| *** klrmn has joined #openstack-swift | 00:08 | |
| *** kiennt has joined #openstack-swift | 00:27 | |
| *** itlinux has joined #openstack-swift | 00:34 | |
| *** mat128 has joined #openstack-swift | 00:35 | |
| *** itlinux has quit IRC | 00:35 | |
| *** itlinux has joined #openstack-swift | 00:36 | |
| *** itlinux has quit IRC | 00:37 | |
| *** jcook has quit IRC | 00:38 | |
| *** jcook has joined #openstack-swift | 00:39 | |
| *** baojg has quit IRC | 00:44 | |
| esnyder | A slightly higher level question (which was motivating the memcache question) is; what strategies are best to handle a large number of reads of the same object at the same time. For instance, if we have a 3 copy storage policy, and some object /a/c/o is stored, and 1000 clients all try to read it at the same time, the disk(s) that /a/c/o lives on are going to get hammered. What are the various alternatives to | 00:46 |
|---|---|---|
| esnyder | improve performance in this situation? | 00:46 |
| *** gkadam has joined #openstack-swift | 00:48 | |
| *** itlinux has joined #openstack-swift | 00:51 | |
| *** gkadam has quit IRC | 00:57 | |
| openstackgerrit | Merged openstack/swift master: swob: Raise RuntimeError instead of IndexError https://review.openstack.org/502484 | 00:57 |
| *** tovin07_ has joined #openstack-swift | 01:00 | |
| *** catintheroof has quit IRC | 01:14 | |
| *** cshastri has joined #openstack-swift | 01:25 | |
| *** alenavolk has joined #openstack-swift | 01:37 | |
| *** kiennt has quit IRC | 01:42 | |
| *** baojg has joined #openstack-swift | 01:47 | |
| *** kiennt has joined #openstack-swift | 01:52 | |
| *** baojg has quit IRC | 01:53 | |
| *** itlinux has quit IRC | 02:06 | |
| *** bkopilov has quit IRC | 02:13 | |
| *** itlinux has joined #openstack-swift | 02:33 | |
| *** itlinux has quit IRC | 02:42 | |
| jrichli | esnyder: correct, only account/container metadata are stored in memcache. | 02:48 |
| jrichli | I don't know how to improve performance in the situation you provided, but the people who may be able to answer your question are at the Openstack PTG in Denver this week. So they may be slow to respond, but if you stay online, you will most likely get an answer eventually. You are more likely to get an answer during work hours for Denver this week. | 02:48 |
| esnyder | jrichli: cool, thanks for the help. | 02:48 |
| *** alenavolk has quit IRC | 03:17 | |
| *** alenavolk has joined #openstack-swift | 03:18 | |
| *** alenavolk has quit IRC | 03:19 | |
| *** alenavolk has joined #openstack-swift | 03:19 | |
| *** alenavolk has joined #openstack-swift | 03:20 | |
| *** alenavolk has joined #openstack-swift | 03:21 | |
| *** bkopilov has joined #openstack-swift | 03:31 | |
| *** alenavolk has joined #openstack-swift | 03:39 | |
| *** kei_yama has quit IRC | 03:41 | |
| *** links has joined #openstack-swift | 03:41 | |
| *** gkadam has joined #openstack-swift | 03:48 | |
| *** kei_yama has joined #openstack-swift | 04:01 | |
| *** gkadam has quit IRC | 04:04 | |
| *** itlinux has joined #openstack-swift | 04:08 | |
| *** gkadam has joined #openstack-swift | 04:13 | |
| *** stewie925 has quit IRC | 04:21 | |
| *** mat128 has quit IRC | 04:26 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:26 | |
| *** SkyRocknRoll has quit IRC | 04:27 | |
| *** stewie925 has joined #openstack-swift | 04:27 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:27 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 04:28 |
| *** klrmn has quit IRC | 04:28 | |
| *** akki has joined #openstack-swift | 04:31 | |
| akki | hello everyone | 04:33 |
| *** psachin has joined #openstack-swift | 04:36 | |
| *** okdas has quit IRC | 04:39 | |
| akki | i am working on two node architecture, one is for controller and other for compute. I have installed the basic and necessary services to launch an instance. Now i am installing the Ceilometer service on the controller node, and according to the guide, we have to make sure "gnocchi" is installed or not. Installing of gnocchi take incoming , indexer and storage drivers as parameters .I have taken "swift" for storage. This is the p | 04:41 |
| *** okdas has joined #openstack-swift | 04:43 | |
| *** okdas has joined #openstack-swift | 04:43 | |
| akki | my query is : Do we need one more node for storage (swift installed on it) or the gnocchi itself manages swift and we have to only configure the gnocchi.conf? | 04:44 |
| akki | please help | 04:45 |
| *** itlinux has quit IRC | 04:51 | |
| *** itlinux has joined #openstack-swift | 04:51 | |
| *** kiennt has quit IRC | 04:54 | |
| *** psachin has quit IRC | 04:55 | |
| *** itlinux has quit IRC | 04:56 | |
| *** akki has quit IRC | 04:59 | |
| *** psachin has joined #openstack-swift | 05:16 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Respond 400 Bad Request when Accept headers fail to parse https://review.openstack.org/502845 | 05:27 |
| *** esnyder has quit IRC | 05:32 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Make write_builder work with fractional replicas https://review.openstack.org/460302 | 05:35 |
| *** alenavolk has quit IRC | 05:46 | |
| *** alenavolk has joined #openstack-swift | 05:47 | |
| *** StevenK_ is now known as StevenK | 06:07 | |
| *** alenavolk has quit IRC | 06:10 | |
| *** rcernin has joined #openstack-swift | 06:16 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Add assertion about last-modified to object post test https://review.openstack.org/502893 | 06:20 |
| *** silor has joined #openstack-swift | 06:35 | |
| *** kiennt has joined #openstack-swift | 06:36 | |
| *** kiennt has left #openstack-swift | 06:37 | |
| *** skudlik has joined #openstack-swift | 06:39 | |
| *** pcaruana has joined #openstack-swift | 06:42 | |
| *** gkadam is now known as gkadam-brb | 06:58 | |
| *** silor has quit IRC | 07:00 | |
| *** tesseract has joined #openstack-swift | 07:00 | |
| *** HTTP_____GK1wmSU has joined #openstack-swift | 07:07 | |
| *** kong has quit IRC | 07:08 | |
| *** HTTP_____GK1wmSU has left #openstack-swift | 07:10 | |
| *** HTTP_____GK1wmSU has joined #openstack-swift | 07:23 | |
| *** gkadam-brb is now known as gkadam | 07:26 | |
| *** HTTP_____GK1wmSU has left #openstack-swift | 07:26 | |
| *** geaaru has joined #openstack-swift | 07:38 | |
| *** oshritf has joined #openstack-swift | 07:46 | |
| *** cbartz has joined #openstack-swift | 07:52 | |
| *** oshritf has quit IRC | 08:09 | |
| *** oshritf has joined #openstack-swift | 08:12 | |
| *** joeljwright has joined #openstack-swift | 08:14 | |
| *** ChanServ sets mode: +v joeljwright | 08:14 | |
| *** mvk has quit IRC | 08:20 | |
| *** oshritf has quit IRC | 08:25 | |
| *** oshritf has joined #openstack-swift | 08:37 | |
| *** oshritf has quit IRC | 08:40 | |
| *** oshritf has joined #openstack-swift | 08:45 | |
| *** mvk has joined #openstack-swift | 08:48 | |
| *** baojg has joined #openstack-swift | 08:55 | |
| *** baojg has quit IRC | 08:59 | |
| *** mvk has quit IRC | 09:27 | |
| *** mvk has joined #openstack-swift | 09:40 | |
| *** tovin07_ has quit IRC | 10:30 | |
| cbartz | Hello, has somebody an opinion on bug 1712507 ? | 10:56 |
| openstack | bug 1712507 in python-swiftclient "add tempurl auth for object operations" [Undecided,New] https://launchpad.net/bugs/1712507 | 10:56 |
| *** bkopilov has quit IRC | 10:58 | |
| *** hoonetorg has quit IRC | 11:08 | |
| *** joeljwright has quit IRC | 11:22 | |
| *** silor has joined #openstack-swift | 11:22 | |
| *** hoonetorg has joined #openstack-swift | 11:25 | |
| *** kei_yama has quit IRC | 11:26 | |
| *** cshastri has quit IRC | 11:47 | |
| *** hseipp has joined #openstack-swift | 11:56 | |
| psachin | cbartz: I think your suggesting is cool. I replied | 12:13 |
| *** mat128 has joined #openstack-swift | 12:15 | |
| cbartz | psachin: Cool, thank you. | 12:17 |
| *** fyxim has joined #openstack-swift | 12:20 | |
| psachin | cbartz: If you cloud provide more details regarding implementation..like when doing GET/POST, how will swift take input param like temp_url_sig etc. | 12:22 |
| *** joeljwright has joined #openstack-swift | 12:26 | |
| *** ChanServ sets mode: +v joeljwright | 12:26 | |
| *** sundbp has joined #openstack-swift | 12:31 | |
| sundbp | i'm looking at running swift using docker containers. as far as I can tell it should be fine as long as I ensure my actual data is mounted from a partition with xattr support (e.g. xfs). am I missing something that will cause swift problems in a container? | 12:32 |
| *** gkadam has quit IRC | 12:36 | |
| *** catintheroof has joined #openstack-swift | 12:48 | |
| *** bkopilov has joined #openstack-swift | 12:58 | |
| cbartz | psachin: I did not spent time in implementation details, as my first goal was to get feedback on the functionality. As I have stated, I could imagine an optional param like --auth-tempurl taking an arg like https://cluster/account/container/prefix?temp_url_prefix=...&temp_url_sig=.... . But these details are most likely be rediscussed in the gerrit. | 13:01 |
| *** baojg has joined #openstack-swift | 13:03 | |
| *** fcantournet has joined #openstack-swift | 13:04 | |
| *** mat128 has quit IRC | 13:13 | |
| *** mat128 has joined #openstack-swift | 13:15 | |
| *** fcantournet has quit IRC | 13:16 | |
| *** hseipp has quit IRC | 13:16 | |
| *** fcantournet has joined #openstack-swift | 13:16 | |
| *** silor has quit IRC | 13:19 | |
| *** silor has joined #openstack-swift | 13:19 | |
| *** links has quit IRC | 13:19 | |
| *** oshritf has quit IRC | 13:22 | |
| *** oshritf has joined #openstack-swift | 13:22 | |
| *** oshritf has quit IRC | 13:30 | |
| *** fcantournet has quit IRC | 13:30 | |
| *** fcantournet has joined #openstack-swift | 13:31 | |
| *** oshritf has joined #openstack-swift | 13:32 | |
| kota_ | good morning | 13:41 |
| mattoliverau | Morning | 13:43 |
| *** fcantournet has quit IRC | 13:44 | |
| *** fcantournet has joined #openstack-swift | 13:44 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: DON'T MERGE: Add debug message to middlewares https://review.openstack.org/503003 | 13:55 |
| kota_ | mattoliverau: o/ | 13:56 |
| *** fcantournet has quit IRC | 13:58 | |
| *** fcantournet has joined #openstack-swift | 13:58 | |
| mattoliverau | kota_: o/ | 13:58 |
| mattoliverau | Christian, Thiago and I are at breakfast downstairs, there a a few spare seats in case others wanted to join :) | 14:05 |
| *** fcantournet has quit IRC | 14:12 | |
| *** fcantournet has joined #openstack-swift | 14:12 | |
| *** fcantournet has quit IRC | 14:25 | |
| *** fcantournet has joined #openstack-swift | 14:26 | |
| *** SkyRocknRoll has quit IRC | 14:30 | |
| *** fcantournet has quit IRC | 14:32 | |
| clayg | zaitcev: I think we should definitely do a hangout or something for the proxy PUT+POST work! torgomatic - when should we do that? | 14:33 |
| clayg | notmyname: can we get a projector!? I want like GIANT pete! | 14:33 |
| *** gyee has joined #openstack-swift | 14:34 | |
| *** vint_bra has joined #openstack-swift | 14:34 | |
| kota_ | clayg: https://ethercalc.openstack.org/Queens-PTG-Discussion-Rooms | 14:52 |
| kota_ | clayg: it's a reservation page for reservable projection room | 14:53 |
| kota_ | Vail and Ballroom C looks to equip a projector? (according to the schedule paper) | 14:54 |
| kota_ | it looks like Ballroom C is available for this afternoon. | 14:54 |
| *** hseipp has joined #openstack-swift | 14:55 | |
| *** tovin07 has joined #openstack-swift | 14:57 | |
| *** cbartz has quit IRC | 14:59 | |
| kota_ | hello tovin07, are you in Denver PTG? | 15:01 |
| *** lucasxu has joined #openstack-swift | 15:01 | |
| tovin07 | hi kota_, unfortunately, i’m not in Denver now | 15:01 |
| kota_ | tovin07: alright, no worries. I just saw you have logged into this channel in the Denver morning time ;-) | 15:02 |
| tovin07 | oh, happy to know that :D | 15:02 |
| tovin07 | I saw that you put OSprofiler as a topic for this PTG | 15:03 |
| tovin07 | sadly, I do not have a chance to go to Denver for this :D | 15:04 |
| *** pcaruana has quit IRC | 15:20 | |
| kota_ | tovin07: if you have progress or opinion for that topic, please add your comment to the etherpad | 15:20 |
| *** zhurong has joined #openstack-swift | 15:20 | |
| *** cschwede has joined #openstack-swift | 15:21 | |
| *** ChanServ sets mode: +v cschwede | 15:21 | |
| *** chlong has joined #openstack-swift | 15:31 | |
| notmyname | good morning | 15:33 |
| acoles | good morning. some of us are in Big Thompson room | 15:33 |
| *** esnyder has joined #openstack-swift | 15:33 | |
| kota_ | good morning acoles and notmyname | 15:34 |
| kota_ | and thanks that info, I'm going to move there (Big Thompson room) | 15:34 |
| notmyname | like yesterday, we're in Big Thompson again for bug triage | 15:34 |
| tdasilva | clayg: i think this is a dupe of that: https://bugs.launchpad.net/swift/+bug/1365330 | 15:52 |
| openstack | Launchpad bug 1365330 in OpenStack Object Storage (swift) "cyclic replication in one cluster" [Undecided,New] | 15:52 |
| clayg | tdasilva: f87a5487 | 15:52 |
| tdasilva | clayg: thanks | 15:53 |
| clayg | tdasilva: 100% agree - dupe | 15:55 |
| *** psachin has quit IRC | 15:55 | |
| *** itlinux has joined #openstack-swift | 15:58 | |
| *** geaaru has quit IRC | 16:01 | |
| *** rcernin has quit IRC | 16:02 | |
| *** joeljwright has quit IRC | 16:03 | |
| zaitcev | clayg, notmyname: I'll need to know a time in advance. In particular, Thursday morning is probably no-go, as I'll have a signing at the time. | 16:03 |
| *** cschwede has quit IRC | 16:05 | |
| *** zhurong has quit IRC | 16:06 | |
| *** hseipp has quit IRC | 16:12 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:13 | |
| *** hseipp has joined #openstack-swift | 16:14 | |
| tdasilva | zaitcev: we were talking about having the meeting next week, that way it becomes a bit more even where everyone is on the video call as opposed to having just you in front of a camera and us in a conference room having side conversations that you can't follow | 16:16 |
| tdasilva | zaitcev: maybe we can schedule a meeting and send out a bluejeans invite | 16:17 |
| *** zhurong has joined #openstack-swift | 16:18 | |
| zaitcev | tdasilva: notmyname replied with that suggestion. | 16:19 |
| *** SkyRocknRoll has quit IRC | 16:20 | |
| *** skudlik has quit IRC | 16:20 | |
| *** oshritf has quit IRC | 16:25 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:33 | |
| notmyname | tdasilva: I had to go back to my hotel room for a bit, but I left my bag in the room | 16:37 |
| notmyname | and power cord | 16:37 |
| *** chsc has joined #openstack-swift | 16:37 | |
| tdasilva | notmyname: yep | 16:37 |
| notmyname | tdasilva: I should be back in 20 minutes, but can you keep an eye on it if people leave? | 16:37 |
| tdasilva | notmyname: yep, np | 16:37 |
| notmyname | thanks | 16:38 |
| *** SkyRocknRoll has quit IRC | 16:38 | |
| *** tovin07 has left #openstack-swift | 16:38 | |
| *** rcernin has joined #openstack-swift | 16:52 | |
| timburke | kota_: mind taking a look at https://review.openstack.org/#/c/411557/ ? i'm thinking i might be able to take a similar approach in https://review.openstack.org/#/c/368264/ | 16:53 |
| patchbot | patch 411557 - swift3 - Differentiate between a missing bucket and a missi... | 16:53 |
| patchbot | patch 368264 - swift3 - Fix the deletion of non-existent keys | 16:53 |
| *** lucasxu has quit IRC | 16:53 | |
| *** klrmn has joined #openstack-swift | 16:54 | |
| kota_ | timburke: will do, thx | 16:54 |
| *** SkyRocknRoll has joined #openstack-swift | 16:55 | |
| *** mvk has quit IRC | 16:58 | |
| *** tesseract has quit IRC | 17:07 | |
| *** skudlik has joined #openstack-swift | 17:12 | |
| *** skudlik has quit IRC | 17:15 | |
| *** honga has quit IRC | 17:17 | |
| *** oshritf has joined #openstack-swift | 17:28 | |
| *** oshritf has quit IRC | 17:34 | |
| *** mvk has joined #openstack-swift | 17:34 | |
| *** oshritf has joined #openstack-swift | 17:40 | |
| *** zhurong has quit IRC | 17:42 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Change symlink behavior with object_post_as_copy = True https://review.openstack.org/503003 | 17:56 |
| *** oshritf has quit IRC | 17:58 | |
| *** hseipp has quit IRC | 18:00 | |
| *** rcernin has quit IRC | 18:07 | |
| *** rmcall has joined #openstack-swift | 18:44 | |
| *** chsc has quit IRC | 18:44 | |
| *** SkyRocknRoll has quit IRC | 19:00 | |
| *** silor has quit IRC | 19:15 | |
| *** hseipp has joined #openstack-swift | 19:26 | |
| *** joeljwright has joined #openstack-swift | 19:28 | |
| *** ChanServ sets mode: +v joeljwright | 19:28 | |
| kota_ | timburke: done on patch 411557 | 19:34 |
| patchbot | https://review.openstack.org/#/c/411557/ - swift3 - Differentiate between a missing bucket and a missi... | 19:34 |
| *** itlinux has quit IRC | 19:43 | |
| *** zhurong has joined #openstack-swift | 19:51 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Fix converging unbalance state at ring builder rebalance https://review.openstack.org/479223 | 19:53 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Accept a trade off of dispersion for balance https://review.openstack.org/503152 | 19:53 |
| *** itlinux has joined #openstack-swift | 19:53 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Accept a trade off of dispersion for balance https://review.openstack.org/503152 | 19:54 |
| *** chsc has joined #openstack-swift | 20:04 | |
| *** chsc has joined #openstack-swift | 20:04 | |
| clayg | kota_: timburke: I spent some time looking at patch 479223 and I think I *almost* understand why the existing ring code had to work the way it did. I _definitely_ know why it can't _keep_ working the way it did without some relaxation of the constraints which lead to lp bug #1699636 and lp bug #1701472 | 20:16 |
| openstack | Launchpad bug 1699636 in OpenStack Object Storage (swift) "some rings won't rebalance" [Undecided,New] https://launchpad.net/bugs/1699636 | 20:16 |
| patchbot | https://review.openstack.org/#/c/479223/ - swift - Fix converging unbalance state at ring builder reb... | 20:16 |
| openstack | Launchpad bug 1701472 in OpenStack Object Storage (swift) "ring builder can converge dead state in rebalance" [Medium,Confirmed] https://launchpad.net/bugs/1701472 | 20:16 |
| *** kallenp has joined #openstack-swift | 20:22 | |
| kallenp | Hello. Please, how can I change account.builder, container.builder and object.builder ? I change IP address of my server. Thanks for a help. | 20:24 |
| clayg | kallenp: there's some syntax for updating node's ip's | 20:24 |
| kallenp | clayg: plese, can you send me the right command ? | 20:26 |
| clayg | `set_info` | 20:26 |
| clayg | I think `swift-ring-builder test.builder set_info` will output some help `swift-ring-builder test.builder set_info --ip 127.0.0.1 --change-ip 128.0.0.1` did a thing for me | 20:28 |
| kallenp | swift-ring-builder account.builder set_info --change-ip 192.168.224.60 | 20:29 |
| kallenp | Ring Builder file is invalid: account.builder | 20:29 |
| kallenp | any idea ? | 20:30 |
| clayg | maybe that builder file is not valid? does it seem happy if you just do a "display" `swift-ring-builder account.builder` | 20:30 |
| kota_ | awesome clayg great! | 20:32 |
| kallenp | I recover .builder files | 20:33 |
| kallenp | now: | 20:34 |
| kallenp | swift-ring-builder account.builder set_info --change-ip 192.168.224.60 | 20:34 |
| kallenp | Invalid <search-value>: '--change-ip' | 20:34 |
| clayg | kallenp: yeah you need to specify which ip you want to change with the --ip argument? | 20:34 |
| clayg | kota_: idk, it's not great - I don't think my change is working 100% either :'( | 20:34 |
| kallenp | clayg: please, what's the right command ? old IP: 192.168.224.160 new IP: 192.168.224.123 | 20:35 |
| clayg | kota_: but there's this strange side effect that if we just don't do gather_parts_for_balance_can_disperse at all we end up with bad intermediate states on the way to bad dispersion with good balance | 20:35 |
| kallenp | And verry sorry, I am only web developer and this is only my private test environment | 20:36 |
| clayg | kallenp: `swift-ring-builder account.builder set_info --ip 192.168.224.160 --change-ip 192.168.224.123` | 20:36 |
| clayg | kallenp: you're doing great, these commands are not terribly ergonomic | 20:36 |
| *** itlinux has quit IRC | 20:37 | |
| kota_ | clayg: i see, I want to follow the change (sorry not yet look at it) | 20:38 |
| kallenp | clayg: man, you're The GENIUS ! Thanks a lot ! | 20:39 |
| kallenp | I also change hostname from storage to devel-storage Is needed any other changes ? | 20:39 |
| clayg | i don't think the rings care much about hostname - that's probably fine | 20:40 |
| kallenp | ok | 20:40 |
| kallenp | this change i permanet or is needed any other commit ? | 20:41 |
| *** rmcall has quit IRC | 20:41 | |
| clayg | so it's a permanet change to the *builder* - you'll want to dump a .ring.gz and distribute that to the nodes in the cluster | 20:41 |
| clayg | easiest way to get a .ring.gz is `swift-ring-builder account.builder rebalance` | 20:42 |
| clayg | you might have to `-f` if you haven't made any other changes - or just `write_ring` if you're sure you have not | 20:42 |
| clayg | ... don't forget to make fixups in the container.builder and object.builder if you need to do that... | 20:42 |
| kallenp | I have only simple and single storage node | 20:43 |
| kallenp | whats the best way ? | 20:43 |
| clayg | `swift-ring-builder account.builder rebalance -f` should be great; then cp account.ring.gz to /etc/swift on your storage nodes | 20:43 |
| clayg | er.. the single storage node ;) | 20:44 |
| *** itlinux has joined #openstack-swift | 20:45 | |
| *** joeljwright has quit IRC | 20:54 | |
| *** alenavolk has joined #openstack-swift | 20:56 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Use swift3's check_signature function https://review.openstack.org/441520 | 20:59 |
| *** kallenp has left #openstack-swift | 21:00 | |
| *** mat128 has quit IRC | 21:03 | |
| timburke | bah, doubled-up spaces | 21:06 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Use swift3's check_signature function https://review.openstack.org/441520 | 21:06 |
| *** alenavolk has quit IRC | 21:07 | |
| *** alenavolk has joined #openstack-swift | 21:11 | |
| *** gkadam has joined #openstack-swift | 21:30 | |
| *** zhurong has quit IRC | 21:35 | |
| *** itlinux has quit IRC | 21:38 | |
| *** itlinux has joined #openstack-swift | 21:39 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Functional tests for staticweb middleware https://review.openstack.org/503262 | 21:45 |
| *** gkadam has quit IRC | 21:49 | |
| *** catintheroof has quit IRC | 21:50 | |
| timburke | clayg: got opinions on https://bugs.launchpad.net/swift/+bug/1646320 ? i was going to mark it incomplete pending us scoping out the work -- but that doesn't quite seem to be what that status is for. i kinda just wanna say Won't Fix? | 21:50 |
| openstack | Launchpad bug 1646320 in OpenStack Object Storage (swift) "account and container database module common.db should use documented sqlite3 error codes instead of strings" [Wishlist,New] | 21:50 |
| *** gkadam has joined #openstack-swift | 21:51 | |
| clayg | you know I think we could close it as wont-fix - hopefully just having the record will remind us why it's so hard to get reasonable taxonomy of sqlite errors using error codes into our codebase? | 21:53 |
| timburke | i'm reminded of torgomatic wanting to have some status for "that's unfortunate" | 21:54 |
| clayg | lol | 21:54 |
| clayg | timburke: OTOH, the upstream bug report does seem to be making some progress | 21:55 |
| clayg | if our py3 efforts eventually allow us to use the sqlite error codes in the case where we're on >py3.7 (and gracefully degrade) - maybe the status is just "deferred" | 21:55 |
| *** redbo has joined #openstack-swift | 21:56 | |
| *** ChanServ sets mode: +v redbo | 21:56 | |
| timburke | clayg: but still -- realistically, when do you see us dropping py27 support? is it really better to have two ways of trying to detect whether to quarantine? that seems to have made our lives harder -- "this DB gets quarantined on my py3 nodes but not py2!" | 21:57 |
| clayg | yeah... idk... it's definitely not a priority? We don't have a good plan to address the deficiency - wontfix seems as appropriate as anything? | 21:58 |
| *** itlinux has quit IRC | 22:00 | |
| *** alenavolk has quit IRC | 22:03 | |
| tdasilva | clayg, timburke, torgomatic: thinking about closing this bug as 'Won't fix': https://bugs.launchpad.net/swift/+bug/1517673 as it doesn't seem to be an issue in Swift itself, thoughts? | 22:16 |
| openstack | Launchpad bug 1517673 in OpenStack Object Storage (swift) "WSGI doesn't support Absolute URL (HTTP/1 compat)" [Undecided,New] | 22:16 |
| clayg | yeah, I mean it's an unfortunate behavior - but the "bug" is in eventlet and eventlet isn't even really at fault because wsgiref has the same bug! | 22:22 |
| clayg | there's some non-zero chance eventlet managed to fix it accidently along the way | 22:22 |
| clayg | did you check if it's still a bug/issue? | 22:22 |
| clayg | i think there's some socket code in one of the comments that would do the trick... | 22:22 |
| timburke | seems reasonable -- even if it *were* something we could fix, i don't know that it'd ever get done. contrast that with something like a socket leak or the rlock issue, where we may actually care enough to push for that patch in eventlet | 22:24 |
| timburke | won't fix -- not in scope for us | 22:25 |
| *** alenavolk has joined #openstack-swift | 22:29 | |
| *** alenavolk has quit IRC | 22:37 | |
| *** catintheroof has joined #openstack-swift | 22:38 | |
| kota_ | timburke, clayg: I'm looking at https://bugs.launchpad.net/swift/+bug/1554006 that reports replicator dumps rsync error when handoff node tried to the primary which has failure device (i.e. unmounted) | 22:41 |
| openstack | Launchpad bug 1554006 in OpenStack Object Storage (swift) "Replicator attempts to rsync data to an unmounted device " [Undecided,New] | 22:41 |
| *** joeljwright has joined #openstack-swift | 22:42 | |
| *** ChanServ sets mode: +v joeljwright | 22:42 | |
| kota_ | I'm feeling that it's in design to call rsync first w/o REPLICATE request in hadoff node. | 22:42 |
| clayg | kota_: I agree replicator definitely does call rsync w/o calling REPLICATE | 22:43 |
| kota_ | clayg: thanks for the answer! | 22:43 |
| clayg | except I think maybe when the remote device is unmounted this behavior is maybe not ideal? | 22:43 |
| clayg | meh, idk, i mean rsync throws an error - so it's probably ok? | 22:44 |
| clayg | we can won't fix that | 22:45 |
| *** sundbp has quit IRC | 22:59 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Disallow fractional replicas in EC policies https://review.openstack.org/503454 | 22:59 |
| *** hseipp has quit IRC | 22:59 | |
| *** vint_bra has quit IRC | 23:00 | |
| *** joeljwright has quit IRC | 23:03 | |
| timburke | torgomatic: remind me: how much of https://github.com/openstack/swift/blob/master/swift/common/utils.py#L86-L91 is still necessary following https://github.com/openstack/swift/commit/6d16079 ? | 23:08 |
| timburke | it makes me rather nervous that we swap out logging._lock with a eventlet.green.threading.RLock... | 23:09 |
| *** jistr has quit IRC | 23:29 | |
| *** kei_yama has joined #openstack-swift | 23:29 | |
| *** jistr has joined #openstack-swift | 23:31 | |
| *** kong has joined #openstack-swift | 23:33 | |
| *** chsc has quit IRC | 23:38 | |
| *** gkadam has quit IRC | 23:44 | |
| *** ianychoi_ has joined #openstack-swift | 23:47 | |
| *** ujjain- has joined #openstack-swift | 23:48 | |
| *** nikivi has quit IRC | 23:48 | |
| *** ianychoi has quit IRC | 23:48 | |
| *** ujjain has quit IRC | 23:48 | |
| *** nikivi has joined #openstack-swift | 23:49 | |
| *** portdirect has quit IRC | 23:49 | |
| *** torgomatic has quit IRC | 23:51 | |
| *** torgomatic has joined #openstack-swift | 23:51 | |
| *** ChanServ sets mode: +v torgomatic | 23:51 | |
| *** portdirect has joined #openstack-swift | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!