
| portante | I was trying not to change the behavior | 18:45 |
| portante | currently, tests are skipped if the proper constraints are not provided | 18:45 |
| clayg | creiht: yay finall got it duped! | 18:46 |
| portante | so instead of skipping them, we run the in-process | 18:46 |
| redbo | I'm gonna go put copyrights on all the empty __init__.py files. | 18:46 |
| portante | but the code today does not gracefully handle a test.conf that does not have auth_host, does it? | 18:46 |
| *** gyee has quit IRC | 18:46 | |
| zaitcev | But you did change. Look, I have a computer. This laptop. In the master, I run ./.functests, they test my server. In your patch tree, I run ./.functests, they force in-process. It's totally changed! | 18:46 |
| creiht | clayg: when I was playing around, I only needed to add a try/except in the close method of LoggingHandlerWeakRef | 18:46 |
| notmyname | gholt: that happened when an HP employee (ie mordred) replaced the entire contents with the pbr stuff | 18:46 |
| redbo | or maybe I'll find any files I modified and add a copyright me. | 18:47 |
| zaitcev | portante: currently the tests are not skipped if constraints are not provided... Well some of them are, but only a small number. | 18:47 |
| clayg | creiht: oh that sounds better! | 18:47 |
| creiht | :) | 18:47 |
| portante | zaitcev: okay, I'll see what I can do | 18:48 |
| creiht | I was just about to submit a patch, and saw you had already thrown something up there | 18:48 |
| zaitcev | portante: but for sure, pulling from /info sounds good, as it will reduce the skipped number even further | 18:48 |
| gholt | notmyname: Well, aside from the long argued sillyness over copyright notices, I was trying to figure out what to put on this file I'm working in that still has OpenStack Foundation, which I understand is not desirable. | 18:48 |
| portante | zaitcev, creiht: can we get the eventlet change in at least? | 18:49 |
| creiht | notmyname: can we just remove the copyright declarations at the top of each file? | 18:49 |
| gholt | I figured I'd grab whatever was in swift/__init__.py -- but... | 18:49 |
| zaitcev | portante: sure | 18:49 |
| creiht | portante: what eventlet change? | 18:49 |
| creiht | sorry been a bit distracted with the logging stuff | 18:49 |
| portante | creiht: the change to swift from eventlet to gevent | 18:49 |
| portante | ;) | 18:50 |
| portante | just kidding | 18:50 |
| creiht | lol | 18:50 |
| gholt | I'll just myself | 18:50 |
| portante | the change to remove threading and use eventlet in its place | 18:50 |
| portante | I'll just ... myself? | 18:50 |
| redbo | insert your own verb | 18:50 |
| gholt | Heheh | 18:50 |
| creiht | portante: ahh, yes I'll take a look again when I get back on the regularly scheduled review train | 18:50 |
| portante | all, so you all want gevent, then? | 18:50 |
| zaitcev | portante: BTW, could you +2 https://review.openstack.org/85909? You reviewed the unified patch once, way before Luis and DATADIR, and it did not change since, so should be 20 seconds +2 for you, right? | 18:50 |
| clayg | creiht: i abandoned mine can you kick something up? I have an env to test it now | 18:50 |
| creiht | clayg: sure | 18:51 |
| portante | zaitcev: I think I was looking at that midstream when I got pulled off, will get back to it | 18:51 |
| zaitcev | I thought the proper meme was "I accidentially the whole Internet". | 18:51 |
| clayg | portante: zaitcev: did in process merge yet? | 18:53 |
| zaitcev | creiht: "the eventlet" is this https://review.openstack.org/85782 together with chunks() moved into scope. Peter just never gives up | 18:53 |
| creiht | zaitcev: lol | 18:53 |
| zaitcev | clayg: no, no. I was just arguing with portante about it a few minutes ago | 18:53 |
| zaitcev | clayg: there was also some forgotten code in there so it also got -1 on that grounds | 18:54 |
| clayg | zaitcev: well bummer | 18:54 |
| portante | clayg, just doin what I am told | 18:54 |
| portante | clayg: can you review the threading->eventlet change on which it depends? | 18:55 |
| zaitcev | yeah, baby steps | 18:55 |
| zaitcev | I am suitably ashamed about bikeshedding one little def chunks() while quickly agreeing to the whole eventlet thing, even including the use_hub(), which I find cargo-culted and unnecessary. | 18:56 |
| *** zhiyan is now known as zhiyan_ | 18:56 | |
| peluse | gholt: you there? | 18:57 |
| zaitcev | redbo explained how it got into servers. Surely the same logic does not apply to tests. And if you want super reliablity, just use use_hub('selects'). | 18:57 |
| zaitcev | oh well | 18:58 |
| gholt | peluse: Yes, but having trouble keeping up with this channel, so be sure to gholt: prefix stuff for me like you are. :) | 18:58 |
| clayg | portante: oh yeah i noticed the depends when I was looking at it last night but forgot to go back to it | 18:58 |
| *** mkollaro has quit IRC | 18:58 | |
| portante | clayg: thanks | 18:59 |
| redbo | I'd like to try switching back to epoll in servers and do a big test, maybe it was some wonky kernel we were using at the time or something. Maybe I'll trick creiht into doing that, he likes that kind of stuff. | 18:59 |
| portante | redbo: I'd be game to test it here as well | 18:59 |
| peluse | gholt: so I have a small request. patch to add policy support to container sync. Could you take a gander for 4 things (1) general review, pretty small change (2) Jenkins fails but all of my tests pass and logically, based on my understanding, they should and the third thing I'll hold off on for now :) | 19:00 |
| peluse | gholt: https://review.openstack.org/#/c/86469/ | 19:00 |
| portante | could we get a patch that makes it easily configurable? | 19:00 |
| zaitcev | we probably could but please don't | 19:00 |
| peluse | gholt: 3 things not 4, man I just can't win w/typing anymore | 19:00 |
| zaitcev | we're swimming in options as we are | 19:00 |
| portante | zaitcev: we need an option so that it is easy to switch a cluster of machines without pushing new code out | 19:01 |
| openstackgerrit | Chuck Thier proposed a change to openstack/swift: Fix logging issue when services stop on py26 https://review.openstack.org/86684 | 19:01 |
| portante | if we don't add one, I'll create a branch with one so that we can make this easily deployable | 19:01 |
| gholt | peluse: Hah, sure thing. I'll put it on my "to do". | 19:01 |
| creiht | clayg: -^ | 19:01 |
| clayg | says the guy that wants to make the entire implementation configurable? I think adding an option it totally reasonable - the best choice probably depends on the platform... | 19:01 |
| zaitcev | portante: well, you have a point... | 19:01 |
| redbo | portante: this is the error that happens when it gets in a bad state. https://github.com/eventlet/eventlet/blob/master/eventlet/hubs/hub.py#L119 | 19:01 |
| portante | reallly? | 19:02 |
| peluse | gholt: thanks! Will see if I can't digest more of the tests in the meantime | 19:02 |
| portante | I just ran into that in in-process functional tests | 19:02 |
| *** jamie_h has quit IRC | 19:03 | |
| portante | I thought that was due to threading, where one thread was running the hub, but a greenlet object got handed to another thread | 19:03 |
| portante | trampoline away a ... pooh! | 19:03 |
| portante | poof! | 19:03 |
| portante | sorry | 19:03 |
| portante | didn't mean to mess up the channel, sorry, cleaning it now quickly | 19:03 |
| openstackgerrit | Chuck Thier proposed a change to openstack/swift: Fix logging issue when services stop on py26 https://review.openstack.org/86684 | 19:03 |
| * portante does anybody have any clorox? | 19:03 | |
| zaitcev | nuke it from orbit. it's the only way to make sure. | 19:03 |
| creiht | I can never seem to submit a patch correctly the first time :/ | 19:04 |
| zaitcev | lemme guess - forgot pep8 | 19:04 |
| redbo | yeah, it's supposed to catch two greenthreads trying to simultaneously read or write a socket | 19:04 |
| creiht | zaitcev: heh, somehow added some extra spaces | 19:04 |
| *** zackf has quit IRC | 19:05 | |
| *** zackf1 has joined #openstack-swift | 19:05 | |
| portante | redbo: so should we have some kind of ssbench scenario set we can hammer on? | 19:05 |
| portante | or were you just going to run it on a node in some staging environment at rax? | 19:06 |
| portante | we are building a 3 proxy, 9 storage node cluster here in a lab, and have a simple 3 node swift cluster immediately available | 19:06 |
| *** Trixboxer has quit IRC | 19:07 | |
| redbo | I'll probably run some sort of benchmark against it | 19:07 |
| clayg | creiht: can you trick redbo or gholt into merging the logging fix - it works fine | 19:07 |
| creiht | clayg: thanks for taking a look | 19:08 |
| zaitcev | logging sounds vaguely familiar | 19:08 |
| redbo | well as long as you promise it works | 19:08 |
| zaitcev | go go gadget https://review.openstack.org/#/q/status:open+project:openstack/swift,n,z | 19:08 |
| zaitcev | wait, what referent = self() | 19:09 |
| zaitcev | you're taking self and invoking __call__() here? right? | 19:10 |
| creiht | zaitcev: that was already there, so don't ask me :) | 19:10 |
| zaitcev | I need to read how weak refs work, then | 19:10 |
| zaitcev | oh that redbo ninjaed me | 19:11 |
| creiht | yeah I'm not entirely sure, but I think some of that stuff torgomatic just pulled from the py27 tree | 19:11 |
| creiht | notmyname: it might be worth considering adding https://review.openstack.org/#/c/86684/ to icehouse | 19:12 |
| notmyname | ya, I was just about to ask about that | 19:13 |
| notmyname | what's the consensus? | 19:13 |
| notmyname | creiht: clayg: portante: zaitcev: redbo: backport that py26 issue or not? | 19:13 |
| creiht | I'm on the fence | 19:13 |
| creiht | though leaning slightly on the side of backport | 19:14 |
| zaitcev | notmyname: I think yes | 19:14 |
| notmyname | creiht: same here | 19:14 |
| redbo | I think it makes it unusable on 2.6. | 19:14 |
| creiht | anyone running a cluster on py26 will see a lot of stack traces in their logs | 19:14 |
| zaitcev | what, really? It looked completely safe, which is why +2 | 19:14 |
| redbo | the problem, not the solution :) | 19:15 |
| creiht | zaitcev: sorry, the above is true without the patch | 19:15 |
| portante | notmyname: backport, it is just ugly when it starting filling logs | 19:15 |
| portante | notmyname: which is any RHEL base system, and/or I think CentOS, right? | 19:16 |
| notmyname | portante: right. which is why I'm leaning to the backport | 19:16 |
| zaitcev | RHEL 6 | 19:16 |
| zaitcev | 7 is 2.7 I think | 19:16 |
| notmyname | zaitcev: right | 19:17 |
| clayg | it works on 2.6.9 - but not 2.6.5 best I can tell... | 19:17 |
| clayg | maybe you could get ubuntu to backport the fix to python 2.6.5's logging? | 19:18 |
| clayg | so is icehouse like out out or just in RC? | 19:18 |
| creiht | I thought we have until Thurs for final cut | 19:19 |
| notmyname | right. we can do another RC | 19:19 |
| notmyname | I'll start turning the wheels on that | 19:19 |
| *** chandan_kumar has joined #openstack-swift | 19:19 | |
| creiht | notmyname: cool | 19:19 |
| creiht | lol... I was just reminded that today is Thursday :) | 19:20 |
| *** davidhadas has joined #openstack-swift | 19:20 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Support to Container Sync https://review.openstack.org/86469 | 19:23 |
| peluse | gholt: ^ might see why the Jenkins tests were failed but would still appeciate a review and then have a test question after that. Not super urgent though, thanks | 19:25 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Add includes of referenced SAIO bin scripts https://review.openstack.org/86681 | 19:27 |
| clayg | portante: ^ the rendering on those literalincludes was messed up cause of :: - i also added test.conf just cause | 19:32 |
| notmyname | "OpenSSL Software Foundation, which funds crypto for most of the Web, set 3-day fundraising record this week: $841.70" | 19:32 |
| portante | clayg: cool, so that means I can +2 my own patch now, huh? ;) | 19:34 |
| clayg | man that's stupid - what good is it for my to have python2.6 installed if the version I have fixes all the bugs in lucid's python that I acctually wanna avoid | 19:34 |
| clayg | portante: fuck that it's just docs - we'll both +2 it and make zaitcev merge it since he was whining about it ;) | 19:34 |
| * clayg hugs zaitcev | 19:34 | |
| *** krtaylor has quit IRC | 19:36 | |
| zaitcev | clayg: at least let me run python setup.py build_sphinx, I want to see what .literalinclude daoes | 19:37 |
| clayg | vagrant-swift-all-in-one has a super hot autodoc script that's pretty great for hacking/reviewing doc changes... | 19:38 |
| clayg | it watches file changes; builds sphix; uploads them to swift | 19:38 |
| clayg | so you just point your web browser at swift and hack away refreshing every so often... | 19:38 |
| clayg | s/at swift/at your saio, autodoc spits out a link/ | 19:39 |
| *** csd has quit IRC | 19:40 | |
| zaitcev | oh... I used file:///q/zaitcev/hail/swift-tip/doc/build/html/development_saio.html as a peasant I am | 19:41 |
| notmyname | backport patch https://review.openstack.org/#/c/86689/ | 19:43 |
| *** wayneeseguin has left #openstack-swift | 19:50 | |
| *** mwstorer has quit IRC | 19:57 | |
| openstackgerrit | gholt proposed a change to openstack/swift: TempURL: Fixed bug with \r or \n in disposition. https://review.openstack.org/86691 | 19:58 |
| *** krtaylor has joined #openstack-swift | 20:03 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Support to Container Sync https://review.openstack.org/86469 | 20:03 |
| clayg | gholt: lol @ copyright attribution on that change | 20:10 |
| gholt | Ssshh. ;) | 20:11 |
| creiht | clayg: I told him he should run his script on all of the files :) | 20:12 |
| clayg | i was *hoping* he scripted it! | 20:12 |
| creiht | gholt: you should claim the copyright of the copyright block since you wrote it | 20:12 |
| clayg | since i'm not in the list I'm not going to explicitly complain that I don't want copyright any files, but i wonder if some of those folks would (assuming you didn't already get their explicit permission) | 20:14 |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Documentation https://review.openstack.org/85824 | 20:17 |
| *** saju_m has quit IRC | 20:28 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Documentation https://review.openstack.org/85824 | 20:29 |
| *** saju_m has joined #openstack-swift | 20:29 | |
| *** gyee has joined #openstack-swift | 20:29 | |
| *** ashish_ has joined #openstack-swift | 20:36 | |
| clayg | gholt: might be nice to open a bug for documenting this "invalid HTTP Content-Disposition header" behavior for weary swift users in the the future that may bump into the problem on older swift deployments... | 20:41 |
| gholt | Ah, yeah | 20:42 |
| gholt | As far as what to do with it, not 100% sure either. urllib quote I think is wrong because it'd make stuff name %xx on people's machines... maybe. I don't actually /know/. :/ | 20:42 |
| clayg | i would sorta imagine most browsers would figure that one out, somebody should test it :D | 20:44 |
| *** ashish_ has quit IRC | 20:44 | |
| gholt | Hey, I'm cool with underscores. :P | 20:47 |
| clayg | nice | 20:48 |
| clayg | gholt: apparently so is torgomatic! | 20:48 |
| gholt | Heheh | 20:48 |
| creiht | isn't torgomatic supposed to be doing something else other than looking at code? :) | 20:48 |
| openstackgerrit | gholt proposed a change to openstack/swift: TempURL: Fixed bug with \r or \n in disposition. https://review.openstack.org/86691 | 20:49 |
| notmyname | from old irc channel logs, "copyright notices are up to personal/project discretion. i think projects benefit from adding and maintaining them, but that's just my opinion" | 20:49 |
| notmyname | from http://eavesdrop.openstack.org/irclogs/%23openstack-infra/%23openstack-infra.2014-04-03.log | 20:50 |
| gholt | That's not too old | 20:50 |
| portante | fruit fly years? | 20:50 |
| creiht | notmyname: so I guess he sets official openstack copyright procedure? :) | 20:51 |
| creiht | I personally think we should just remove them all | 20:52 |
| notmyname | creiht: oh, so you set official openstack copyright procedure now? ;-) | 20:52 |
| creiht | no | 20:52 |
| creiht | just expressing my opinion | 20:53 |
| gholt | Yeah, I'm not sure who to trust, somebody named fungi or somebody named creiht. :P | 20:53 |
| creiht | lol | 20:53 |
| notmyname | the only "official" thing I've ever found is https://wiki.openstack.org/wiki/Documentation/Copyright | 20:54 |
| notmyname | which doesn't say much | 20:54 |
| creiht | yeah it has always been ambiguous | 20:55 |
| notmyname | I'm pretty sure just having the individual authors listed out is wrong, though :-) | 20:55 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Expose container info on deleted containers. https://review.openstack.org/86708 | 20:55 |
| notmyname | I'd be fine with removing them too | 20:56 |
| creiht | notmyname: I think we need to comission a 2 year study from the board and call it defcopyright | 21:00 |
| notmyname | creiht: and our deliverable will be the process by which we can determine what the right copyright policy is? | 21:01 |
| *** csd has joined #openstack-swift | 21:01 | |
| creiht | notmyname: well I'm sure first we would have to define what required sections of code must maintain a copyright | 21:04 |
| gholt | clayg: Finally got the testing done. Url quoting won't work. Well, I mean it'd "work" but you end up with %xx named files downloaded. | 21:07 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Load constraints from cluster info dictionary https://review.openstack.org/86710 | 21:08 |
| openstackgerrit | Samuel Merritt proposed a change to openstack/swift: Check swift.conf MD5 with recon https://review.openstack.org/86114 | 21:08 |
| *** piyush2 has left #openstack-swift | 21:08 | |
| gholt | clayg: http://stackoverflow.com/questions/93551/how-to-encode-the-filename-parameter-of-content-disposition-header-in-http | 21:10 |
| gholt | :) | 21:10 |
| portante | zaitcev, acoles: ^^^ | 21:10 |
| portante | functional tests no longer rely on direct constraints, only use cluster info of target swift instance for functional tests | 21:10 |
| portante | torgomatic: ^^^ | 21:10 |
| portante | this is not tied to the in-process stuff | 21:11 |
| gholt | clayg: So mostly we just need to at least return valid html, which it wasn't. Past that, whatever goes for the moment. | 21:11 |
| gholt | clayg: torgomatic: I can always use quote(name, safe='/ +-blah') | 21:18 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Keep status_changed_at up-to-date with status changes. https://review.openstack.org/86714 | 21:19 |
| *** peterbusque has quit IRC | 21:23 | |
| *** peterbusque has joined #openstack-swift | 21:23 | |
| *** chandan_kumar has quit IRC | 21:23 | |
| *** peterbusque has quit IRC | 21:25 | |
| torgomatic | gholt: yeah, I don't actually have that strong an opinion on quote vs. underscores; either one is way better than invalid HTTP headers | 21:33 |
| torgomatic | quote seems *slightly* better, but... meh | 21:33 |
| clayg | gholt: that's weird, when I linked that so post in my inline comment my reading seemed to indicate quote was the way to go - which browser was too dumb to figure out thos %XX in the filename - do you bother to test the filename*=UTF-8'' thing? | 21:37 |
| *** Guest____ has joined #openstack-swift | 21:38 | |
| gholt | safari, I didn't do the random punctuation thing, no | 21:38 |
| gholt | I'm cool with quote with extra safe chars, seems reasonable | 21:38 |
| gholt | Just figure I'll wait until Monday to commit that in case a strong other opinion comes up. | 21:39 |
| creiht | I think gerrit just likes to troll me | 21:40 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Merge container storage_policy_index values. https://review.openstack.org/86720 | 21:44 |
| tdasilva | peluse: I just did a quick pass of your storage policy doc patch and noticed that there's a TODO for adding info to the SAIO instructions. I was wondering if that wouldn't complicate the saio deployment even more. For some reason I see SP as a more advanced feature and SAIO as a way for people new to swift just get started with it...what do you think? | 21:45 |
| notmyname | tdasilva: while the SAIO may be used as a "getting started" thing, that's not the intent. the SAIO is for contributors doing dev work | 21:46 |
| *** krtaylor has quit IRC | 21:46 | |
| tdasilva | notmyname: I see and in part I agree with you, but then I wonder if it would make sense to then create sections as "here's what you need to do" to just try it out and here's the complete instructions to setup a dev. environment... | 21:48 |
| notmyname | tdasilva: yes. I think the bare minimum to get a distributed storage system up and running starts to look close to the SAIO, therefore nobody has bothered making a second, seprate doc | 21:49 |
| zaitcev | portante: so, the master cannot func-test my Grizzly cluster anymore, is that how it is? | 21:50 |
| zaitcev | on n/m | 21:52 |
| zaitcev | curl -v -k https://rhev-a24c-01.mpc.lab.eng.bos.redhat.com/info | 21:53 |
| clayg | zaitcev: well just cause info is there doesn't mean all of the constraints are listed... but if you're running and old deployment you can always checkout the functests that go with that version yeah? | 21:57 |
| zaitcev | clayg: I figure I probably could | 22:01 |
| clayg | heh | 22:01 |
| clayg | notmyname: so i was wrong, the overhead of keeping changes I need for the reconciler uptodate against feature/ec and pre-storage-policies is going to be too high | 22:02 |
| notmyname | clayg: bah | 22:02 |
| notmyname | clayg: I'm fine with keeping the wiki page updated | 22:02 |
| notmyname | clayg: it's not a scalable solution, but we're only talking about a few contributors and patches for a short about of time | 22:03 |
| clayg | notmyname: I think it's going to end up being better just getting stuff merged to ec and then pull them out to master after that | 22:03 |
| clayg | I think the pre-storage-policy topic is a good idea - I think it's just better for changes that are already on master | 22:04 |
| notmyname | changes already on masteR? | 22:04 |
| clayg | yeah changes that are already on feature/ec and could be pulled onto mater - sorry | 22:04 |
| notmyname | ok, that makes sense. that's what we talked about (or at least I understood) | 22:05 |
| clayg | part of the problem is it's easier to cherry pick a diff against master over to ec than the other way around | 22:05 |
| clayg | notmyname: well i've been running into a lot of stuff working on the reconciler, but I end up with patches like this that just suck to maintain -> https://review.openstack.org/#/c/86714/ | 22:06 |
| clayg | it's a mirror of a depdendency chain that's purposed to master - such a chore when I need to go back and fix the tip of the chain on master and then redo the cherry picks to feature/ec so I can rebase the replicator and reconciler changes - ick | 22:07 |
| notmyname | makes sense | 22:08 |
| notmyname | clayg: so this seems like no change other than the lack of same topic across patches | 22:09 |
| *** yuanz has joined #openstack-swift | 22:09 | |
| clayg | notmyname: well... no I don't think we loose the topic name across patches - that was easy | 22:10 |
| notmyname | clayg: ah, ok | 22:10 |
| clayg | notmyname: i'm just not going to be doing anymore for stuff that I want on feature/ec | 22:11 |
| notmyname | ... | 22:11 |
| notmyname | I'm having hard time following. phone call? | 22:12 |
| *** yuan has quit IRC | 22:12 | |
| clayg | so like instead of feature/ec -> mirror-pre-storage-policy-change -> reconciler, it's gunna be feature/ec -> some-change-we'll-eventually-have-to-dig-out-of-feature/ec -> reconciler | 22:12 |
| clayg | nah, i gotta sign off for a bit - baseball time - we can talk about it tomorrow | 22:15 |
| notmyname | ok | 22:15 |
| *** saju_m has quit IRC | 22:20 | |
| gholt | peluse: That container sync policy stuff seems reasonable to me. I'm not sure how much I should read into a lot of that just yet. But seems totally fine. | 22:27 |
| peluse | gholt: thanks. still trying to figure out the Jenkis failure but a few more shotgun type things to try first. I can't think of any reason why you can't/shouldn't be able to sync containers with different policies, can you? (not realted to the failure BTW) | 22:36 |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Support to Container Sync https://review.openstack.org/86469 | 22:38 |
| *** zackf1 has quit IRC | 22:41 | |
| notmyname | this coming sunday (right before the red hat summit), there is going to be a hackathon on a few different storage projects, including swift | 22:49 |
| notmyname | any ideas on good topics for people to tackle? | 22:50 |
| openstackgerrit | paul luse proposed a change to openstack/swift: Add Storage Policy Documentation https://review.openstack.org/85824 | 23:06 |
| *** jergerber has quit IRC | 23:29 | |
| *** yuan has joined #openstack-swift | 23:44 | |
| *** yuanz has quit IRC | 23:46 | |
| *** Guest____ has quit IRC | 23:47 | |
| zaitcev | Make them run down the bug list at Launchpad | 23:47 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!