| kota_ | good morning | 00:02 |
|---|---|---|
| *** sams-gleb has joined #openstack-swift | 00:06 | |
| mattoliverau | kota_: morning | 00:10 |
| kota_ | morning mattoliverau | 00:11 |
| kota_ | acoles: thanks for updating the patches | 00:21 |
| kota_ | acoles: looking at the irc log, we should consider what could be a problem with uuid case. i didn't think the case if someone could copy/paste the builder file??? | 00:22 |
| kota_ | and it could have same uuid :/ | 00:23 |
| kota_ | it could be just a mitagation but some commands like "reset" could be useful? | 00:25 |
| kota_ | i don't think so much because what he should do is just creating brand-new ring builder file for the case. | 00:26 |
| kota_ | the ring.gz file is usually copy/paste to deploy into the cluster env, but copy/pasting *builder* file sounds crazy... | 00:27 |
| *** tonanhngo has quit IRC | 00:35 | |
| *** catintheroof has joined #openstack-swift | 00:35 | |
| *** gkadam has quit IRC | 00:40 | |
| *** tonanhngo has joined #openstack-swift | 00:45 | |
| *** gyee has quit IRC | 00:47 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: fixup! Add container sharding to Swift containers https://review.openstack.org/450491 | 00:49 |
| *** tonanhngo has quit IRC | 00:50 | |
| *** jamielennox is now known as jamielennox|away | 00:54 | |
| timburke | mattoliverau: to warn you, i'm out later this week and all of next week, but i'll get as far as i can on sharding before i go | 00:56 |
| mattoliverau | timburke: kk, you've already given me alot to work on :) and thanks. | 00:58 |
| *** jamielennox|away is now known as jamielennox | 00:59 | |
| openstackgerrit | junboli proposed openstack/swift master: Use swift tempurl instaed of swift-temp-url https://review.openstack.org/450494 | 01:02 |
| *** vint_bra has joined #openstack-swift | 01:20 | |
| tone_z | Good morning! | 01:20 |
| *** timburke has quit IRC | 01:25 | |
| *** AndyWojo has quit IRC | 01:25 | |
| *** hugokuo has quit IRC | 01:25 | |
| *** vint_bra has quit IRC | 01:26 | |
| *** vint_bra has joined #openstack-swift | 01:26 | |
| mattoliverau | tone_z: morning! | 01:27 |
| *** AndyWojo has joined #openstack-swift | 01:27 | |
| *** timburke has joined #openstack-swift | 01:28 | |
| *** ChanServ sets mode: +v timburke | 01:28 | |
| tone_z | mattoliverau: morning! | 01:28 |
| *** hugokuo has joined #openstack-swift | 01:28 | |
| *** sams-gleb has quit IRC | 01:40 | |
| *** catintheroof has quit IRC | 01:41 | |
| *** JimCheung has quit IRC | 01:41 | |
| *** zhurong has joined #openstack-swift | 01:53 | |
| *** vint_bra has quit IRC | 02:14 | |
| kota_ | i wonder if someone knows func-encryption gate could be broken? | 02:21 |
| kota_ | the patch acoles updated failed at http://logs.openstack.org/21/441921/6/check/gate-swift-tox-xfs-tmp-func-encryption-ubuntu-xenial/857bfd6/console.html but it should not be related to the func tests | 02:22 |
| kota_ | that is because it touches only offline ring tools | 02:23 |
| *** sams-gleb has joined #openstack-swift | 02:38 | |
| *** sams-gleb has quit IRC | 02:42 | |
| openstackgerrit | junboli proposed openstack/swift master: Fix subclass call super class's constructor https://review.openstack.org/450523 | 02:45 |
| *** _JZ_ has quit IRC | 03:01 | |
| *** links has joined #openstack-swift | 03:35 | |
| zaitcev | Guys. We have someone running Swift in Docker containers, right? I heard Project Runway does that. | 03:35 |
| zaitcev | How do they handle logging? The only thing that comes to mind is to install rsyslog in the container, than VOLUME /var/log. This has the disadvantage of not having any logrotate. Ideally I'd like to ship logs outside of the running container somehow. | 03:37 |
| *** blair has quit IRC | 03:51 | |
| *** zhurong has quit IRC | 03:57 | |
| *** blair has joined #openstack-swift | 03:57 | |
| kota_ | zaitcev: I didn't actually, but i think storlets also does simliar thing (log to /var/log in the container and then host redirect it to the rsyslogd) so eranrom may know something on that | 04:06 |
| *** jamielennox is now known as jamielennox|away | 04:09 | |
| mattoliverau | zaitcev: runway might go to docker, but atm (unless its changed) is't LXC. You can do bind mounting in LXC (or the equivelnt) which means you can place the logs outside the container if need be. | 04:18 |
| mattoliverau | but yeah, you probably need a syslog | 04:18 |
| zaitcev | well, bind works I guess | 04:19 |
| zaitcev | The host has its own Swift running, I'm concerned the logs will intermingle | 04:19 |
| kota_ | mattoliverau: i think it's better than /var/log thing | 04:19 |
| kota_ | wow, both inside/outside swift is running!? | 04:19 |
| mattoliverau | zaitcev: if thats the case then you might want to add a rule to syslog to change the naming slightly. | 04:20 |
| kota_ | interesting | 04:20 |
| mattoliverau | you can talk to syslog as a dev, so you might be able to push through the dev, and then talk to it directly. not that that is supported in Swift.. but could be an option. | 04:20 |
| zaitcev | right, I meant to tinker with it, just not sure of success | 04:21 |
| mattoliverau | then you won't need an addiontal syslog, though you'd need to do some magic on the hosts to put things in the right place. | 04:21 |
| *** psachin has joined #openstack-swift | 04:22 | |
| *** klrmn has quit IRC | 04:27 | |
| *** jamielennox|away is now known as jamielennox | 04:30 | |
| *** MarkMielke has quit IRC | 04:35 | |
| *** sams-gleb has joined #openstack-swift | 04:39 | |
| *** sams-gleb has quit IRC | 04:44 | |
| *** zhurong has joined #openstack-swift | 04:55 | |
| mahatic | good morning | 04:57 |
| mahatic | notmyname: after the submissions over here - http://forumtopics.openstack.org/, is there anything the community can do, like voting? I don't think so, I believe the TC and co will decide what's gonna be at the forum? | 04:58 |
| kota_ | mahatic: morning, i don't think we can vote it | 05:04 |
| kota_ | mahatic: the format looks like similar with the older design summit session proposal (maybe G or H, IIRC) and then... | 05:05 |
| kota_ | mahatic: ah, it might be able to *review* it to give your feedback | 05:06 |
| kota_ | I can see the box to write something, e.g. http://forumtopics.openstack.org/cfp/details/33 | 05:07 |
| mahatic | kota_: oic | 05:07 |
| kota_ | the intersting point to me is we have only 40 proposal in total | 05:07 |
| mahatic | kota_: but not sure if they'd consider those "comments" right | 05:07 |
| kota_ | the last item on top of the page shows, http://forumtopics.openstack.org/cfp/details/40 | 05:07 |
| kota_ | mahatic: yeah, need to ask | 05:08 |
| kota_ | 40 proposals, swift takes 8 of 40 -->>> 20% of openstack is swift :P | 05:09 |
| mahatic | :D yes! | 05:09 |
| kota_ | idk it works as expected by tc | 05:09 |
| mattoliverau | mahatic: morning | 05:10 |
| mahatic | mattoliverau: o/ | 05:10 |
| mahatic | kota_: re 20%, so we still have opportunity to up the number of proposals for swift then ;) | 05:11 |
| kota_ | mahatic: exactly ;-) | 05:11 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient master: Updated from global requirements https://review.openstack.org/89250 | 05:17 |
| *** rcernin has joined #openstack-swift | 05:38 | |
| *** sams-gleb has joined #openstack-swift | 05:42 | |
| *** sams-gleb has quit IRC | 05:48 | |
| *** zaitcev has quit IRC | 05:50 | |
| openstackgerrit | junboli proposed openstack/swift master: update doc infos to ocata https://review.openstack.org/450571 | 05:53 |
| *** ChubYann has quit IRC | 05:54 | |
| *** bkopilov_ has quit IRC | 06:19 | |
| openstackgerrit | Merged openstack/swift master: Remove unused returned value object_path from yield_hashes() https://review.openstack.org/450269 | 06:19 |
| *** bkopilov has quit IRC | 06:20 | |
| *** silor has joined #openstack-swift | 06:26 | |
| *** silor1 has joined #openstack-swift | 06:31 | |
| *** silor has quit IRC | 06:33 | |
| *** silor1 is now known as silor | 06:33 | |
| *** jamielennox is now known as jamielennox|away | 06:36 | |
| *** hseipp has joined #openstack-swift | 06:45 | |
| *** sams-gleb has joined #openstack-swift | 06:49 | |
| *** mtreinish has quit IRC | 06:57 | |
| *** mtreinish has joined #openstack-swift | 06:58 | |
| *** admin6 has joined #openstack-swift | 07:12 | |
| alecuyer | good morning | 07:13 |
| *** pcaruana has joined #openstack-swift | 07:13 | |
| *** sams-gleb has quit IRC | 07:17 | |
| *** tesseract has joined #openstack-swift | 07:18 | |
| *** cbartz has joined #openstack-swift | 07:20 | |
| *** jamielennox|away is now known as jamielennox | 07:21 | |
| *** jaosorior has joined #openstack-swift | 07:29 | |
| *** geaaru has joined #openstack-swift | 07:29 | |
| *** rcernin has quit IRC | 07:29 | |
| *** rcernin has joined #openstack-swift | 07:47 | |
| *** amoralej|off is now known as amoralej | 07:53 | |
| *** furlongm has joined #openstack-swift | 07:57 | |
| *** bkopilov_ has joined #openstack-swift | 08:00 | |
| *** bkopilov has joined #openstack-swift | 08:01 | |
| acoles | good morning | 08:04 |
| *** rcernin has quit IRC | 08:06 | |
| *** rcernin has joined #openstack-swift | 08:07 | |
| *** hseipp has quit IRC | 08:09 | |
| *** rcernin has quit IRC | 08:10 | |
| *** rcernin has joined #openstack-swift | 08:10 | |
| acoles | mahatic: in case you didn't notice, most of the composite ring patch you were reviewing has been squashed into patch 441921 | 08:12 |
| patchbot | https://review.openstack.org/#/c/441921/ - swift - Add Composite Ring Functionality | 08:12 |
| kota_ | acoles: morning | 08:12 |
| kota_ | and the composite ring one is under reviewing to me | 08:12 |
| kota_ | acoles: just notification on that | 08:12 |
| acoles | mahatic: sorry I should have left a note on patch 449310 to say that | 08:12 |
| patchbot | https://review.openstack.org/#/c/449310/ - swift - WIP Use uuid to differentiate component rings in c... | 08:12 |
| kota_ | i mean patch 441921 | 08:13 |
| patchbot | https://review.openstack.org/#/c/441921/ - swift - Add Composite Ring Functionality | 08:13 |
| mahatic | acoles: good morning | 08:13 |
| acoles | kota_: hi! I hope I have not confused too much with squashing into 441921 | 08:13 |
| * kota_ is still thinking about "check_against_existing" | 08:14 | |
| acoles | kota_: you're right - crazy to copy builder files, but our job is to plan for "crazy" sometimes :) the bigger concern is lost builder file and using write-builder from a ring.gz | 08:15 |
| kota_ | acoles: ok, before thinking too much, can i leave my draft comment to ask you? | 08:15 |
| acoles | but I left the uuid ideas in the follow on patch | 08:15 |
| kota_ | yup | 08:15 |
| acoles | kota_: of course | 08:15 |
| mahatic | acoles: I did, I glanced to see the diff. I no longer see the "tuple" part as well, thanks. | 08:15 |
| kota_ | k, do it now. | 08:15 |
| acoles | mahatic: yep I fixed the parentheses you spotted | 08:15 |
| kota_ | acoles: I didn't run all code yet but i concerned at https://review.openstack.org/#/c/441921/6/swift/common/ring/composite_builder.py@214 mostly. | 08:17 |
| *** mvk has quit IRC | 08:17 | |
| acoles | kota_: that test failure is concerning...but it passed on recheck I think. hmmm | 08:19 |
| acoles | kota_: what was your concern here https://review.openstack.org/#/c/441921/6/swift/common/ring/builder.py@1673 ? | 08:21 |
| *** kirill_ has joined #openstack-swift | 08:21 | |
| kota_ | acoles: on the func test perspective, i don't think it's related to the change because the existing swift functional tests never calls the new functionality (compose_rings, whatever) right? | 08:21 |
| acoles | is there a way to easily get a link to a specific comment in gerrit? | 08:21 |
| acoles | kota_: agree re func test, just worrying that it failed in general, but can't see how it relates to the patch | 08:22 |
| kota_ | acoles: ah - sorry, https://review.openstack.org/#/c/441921/6/swift/common/ring/builder.py@1673 <- i should remove the comment, just I was starting to review from there and I was made sense already we will need it in composite_builder.py | 08:23 |
| acoles | kota_: ok. TBH IDK if me calling it "source" was such a great thing. It's ok. I am just frustrated by not having a *perfect* way to track builder "identity" :/ | 08:25 |
| acoles | kota_: re. your comments at https://review.openstack.org/#/c/441921/6/swift/common/ring/composite_builder.py@214 ... | 08:25 |
| *** oshritf has joined #openstack-swift | 08:27 | |
| kirill_ | hello everyone! is this an appropriate place to ask a question about swift3? | 08:28 |
| kota_ | kirill_: it's ok here but #swift3 channel also works and it's for only swift3 rather than general openstack-swift | 08:30 |
| acoles | kota_: for now I deliberately kept this very simple - you cannot add or remove rings, only compose rings and re-compose the same rings, for example after a component rebalance. test_write_composite_ring_errors asserts that new builders must be same length as old. Once we understand other use cases (add, remove) we can and support but for no I was hoping to keep it simple. | 08:31 |
| kota_ | acoles: ok, I like the opinion, "keep it simple" | 08:33 |
| kota_ | acoles: that means... oh, I misread one thing | 08:34 |
| kota_ | acoles: "if the new component shorter than old component, that's ok i.e. we can remove a (or some) builder(s) in the new_component" <- this is false | 08:34 |
| kota_ | i skipped to look at the errors.append() at L213 so if the length is different, anyway, we will get a ValueError | 08:35 |
| kota_ | s/skipped/missed/ | 08:36 |
| acoles | kota_: yes, if new is shorter then there is an IndexError which is converted to an error | 08:36 |
| acoles | kota_: I have a question for you... | 08:37 |
| kota_ | and current code looks still simple. if you want a new ring from different component, please you make it as different ring name. | 08:37 |
| kota_ | acoles: sure | 08:37 |
| acoles | kota_: yes, for now if you want to add/remove components then start again with a new composite.ring.gz | 08:38 |
| acoles | kota_: ok, my question is about the source=builder_file | 08:38 |
| kota_ | acoles: ok, please go ahead | 08:39 |
| acoles | kota_: builder_file can be relative path or an absolute path, so long as open() can find it. But. the suggestion on etherpad is that we *require* component ring builders to be in same dir as the composite-ring.gz, (again to keep things simple and not too brittle if paths change) | 08:40 |
| acoles | kota_: I don't want to make the builder.source be forced to have only the file name | 08:41 |
| kota_ | acoles: let me check the etherpad, I didn't remember to write the requirement that we should locate it in a dir | 08:42 |
| acoles | i.e. basename - feels like that should not be something enforced in builder.py | 08:42 |
| acoles | kota_: I think it was mattoliverau suggestion | 08:42 |
| kota_ | acoles: i found at L100 | 08:44 |
| acoles | kota_: yes line 100, | 08:44 |
| acoles | kota_: so I am wondering if it is ok for composite_builder.py to be as it is now in the patch i.e. simply check that builder.source values are the same for each component, and then make whatever calls into composite_builder be responsible for what the value of builder.source is? | 08:45 |
| *** rcernin has quit IRC | 08:46 | |
| acoles | so if the sources should all be basenames, then the caller (a CLI for example) should ensure that the builders are loaded that way using only basename for builder_file | 08:46 |
| *** rcernin has joined #openstack-swift | 08:46 | |
| *** mvk has joined #openstack-swift | 08:47 | |
| acoles | but if a caller wants to use some other way to track builders (e.g. relative path) then the caller can set builder.source appropriately | 08:47 |
| kota_ | and the functionality in composite_builder doesn't care of it? | 08:47 |
| acoles | composite_builder just checks that the builder.source has not changed for each component | 08:48 |
| *** rcernin_ has joined #openstack-swift | 08:48 | |
| acoles | IDK, I am trying to avoid putting something very rigid into composite_builder | 08:48 |
| acoles | at least, for now, when we are still figuring out exact use case | 08:49 |
| acoles | for a CLI for example | 08:49 |
| kota_ | acoles: I'm feeling the identifier(s) is (are) cared by CLI as possible could be better and then composite_builder will trust the identifiers, just checking the sameness | 08:51 |
| kota_ | so... | 08:51 |
| kota_ | that was why I removed the source file path in the patch, previously | 08:52 |
| acoles | kota_: ok. so let's leave the patch as it is but think some more about that issue. I am curious what we may learn about how others manage builders, ring rebalancing etc, other than using swift-ring-builder. | 08:52 |
| acoles | kota_: ah, which version - I looked back to version 3 for your previous use of builder file | 08:53 |
| kota_ | acoles: i think similar issue still exists in swift ring builder | 08:53 |
| kota_ | acoles: that is because the dev check (existing ip/port/dev check) is in cli tool | 08:54 |
| kota_ | swift.common.ring.RingBuilder.add_dev doesn't care the same failure domain | 08:54 |
| kota_ | set the devs directly could be dengerous, doesn't it? | 08:55 |
| kota_ | I hope validate() method work to check before finishing up the rebalance | 08:57 |
| kota_ | though | 08:57 |
| kota_ | acoles: re: which version, I remove the builder_file in patch set 4 | 08:59 |
| kota_ | removed | 08:59 |
| admin6 | acoles: kota: Hi guys, I’ve not been around for a while working on othe projects, but I’m now looking back (a bit) into swift. :-) | 08:59 |
| acoles | admin6: hi | 08:59 |
| acoles | kota_: where did you remove the source file path? I can't see that yet in patchset 3 | 09:00 |
| admin6 | I’d like to highlight back on bugs 1631144 and 1633647. I’m (still) using swift 2.7 and applied the different patches (387172 and 389746) provided to correct the bug. and it has greatly cleaned my corrupted fragments | 09:00 |
| openstack | bug 1631144 in OpenStack Object Storage (swift) "Swift Erasure Coding : Error when decoding fragment" [Critical,Fix released] https://launchpad.net/bugs/1631144 | 09:00 |
| openstack | bug 1633647 in OpenStack Object Storage (swift) "bad fragment data not detected in audit" [Critical,Fix released] https://launchpad.net/bugs/1633647 | 09:00 |
| kota_ | acoles: https://review.openstack.org/#/c/441921/4/swift/common/ring/builder.py this one? | 09:01 |
| kota_ | it's patch set 4 | 09:01 |
| admin6 | However, It seems that I still have some new corrupted erasure coded fragment (in a far lower proportion than during last autumn). I think they are not detected by the patched auditor. | 09:01 |
| kota_ | hi admin6 | 09:01 |
| acoles | admin6: :( | 09:02 |
| kota_ | :/ | 09:02 |
| admin6 | May I send you some examples files so you can have a look ? | 09:02 |
| openstackgerrit | liuyamin proposed openstack/python-swiftclient master: Fix some reST field lists in docstrings https://review.openstack.org/450656 | 09:04 |
| kota_ | admin6: i could look at | 09:05 |
| acoles | admin6: yes please. are these fragments that have been corrupted since you applied the patches? or corrupted frags that existed before but have still not been cleaned up? | 09:06 |
| acoles | kota_: re file path, OIC sorry I misunderstood you. | 09:08 |
| admin6 | as far as I’ve seen, it’s always ‘reconstructed’ fragments that are corrupted. I’ve applied the patch on november but I still have new corrupted fragments that are generated after the application of the patch. | 09:11 |
| acoles | admin6: ok. if you can file a bug on launchpad that would be very helpful | 09:13 |
| admin6 | acoles: Ok | 09:14 |
| acoles | thanks | 09:14 |
| acoles | kota_: another question for you ... :) should we squash 446340 into 441921? | 09:18 |
| acoles | patch 446340 patch 441921 | 09:18 |
| patchbot | https://review.openstack.org/#/c/446340/ - swift - Add ring metadata space to the RingData serialization | 09:18 |
| patchbot | https://review.openstack.org/#/c/441921/ - swift - Add Composite Ring Functionality | 09:18 |
| acoles | patchbot: it's so good to have you back | 09:18 |
| kota_ | no strong opinion from me on that, if you feel it will be easy to review, I'll do that. | 09:20 |
| kota_ | i just thought, the metadata space could be worth also current existing Ring | 09:20 |
| kota_ | e.g. add the version number to the ring.gz we can confirm the version of the ring.gz and its consistency with the builder file | 09:21 |
| kota_ | ah, but... swift-ring-builder tool already may have the capability? | 09:22 |
| kota_ | I saw some checks in the recent playing with ring-builder... | 09:22 |
| acoles | sure. I just wonder if it will merge on its own without an immediate use. the first use for it is the composite ring. and its "only" ~30 lines diff | 09:23 |
| acoles | IDK we can wait for other reviewers opinions | 09:23 |
| *** gabor_antal_ is now known as gabor_antal | 09:23 | |
| kota_ | https://github.com/openstack/swift/blob/master/swift/cli/ringbuilder.py#L500-L509 | 09:24 |
| acoles | kota_: actually I just realised that i made some small tweaks to the ring metadata in the composite ring patch so need to get those changes into the same patch one way or the other | 09:25 |
| kota_ | hmm.. currently swift-ring-builder can show the diff with the builder file and the ring.gz, although we can not check the gz version. | 09:25 |
| acoles | kota_: https://review.openstack.org/#/c/441921/6/swift/common/ring/ring.py | 09:25 |
| kota_ | looking | 09:25 |
| kota_ | ah, ok i saw it once | 09:26 |
| kota_ | hmm... I'm not sure too which way can work *farst* to us :/ | 09:29 |
| kota_ | fast | 09:29 |
| * acoles needs coffee, bbiab | 09:30 | |
| kota_ | ah... the commit message at patch 446340 looks stale, it tells about md5 which we should not use. | 09:35 |
| patchbot | https://review.openstack.org/#/c/446340/ - swift - Add ring metadata space to the RingData serialization | 09:35 |
| kota_ | anyway, I need to circle back to it | 09:35 |
| kota_ | acoles: I walked through all changes https://review.openstack.org/#/c/441921/ and then I'm realizing it's ok to squash up into the pearent patch 446340 | 09:44 |
| patchbot | patch 446340 - swift - Add ring metadata space to the RingData serialization | 09:44 |
| kota_ | ah, patch bot cannot refer the URL, just take care of "patch" phrase. | 09:44 |
| acoles | kota_: ok let's squash them. | 09:46 |
| kota_ | acoles: k, could you do it? | 09:46 |
| kota_ | or i should do? | 09:47 |
| acoles | kota_: yes no problem I will do it | 09:47 |
| kota_ | acoles: thx! | 09:47 |
| kota_ | acoles: because of your kindness, i am able to go dinner :D | 09:47 |
| acoles | hehe. have a good evening | 09:48 |
| kota_ | ;-) | 09:48 |
| *** cbartz has quit IRC | 09:50 | |
| *** csmart has quit IRC | 10:04 | |
| portante | zaitcev, we have been working on logging solutions with containers now for a while, happy share if you want to tinker | 10:15 |
| *** mvk has quit IRC | 10:16 | |
| *** csmart has joined #openstack-swift | 10:17 | |
| *** sams-gleb has joined #openstack-swift | 10:18 | |
| *** zhurong has quit IRC | 10:20 | |
| *** sams-gleb has quit IRC | 10:22 | |
| *** zhurong has joined #openstack-swift | 10:28 | |
| *** mvk has joined #openstack-swift | 10:31 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Add Composite Ring Functionality https://review.openstack.org/441921 | 10:35 |
| *** bkopilov_ has quit IRC | 10:36 | |
| *** bkopilov has quit IRC | 10:36 | |
| * mahatic is off work until the weekend. Back to work on 3rd April | 10:41 | |
| acoles | mahatic: have a good break | 10:42 |
| mahatic | acoles: thank you | 10:42 |
| openstackgerrit | XieYingYun proposed openstack/swift master: Fix some reST field lists in docstrings https://review.openstack.org/449890 | 11:11 |
| *** furlongm has quit IRC | 11:19 | |
| *** zhurong has quit IRC | 11:25 | |
| *** cbartz has joined #openstack-swift | 11:26 | |
| *** ma9_ has joined #openstack-swift | 11:27 | |
| *** cbartz has quit IRC | 11:28 | |
| *** ma9_ has quit IRC | 11:29 | |
| *** cbartz has joined #openstack-swift | 11:30 | |
| *** furlongm has joined #openstack-swift | 11:31 | |
| *** ma9_ has joined #openstack-swift | 11:32 | |
| *** openstackgerrit has quit IRC | 11:33 | |
| *** ma9_ has quit IRC | 11:37 | |
| *** openstackgerrit has joined #openstack-swift | 11:39 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 11:39 |
| *** NM has joined #openstack-swift | 11:40 | |
| *** ma9_ has joined #openstack-swift | 11:41 | |
| *** jordanP has joined #openstack-swift | 12:00 | |
| *** sams-gleb has joined #openstack-swift | 12:01 | |
| *** ma9_ has quit IRC | 12:15 | |
| *** chlong has joined #openstack-swift | 12:16 | |
| *** sams-gle_ has joined #openstack-swift | 12:39 | |
| *** ma9_ has joined #openstack-swift | 12:41 | |
| *** sams-gleb has quit IRC | 12:42 | |
| *** klamath has joined #openstack-swift | 12:45 | |
| openstackgerrit | Karen Chan proposed openstack/swift master: Store version id if restoring object from archive https://review.openstack.org/437523 | 12:53 |
| openstackgerrit | Karen Chan proposed openstack/swift master: Add X-Backend-Versioning-Mode-Override https://review.openstack.org/437196 | 12:56 |
| *** amoralej is now known as amoralej|lunch | 13:08 | |
| *** kirill_ has quit IRC | 13:18 | |
| *** kei_yama has quit IRC | 13:19 | |
| *** links has quit IRC | 13:22 | |
| *** ma9_ has quit IRC | 13:46 | |
| *** jordanP has quit IRC | 13:49 | |
| *** ma9_ has joined #openstack-swift | 13:53 | |
| *** _JZ_ has joined #openstack-swift | 13:54 | |
| *** hseipp has joined #openstack-swift | 13:58 | |
| *** amoralej|lunch is now known as amoralej | 14:03 | |
| *** ma9_ has quit IRC | 14:03 | |
| *** ma9_ has joined #openstack-swift | 14:10 | |
| *** jaosorior has quit IRC | 14:12 | |
| *** vint_bra has joined #openstack-swift | 14:44 | |
| cbartz | tdasilva: Could you please take a look at patch 423377 ? | 14:56 |
| patchbot | https://review.openstack.org/#/c/423377/ - python-swiftclient - ISO 8601 timestamps for tempurl | 14:56 |
| tdasilva | cbartz: yes, I started to look at it, just didn't have a chance to finish testing | 14:57 |
| tdasilva | but i'm happy with it overall, will try to finish up today | 14:58 |
| cbartz | tdasilva: Ok, thx. | 14:58 |
| *** ma9_1 has joined #openstack-swift | 15:02 | |
| *** ma9_ has quit IRC | 15:04 | |
| *** gabor_antal has quit IRC | 15:05 | |
| *** cbartz has left #openstack-swift | 15:14 | |
| *** oshritf has quit IRC | 15:15 | |
| *** klrmn has joined #openstack-swift | 15:15 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Test and fixups for ContainerSharder config https://review.openstack.org/450845 | 15:21 |
| *** rcernin has quit IRC | 15:31 | |
| *** rcernin_ has quit IRC | 15:31 | |
| *** tonanhngo has joined #openstack-swift | 15:51 | |
| notmyname | good morning | 16:13 |
| openstackgerrit | Gábor Antal proposed openstack/swift master: Use more specific asserts in test/unit/common https://review.openstack.org/342781 | 16:15 |
| notmyname | is gabor antal here in IRC? | 16:16 |
| *** ma9_1 has quit IRC | 16:18 | |
| *** psachin has quit IRC | 16:24 | |
| *** pcaruana has quit IRC | 16:26 | |
| *** JimCheung has joined #openstack-swift | 16:26 | |
| *** sams-gle_ has quit IRC | 16:27 | |
| *** bkopilov has joined #openstack-swift | 16:31 | |
| *** bkopilov_ has joined #openstack-swift | 16:31 | |
| *** chlong has quit IRC | 16:34 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:36 | |
| *** gabor_antal has joined #openstack-swift | 16:43 | |
| *** sams-gleb has joined #openstack-swift | 16:44 | |
| *** chlong has joined #openstack-swift | 16:49 | |
| *** hseipp has quit IRC | 16:49 | |
| *** chsc has joined #openstack-swift | 16:50 | |
| *** chsc has quit IRC | 16:50 | |
| *** chsc has joined #openstack-swift | 16:50 | |
| *** gabor_antal_ has joined #openstack-swift | 16:55 | |
| *** gabor_antal has quit IRC | 16:55 | |
| *** Renich has quit IRC | 17:03 | |
| *** geaaru has quit IRC | 17:04 | |
| *** mvk has quit IRC | 17:13 | |
| jrichli | notmyname: I see gabor_antal_ is here now | 17:14 |
| gabor_antal_ | hey there | 17:14 |
| *** gabor_antal_ is now known as gabor_antal | 17:14 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 17:15 | |
| jrichli | gabor_antal: notmyname was looking for you earlier. he might be busy now, donno. but i expect he will reply soon | 17:16 |
| notmyname | gabor_antal: hello! | 17:16 |
| notmyname | gabor_antal: I was looking at patch 342781 | 17:16 |
| patchbot | https://review.openstack.org/#/c/342781/ - swift - Use more specific asserts in test/unit/common | 17:16 |
| gabor_antal | oh, okay, i'll be here for like half an hour | 17:16 |
| notmyname | gabor_antal: before I say anything else, I want to make clear that I agree with you said in the commit message. and so far in my looking at the changes, it seems ok | 17:18 |
| notmyname | (what a terrible intro) | 17:18 |
| *** SkyRocknRoll has quit IRC | 17:18 | |
| *** amoralej is now known as amoralej|off | 17:18 | |
| gabor_antal | it is :) | 17:18 |
| notmyname | but it's a lot of code churn for something that can't be enforced (eg linting checks) | 17:19 |
| notmyname | and being in tests, it has the chance of some odd side effects (like the comment about truthy not being the same as equal | 17:20 |
| notmyname | again, the patch isn't wrong or bad itself | 17:20 |
| notmyname | but in the larger context of review and maintenance, why? | 17:20 |
| gabor_antal | Yeah I know it, but the first version when I created the patch, I did not think over that equal change. | 17:20 |
| notmyname | part of me want's the patch because it's "more" correct. but the other part of me doesn't because it's code churn fixing stuff that isn't broken | 17:22 |
| notmyname | gabor_antal: so I wanted to ask you about your thoughts on it, share my thoughts, and see where we are | 17:22 |
| notmyname | and again, I don't think the patch is wrong or bad itself, and I agree with your commit message :-) | 17:23 |
| gabor_antal | In fact, most of the rules are already merged to the main hacking project, here: https://review.openstack.org/#/c/354185/ | 17:23 |
| notmyname | oh, interesting | 17:24 |
| gabor_antal | And beforce the question, hacking is not a requirement which is automatically proposed to the projects. | 17:24 |
| gabor_antal | We faced in nova too, so i checked it and it is blacklisted to the proposal bot | 17:24 |
| gabor_antal | I can feel your point, but honestly I think this change is not useless. | 17:26 |
| gabor_antal | And beyond the fact that the code is more readable, also on error, a better error message is thrown. | 17:27 |
| gabor_antal | And I think this is the main reason for the update. I have faced several times the "False is not True." error message, but this can be changed as it can be seen in the patches | 17:28 |
| notmyname | gabor_antal: sorry. a few conversations going on at once :-) | 17:32 |
| gabor_antal | notmyname np, but i'll be leave shortly, so maybe we can continue tomorrow | 17:38 |
| notmyname | gabor_antal: I think i would like your patch a lot more if it also added/included the hacking check. however, I don't know what else that will show | 17:38 |
| gabor_antal | That's not a bad idea, I'll check it tomorrow | 17:41 |
| *** tonanhngo has quit IRC | 17:43 | |
| notmyname | gabor_antal: and thanks for your patch. I definitely appreciate it :-) | 17:44 |
| *** zaitcev has joined #openstack-swift | 17:44 | |
| *** ChanServ sets mode: +v zaitcev | 17:44 | |
| *** mvk has joined #openstack-swift | 17:52 | |
| gabor_antal | notmyname: I updated the HACKING on my local computer, and enabled the currently available extensions, and there lots of more things to do with them. But I think the above hacking patch is not in the release yet | 18:05 |
| notmyname | gabor_antal: ah, ok | 18:05 |
| gabor_antal | But I will check it tomorrow more precisely | 18:05 |
| *** tesseract has quit IRC | 18:06 | |
| gabor_antal | H203 is part of the changes, it says: "Use assertIs(Not)None to check for None (off by default) Unit test assertions tend to give better messages for more specific assertions." | 18:06 |
| *** JimCheung has quit IRC | 18:19 | |
| *** JimCheung has joined #openstack-swift | 18:20 | |
| *** JimCheung has quit IRC | 18:24 | |
| *** gyee has joined #openstack-swift | 18:26 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: fixup! Add container sharding to Swift containers https://review.openstack.org/450491 | 18:32 |
| timburke | another stable failure: http://logs.openstack.org/periodic-stable/periodic-swift-python27-mitaka/c04fdce/console.html#_2017-03-28_06_08_32_098691 | 18:34 |
| timburke | those KeyErrors seem suspicious -- something with patch/unpatch_policies? i feel like i saw something about that lately... | 18:37 |
| *** Renich has joined #openstack-swift | 18:41 | |
| *** ujjain has quit IRC | 19:12 | |
| *** ujjain has joined #openstack-swift | 19:18 | |
| *** ujjain has joined #openstack-swift | 19:18 | |
| *** SkyRocknRoll_ has quit IRC | 19:18 | |
| *** ujjain has quit IRC | 19:22 | |
| *** ujjain has joined #openstack-swift | 19:27 | |
| *** ujjain has joined #openstack-swift | 19:27 | |
| *** ujjain has quit IRC | 19:41 | |
| *** SkyRocknRoll has joined #openstack-swift | 19:53 | |
| acoles | timburke: agree, the one I looked at last week when you raised this in the meeting also had a KeyError looking up a policy, in ssync test, and I also wondered if some policy patching was not being undone | 19:55 |
| *** SkyRocknRoll has quit IRC | 19:59 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Tests and fixups for sharding stats reporting https://review.openstack.org/450957 | 20:00 |
| *** SkyRocknRoll has joined #openstack-swift | 20:01 | |
| *** klrmn has quit IRC | 20:04 | |
| *** klrmn has joined #openstack-swift | 20:04 | |
| *** silor has quit IRC | 20:28 | |
| *** vinsh has joined #openstack-swift | 21:29 | |
| *** sams-gleb has quit IRC | 21:35 | |
| *** sams-gleb has joined #openstack-swift | 21:36 | |
| *** sams-gleb has quit IRC | 21:36 | |
| *** NM has quit IRC | 21:36 | |
| mattoliverau | Morning | 21:56 |
| *** JimCheung has joined #openstack-swift | 21:59 | |
| *** sams-gleb has joined #openstack-swift | 22:37 | |
| notmyname | tonyb_: oh! new eventlet has been accepted by openstack requirements | 22:38 |
| *** _JZ_ has quit IRC | 22:38 | |
| *** sams-gleb has quit IRC | 22:41 | |
| mattoliverau | acoles: \o/ sharding patch! | 22:54 |
| mattoliverau | timburke: Nice, get_item_since test. I had a quick go at a get_item_since, so might add an additional check to yours (and see if you like it), that is if you dont mind. | 22:56 |
| timburke | mattoliverau: absolutely! and as i left in a comment on my followup, it needs ts.internal | 22:57 |
| notmyname | are you *sure* that it's ts.internal? | 22:57 |
| mattoliverau | timburke: cool, I'll add that too :) | 22:57 |
| notmyname | as opposed to ts.internal? | 22:57 |
| timburke | notmyname: or maybe ts.infernal? | 22:58 |
| mattoliverau | meh, tomato, tomato | 22:58 |
| notmyname | I think it's more "meh, pickle, pickle" | 22:59 |
| *** vint_bra has quit IRC | 23:00 | |
| *** klamath has quit IRC | 23:25 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Add functests for disallowed COPYs into a versioned container https://review.openstack.org/451111 | 23:28 |
| *** catintheroof has joined #openstack-swift | 23:37 | |
| mattoliverau | notmyname: btw, tonyb_ is off this week | 23:42 |
| *** gyee has quit IRC | 23:43 | |
| *** NM has joined #openstack-swift | 23:47 | |
| timburke | notmyname: i *think* https://review.openstack.org/#/c/451118/ will get us our linky? or, set up jobs such that we'll get our linky the next time there's a commit that touches the releasenotes/ subtree? | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!