| *** vinsh_ has quit IRC | 00:02 | |
| *** vinsh has joined #openstack-swift | 00:04 | |
| *** brimestone has quit IRC | 00:05 | |
| tdasilva | notmyname: this doesn't look right, does it? https://review.openstack.org/#/c/562428/ | 00:25 |
|---|---|---|
| patchbot | patch 562428 - swift - Merge branch 'feature/s3api' into master | 00:25 |
| *** yashmurty has joined #openstack-swift | 00:42 | |
| *** ukaynar has quit IRC | 00:46 | |
| *** ukaynar has joined #openstack-swift | 00:47 | |
| *** drewn3ss_ has joined #openstack-swift | 00:49 | |
| *** drewn3ss has quit IRC | 00:50 | |
| *** drewn3ss_ has quit IRC | 00:50 | |
| *** drewn3ss has joined #openstack-swift | 00:51 | |
| notmyname | tdasilva: how so? | 00:51 |
| tdasilva | notmyname: i guess i have the same question as Pete | 00:51 |
| tdasilva | there's no code to review | 00:51 |
| *** ukaynar has quit IRC | 00:52 | |
| notmyname | ah | 00:53 |
| notmyname | hmm... how should we do this? the code is in the ancestor. so to see it, grab that patch and run it. diff the whole thing against master | 00:56 |
| notmyname | I didn't make a single commit with all the code. just the merge commit to maintain the history | 00:58 |
| *** vinsh has quit IRC | 01:03 | |
| zaitcev | So that means that I cannot leave comments to the code, unless I file a review with my proposed changes. | 01:03 |
| zaitcev | git review -d 562428 appears to include the swift/common/middleware/s3api, so that's something | 01:04 |
| notmyname | any ideas on how to make it better? | 01:06 |
| timburke | i mean, we *could* do it as a single commit with all the code dropped in... | 01:07 |
| timburke | idk if that's much of any better or not | 01:07 |
| tdasilva | idk...i guess i was in the same boat as zaitcev expecting to see all the code...trying to remember how we did other feature branches... | 01:08 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: WIP: Add "Manual" Tiering Feature https://review.openstack.org/546958 | 01:08 |
| timburke | much closer to the "here's all the code" approach | 01:08 |
| zaitcev | Like I sad, I see all the code just fine. I just cannot put any review comments on it. | 01:10 |
| timburke | at least for crypto -- we took all the work from the feature branch, broke it up into a few commits that told a story, proposed to a review branch. this'd be similar, but the story is "here's all of swift3, renamed as s3api" | 01:11 |
| timburke | since that story fits in a single commit, we could get away with skipping the branch | 01:11 |
| tdasilva | here's crypto: https://review.openstack.org/#/c/328208/ | 01:11 |
| patchbot | patch 328208 - swift (feature/crypto-review) - Enable object body and metadata encryption (MERGED) | 01:11 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: WIP: Add "Manual" Tiering Feature https://review.openstack.org/546958 | 01:12 |
| timburke | here's the merge commit -- https://review.openstack.org/#/c/336407/ | 01:14 |
| patchbot | patch 336407 - swift - Merge at-rest encryption feature into master (MERGED) | 01:14 |
| timburke | which, by that time, we'd already agreed on all the step along the way | 01:14 |
| *** cshastri has joined #openstack-swift | 01:19 | |
| *** germs has quit IRC | 01:30 | |
| *** germs has joined #openstack-swift | 01:30 | |
| *** germs has quit IRC | 01:30 | |
| *** germs has joined #openstack-swift | 01:30 | |
| openstackgerrit | Merged openstack/swift feature/deep: Stop replicating object rows when container could be sharded https://review.openstack.org/549678 | 01:39 |
| openstackgerrit | Merged openstack/swift feature/deep: Call _sync_other_items from _choose_replication_mode https://review.openstack.org/562340 | 01:46 |
| *** m_kazuhiro has joined #openstack-swift | 01:47 | |
| m_kazuhiro | good morning | 01:48 |
| mattoliverau | m_kazuhiro: morning | 01:48 |
| mattoliverau | again | 01:48 |
| m_kazuhiro | mattoliverau: morning again :) | 01:49 |
| openstackgerrit | Merged openstack/swift master: Deprecate auth_uri option https://review.openstack.org/562093 | 01:56 |
| notmyname | let me know what you want to do with the s3api branch. I don't care too much if it's a chain+merge commit or if it's all in one commit. I'm happy to make either happen, and I think kota_ + reviewers should choose whatever will make them most production | 02:15 |
| m_kazuhiro | IMO, chain + merge commit is easier to review. I want to understand and review codes one by one. | 02:23 |
| *** wangyx has joined #openstack-swift | 02:25 | |
| *** psachin has joined #openstack-swift | 02:42 | |
| openstackgerrit | wangqi proposed openstack/swift master: Modify the empty list ensure method https://review.openstack.org/562451 | 02:46 |
| *** cshastri has quit IRC | 03:35 | |
| *** m_kazuhiro has quit IRC | 03:42 | |
| *** m_kazuhiro has joined #openstack-swift | 03:54 | |
| *** cshastri has joined #openstack-swift | 03:58 | |
| *** germs has quit IRC | 04:12 | |
| *** hoonetorg has quit IRC | 04:27 | |
| *** yashmurty has quit IRC | 04:36 | |
| *** yashmurty has joined #openstack-swift | 04:39 | |
| *** hoonetorg has joined #openstack-swift | 04:41 | |
| *** f0o has quit IRC | 05:16 | |
| *** links has joined #openstack-swift | 05:18 | |
| *** armaan has joined #openstack-swift | 05:20 | |
| *** yashmurty has quit IRC | 05:23 | |
| kota_ | back to online | 05:24 |
| *** yashmurty has joined #openstack-swift | 05:26 | |
| *** ccamacho has quit IRC | 05:38 | |
| *** geaaru has quit IRC | 05:39 | |
| *** armaan has quit IRC | 05:51 | |
| kota_ | hmm creating one commit to propose master, it would be obviously a huge patch. | 05:52 |
| kota_ | https://gist.github.com/bloodeagle40234/9987eff3270f1b63b63c7e4a447e054f | 05:52 |
| kota_ | however, i prefer to prepare a clean commit (one or a few patches) to propose master. | 05:53 |
| *** f0o has joined #openstack-swift | 05:54 | |
| kota_ | a merge commit can have pros to keep the commit logs originally written in feature/s3api but the cons is stability is not guaranteed for each commit because feature/s3api is assumed fast develop with a few reviewers. | 05:55 |
| kota_ | creating a (or few) brand-new commit to propose to the master is the easiest to comment/know the diff on the gerrit imo. cons is it maybe a big. | 05:57 |
| *** TobbeCN has joined #openstack-swift | 05:57 | |
| kota_ | evne we take a creating-a-commit-way, maybe the step is, fetch feature/s3api, rebase it to the master's HEAD, make then as a single commit via rebase -i? | 05:58 |
| kota_ | i'm not so worried about the way on the perspective to save the commit log because we could have a tag for the development-history (in term of swift3, it's migration history) as a reference | 05:59 |
| kota_ | as well as we did ec/storage_policy like https://github.com/openstack/swift/tree/erasure_code_dev_history | 06:01 |
| *** itlinux has joined #openstack-swift | 06:02 | |
| kota_ | the consideration point is how many patches if we want to separate the single large commit but the perspective of stability, my vote is one single commit anyway. | 06:05 |
| kota_ | if we want to separate the patches for revews, we should go feature/s3api-review branch as well as crypt-review (, ec or sp-review), then make the authorized code int a single patch like as Tim suggested | 06:06 |
| kota_ | oh... no? | 06:07 |
| *** itlinux has quit IRC | 06:09 | |
| kota_ | ah... in the crypt case, the trigger looks like just a merge commit but i think it's the merge from feature/crypt-review, not from feature/crypt directly. | 06:11 |
| *** itlinux has joined #openstack-swift | 06:12 | |
| kota_ | so IIRC acoles maintained the feature/crypt-review to be a few (5-6?) changes build from feature/crypto, then each commit is independent. | 06:13 |
| kota_ | then, after review all commits in the -review branch, we get simply the merge commit into a master. | 06:14 |
| kota_ | so... | 06:14 |
| kota_ | anyway, what we need is create commit(s) to review the code to the master, while the task, we need to determine single commit vs multi commits. | 06:15 |
| *** armaan has joined #openstack-swift | 06:16 | |
| kota_ | though i could manage the review branch, i wonder how many i can separate the patch to smaller ones because the components in the s3api module are rerated each other... | 06:18 |
| kota_ | just separating docs/code/tests doesn't seem to be worth to me for now. | 06:18 |
| kota_ | notmyname, m_kazuhiro, zaitcev, tdasilva, timburke, mattoliverau: how do you think of? | 06:19 |
| *** pcaruana has joined #openstack-swift | 06:21 | |
| *** itlinux has quit IRC | 06:32 | |
| *** itlinux has joined #openstack-swift | 06:37 | |
| m_kazuhiro | kota_: As you said, if s3api modules are related each other, I think single commit is better way... | 06:38 |
| m_kazuhiro | kota_: In your expectation, how many lines of code will be included in the single commit? | 06:42 |
| kota_ | m_kazuhiro: https://gist.github.com/bloodeagle40234/9987eff3270f1b63b63c7e4a447e054f | 06:43 |
| kota_ | it looks to have duplicate counts and not rebased on the master yet. | 06:44 |
| kota_ | that is simple result "git diff master feature/s3api --stat" | 06:44 |
| *** paramite has joined #openstack-swift | 06:49 | |
| *** m_kazuhiro has quit IRC | 06:56 | |
| *** ccamacho has joined #openstack-swift | 06:58 | |
| *** itlinux has quit IRC | 07:02 | |
| *** tesseract has joined #openstack-swift | 07:10 | |
| *** rcernin has quit IRC | 07:33 | |
| acoles | good morning | 07:36 |
| acoles | kota_: for crypto-review we had approx 5 final patches. that made sense because there we 1. non-behavioural changes to existing code, refactor, test infrastructure etc to support crypto 2. new 'generic' features such as transient sysmeta 3. the crypto pieces 4. doc (I don't remember 5 :)) | 07:38 |
| *** geaaru has joined #openstack-swift | 07:38 | |
| acoles | oh yes, there was the 'generic' container update header override mechanism | 07:39 |
| acoles | kota_: the advantage of keeping doc separate and last in chain is that it tends to get a lot of changes so they can be made without needing new patchsets for the code, plus anyone can push doc changes without impacting the code patches | 07:40 |
| *** pcaruana has quit IRC | 07:45 | |
| *** pcaruana has joined #openstack-swift | 07:46 | |
| acoles | clayg: did you have a follow up patch for https://review.openstack.org/#/c/549678 on your machine? you marked some stuff 'Done'... | 07:57 |
| patchbot | patch 549678 - swift (feature/deep) - Stop replicating object rows when container could ... (MERGED) | 07:57 |
| *** cbartz has joined #openstack-swift | 08:00 | |
| *** cbartz has quit IRC | 08:00 | |
| *** cbartz has joined #openstack-swift | 08:01 | |
| *** hseipp has joined #openstack-swift | 08:08 | |
| *** mikecmpbll has joined #openstack-swift | 08:09 | |
| *** gkadam has joined #openstack-swift | 08:16 | |
| *** itlinux has joined #openstack-swift | 08:28 | |
| kota_ | acoles: sounds good lessons | 09:08 |
| *** itlinux has quit IRC | 09:09 | |
| kota_ | thinking of scheme, what i can do is to make 2 patch sets, docs and codes. | 09:10 |
| kota_ | because swift3 was obviously an independent middleware from swift so the simple import should not require refactor and 'generic' change for the existing code. | 09:11 |
| kota_ | but you all knows a bunch of lines is inserted for importing :P | 09:13 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Structural reorganisation of sharder.py https://review.openstack.org/562530 | 09:20 |
| acoles | kota_: that seems to make sense, code + docs | 09:50 |
| *** armaan has quit IRC | 09:55 | |
| *** armaan has joined #openstack-swift | 09:55 | |
| *** kei_yama has quit IRC | 10:07 | |
| *** mvk has quit IRC | 10:47 | |
| *** armaan has quit IRC | 10:59 | |
| *** armaan has joined #openstack-swift | 10:59 | |
| *** silor has joined #openstack-swift | 11:22 | |
| *** armaan has quit IRC | 11:37 | |
| *** mvk has joined #openstack-swift | 11:44 | |
| *** yashmurty has quit IRC | 11:48 | |
| *** psachin has quit IRC | 11:52 | |
| *** yashmurty has joined #openstack-swift | 11:58 | |
| *** armaan has joined #openstack-swift | 12:00 | |
| *** psachin has joined #openstack-swift | 12:01 | |
| *** armaan has quit IRC | 12:11 | |
| *** armaan has joined #openstack-swift | 12:17 | |
| *** armaan has quit IRC | 12:22 | |
| openstackgerrit | Merged openstack/swift feature/deep: Structural reorganisation of sharder.py https://review.openstack.org/562530 | 12:26 |
| *** NM has joined #openstack-swift | 12:27 | |
| *** bob_cheesey has quit IRC | 12:28 | |
| *** cshastri has quit IRC | 12:32 | |
| *** silor has quit IRC | 13:00 | |
| *** armaan has joined #openstack-swift | 13:02 | |
| *** wangyx has quit IRC | 13:12 | |
| *** wangyx has joined #openstack-swift | 13:13 | |
| *** jistr is now known as jistr|mtg | 13:14 | |
| *** wangyx has quit IRC | 13:17 | |
| *** vinsh has joined #openstack-swift | 13:19 | |
| *** cshastri has joined #openstack-swift | 13:24 | |
| *** vinsh_ has joined #openstack-swift | 13:25 | |
| *** vinsh has quit IRC | 13:28 | |
| *** psachin has quit IRC | 13:33 | |
| *** TobbeCN has quit IRC | 13:35 | |
| *** TobbeCN has joined #openstack-swift | 13:36 | |
| *** wangyx has joined #openstack-swift | 13:36 | |
| *** TobbeCN has quit IRC | 13:40 | |
| *** wangyx has quit IRC | 13:40 | |
| *** cshastri has quit IRC | 13:41 | |
| *** vinsh_ has quit IRC | 13:46 | |
| *** vinsh has joined #openstack-swift | 13:47 | |
| tdasilva | kota_: i like your plan of code + docs patch... | 13:51 |
| *** yashmurty has quit IRC | 13:55 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Store root/parent container in shard ranges table https://review.openstack.org/562603 | 14:04 |
| *** NM has quit IRC | 14:09 | |
| *** jistr|mtg is now known as jistr | 14:10 | |
| *** wangyx has joined #openstack-swift | 14:18 | |
| *** wangyx__ has joined #openstack-swift | 14:19 | |
| *** wangyx has quit IRC | 14:23 | |
| *** armaan has quit IRC | 14:28 | |
| *** brimestone has joined #openstack-swift | 14:55 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Logging and exception handling cleanup https://review.openstack.org/562642 | 14:56 |
| *** vinsh has quit IRC | 14:58 | |
| *** TobbeCN has joined #openstack-swift | 15:04 | |
| *** TobbeCN has quit IRC | 15:08 | |
| *** cshastri has joined #openstack-swift | 15:18 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Take lock before unlinking db files https://review.openstack.org/559796 | 15:22 |
| *** ccamacho has quit IRC | 15:23 | |
| *** germs has joined #openstack-swift | 15:24 | |
| *** germs has quit IRC | 15:24 | |
| *** germs has joined #openstack-swift | 15:24 | |
| *** PsionTheory has joined #openstack-swift | 15:38 | |
| *** armaan has joined #openstack-swift | 15:42 | |
| *** ccamacho has joined #openstack-swift | 15:42 | |
| *** ccamacho has quit IRC | 15:43 | |
| *** ccamacho has joined #openstack-swift | 15:43 | |
| *** armaan has quit IRC | 15:47 | |
| *** armaan has joined #openstack-swift | 15:48 | |
| *** mvk has quit IRC | 15:59 | |
| *** links has quit IRC | 16:01 | |
| *** armaan has quit IRC | 16:02 | |
| *** pcaruana has quit IRC | 16:03 | |
| *** ccamacho has quit IRC | 16:06 | |
| *** geaaru has quit IRC | 16:07 | |
| *** cbartz has quit IRC | 16:07 | |
| *** wangyx__ has quit IRC | 16:12 | |
| *** wangyx has joined #openstack-swift | 16:12 | |
| *** wangyx has quit IRC | 16:17 | |
| notmyname | good morning | 16:17 |
| *** itlinux has joined #openstack-swift | 16:19 | |
| *** hseipp has quit IRC | 16:25 | |
| *** mvk has joined #openstack-swift | 16:25 | |
| *** ukaynar_ has joined #openstack-swift | 16:32 | |
| *** mikecmpbll has quit IRC | 16:39 | |
| *** ukaynar_ has quit IRC | 16:43 | |
| *** NM has joined #openstack-swift | 16:43 | |
| *** brimestoned has joined #openstack-swift | 16:44 | |
| *** brimestone has quit IRC | 16:45 | |
| *** jiaopengju has quit IRC | 16:47 | |
| *** ukaynar_ has joined #openstack-swift | 16:53 | |
| *** brimestoned has quit IRC | 16:56 | |
| notmyname | about one month until swift's 9th birthday (may 17) | 16:58 |
| *** itlinux has quit IRC | 17:02 | |
| *** cshastri has quit IRC | 17:06 | |
| *** ukaynar_ has quit IRC | 17:08 | |
| *** gkadam has quit IRC | 17:09 | |
| *** armaan has joined #openstack-swift | 17:24 | |
| notmyname | "The fourth Project Teams Gathering will be held September 10-14th back at the Renaissance Stapleton Hotel in Denver, Colorado" | 17:25 |
| notmyname | and "...this time around the Ops Meetup will be co-located with the Denver PTG" | 17:25 |
| tdasilva | same place, right? | 17:35 |
| notmyname | yes | 17:35 |
| tdasilva | oh wow, prices doubled | 17:37 |
| timburke | and that's when you're early | 17:37 |
| acoles | the train place? | 17:51 |
| *** mvk has quit IRC | 17:52 | |
| notmyname | yes | 17:52 |
| timburke | apparently that situation's gotten better. apparently. | 17:52 |
| *** mikecmpbll has joined #openstack-swift | 17:55 | |
| *** armaan has quit IRC | 17:56 | |
| tdasilva | acoles: what was the hotel that kota_ was staying at again? ;) | 17:59 |
| acoles | tdasilva: there was one smaller hotel just down the street, I forget the name | 18:00 |
| tdasilva | i just remember that someone, i think it was kota_ stayed farther away from the train line... | 18:01 |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Rename replication stats for sharding databases deferred https://review.openstack.org/562772 | 18:02 |
| tdasilva | anyhow, looks like the problem is no longer there... | 18:02 |
| *** cshastri has joined #openstack-swift | 18:05 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Rename replication stats for sharding databases deferred https://review.openstack.org/562772 | 18:05 |
| *** NM has quit IRC | 18:08 | |
| *** NM has joined #openstack-swift | 18:10 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: WIP Make first GET to root return objects or shard ranges as appropriate https://review.openstack.org/562777 | 18:18 |
| *** cshastri has quit IRC | 18:19 | |
| *** guimaluf has joined #openstack-swift | 18:20 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Use string values when passing db state in headers https://review.openstack.org/562780 | 18:22 |
| *** TobbeCN has joined #openstack-swift | 18:36 | |
| *** TobbeCN has quit IRC | 18:41 | |
| *** tesseract has quit IRC | 18:55 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Use non-ascii container name in a probe test https://review.openstack.org/561856 | 19:02 |
| *** early` has quit IRC | 19:52 | |
| *** ukaynar has joined #openstack-swift | 19:54 | |
| *** early has joined #openstack-swift | 19:55 | |
| *** ukaynar has quit IRC | 19:58 | |
| *** gyee has joined #openstack-swift | 20:32 | |
| *** itlinux has joined #openstack-swift | 20:36 | |
| *** itlinux has quit IRC | 20:41 | |
| *** paramite has quit IRC | 20:45 | |
| *** armaan has joined #openstack-swift | 21:14 | |
| *** NM has quit IRC | 21:30 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Use string values when passing db state in headers https://review.openstack.org/562780 | 21:31 |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Fix flakey unit test https://review.openstack.org/562851 | 22:17 |
| *** knylander has quit IRC | 22:18 | |
| *** mvk has joined #openstack-swift | 22:24 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Drop items query param https://review.openstack.org/562856 | 22:29 |
| *** rcernin has joined #openstack-swift | 22:30 | |
| mattoliverau | morning | 22:33 |
| timburke | mattoliverau: o/ | 22:34 |
| mattoliverau | hmm, september 10 - 14... 10th is the calulated due date of the baby... so I'm pretty sure I wont be at the PTG. | 22:35 |
| *** TobbeCN has joined #openstack-swift | 22:37 | |
| *** TobbeCN has quit IRC | 22:41 | |
| *** kallenp has joined #openstack-swift | 22:43 | |
| kallenp | Hallo, howto swift list by date ? | 22:44 |
| *** Sukhdev has joined #openstack-swift | 22:49 | |
| *** armaan has quit IRC | 22:50 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Use non-ascii container name in a probe test https://review.openstack.org/561856 | 22:50 |
| *** armaan has joined #openstack-swift | 22:50 | |
| timburke | kallenp: list objects within a container, sorted by last modified? unfortunately, swift can't -- it only has an index on (deleted, name) so it can return results in lexicographical order | 22:51 |
| timburke | now, if you can structure your object names to start with the date uploaded... | 22:52 |
| *** kallenp has left #openstack-swift | 23:10 | |
| clayg | timburke: with the data architect services like a boss | 23:16 |
| *** kei_yama has joined #openstack-swift | 23:17 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Narrow race in _replicate_object https://review.openstack.org/561091 | 23:26 |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep: Narrow race in _replicate_object https://review.openstack.org/561091 | 23:26 |
| *** mikecmpbll has quit IRC | 23:40 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!