Tuesday, 2017-10-03

*** itlinux has joined #openstack-swift00:35
thurloatnotmyname: thanks for the thorough answer. will definitely be readon00:38
*** wes_dillingham has quit IRC00:40
*** wes_dillingham has joined #openstack-swift00:41
*** lnxnut has joined #openstack-swift00:42
*** tovin07_ has joined #openstack-swift00:46
*** lnxnut has quit IRC00:53
kota_good morning01:21
*** MeltedLux has quit IRC01:25
*** cshastri has joined #openstack-swift01:26
*** itlinux has quit IRC01:29
*** itlinux has joined #openstack-swift01:29
*** MeltedLux has joined #openstack-swift01:36
*** itlinux has quit IRC01:48
*** lnxnut has joined #openstack-swift01:50
*** szaher has quit IRC01:57
*** MeltedLux has quit IRC01:59
*** lnxnut has quit IRC02:01
*** szaher has joined #openstack-swift02:07
*** itlinux has joined #openstack-swift02:31
*** itlinux has quit IRC02:44
*** lnxnut has joined #openstack-swift02:59
*** lnxnut has quit IRC03:07
*** chsc has joined #openstack-swift03:10
*** chsc has joined #openstack-swift03:10
*** chsc has quit IRC03:15
*** gyee has joined #openstack-swift03:37
*** gyee has quit IRC03:38
*** kei_yama has quit IRC03:45
*** Dinesh_Bhor has joined #openstack-swift03:47
*** gkadam has joined #openstack-swift03:52
*** kei_yama has joined #openstack-swift04:02
*** lnxnut has joined #openstack-swift04:04
*** SkyRocknRoll has joined #openstack-swift04:07
*** mingyu has quit IRC04:17
*** mingyu has joined #openstack-swift04:19
*** mingyu has quit IRC04:19
*** abhitechie has joined #openstack-swift04:20
*** lnxnut has quit IRC04:21
*** pcaruana has joined #openstack-swift05:01
*** psachin has joined #openstack-swift05:04
*** mabrams has joined #openstack-swift05:04
*** skudlik has joined #openstack-swift05:13
*** lnxnut has joined #openstack-swift05:18
*** mingyu has joined #openstack-swift05:20
*** pcaruana has quit IRC05:23
*** mingyu has quit IRC05:25
*** lnxnut has quit IRC05:28
*** mat128 has quit IRC05:36
*** rcernin has joined #openstack-swift05:53
*** ChubYann has quit IRC05:54
*** mingyu has joined #openstack-swift05:55
*** mvk has quit IRC06:13
*** tovin07_ has quit IRC06:22
*** lnxnut has joined #openstack-swift06:24
*** hoonetorg has quit IRC06:27
*** lnxnut has quit IRC06:34
*** hoonetorg has joined #openstack-swift06:41
*** pcaruana has joined #openstack-swift06:44
*** MeltedLux has joined #openstack-swift07:09
*** tesseract has joined #openstack-swift07:17
*** oshritf has joined #openstack-swift07:19
*** SkyRocknRoll has quit IRC07:20
*** oshritf has quit IRC07:21
*** oshritf has joined #openstack-swift07:26
*** oshritf has quit IRC07:29
*** lnxnut has joined #openstack-swift07:31
*** oshritf has joined #openstack-swift07:32
*** SkyRocknRoll has joined #openstack-swift07:37
*** Guest36 has joined #openstack-swift07:39
*** oshritf has quit IRC07:40
*** lnxnut has quit IRC07:41
*** oshritf has joined #openstack-swift07:43
*** oshritf has quit IRC07:53
*** Guest36 has quit IRC07:57
*** oshritf has joined #openstack-swift08:02
*** oshritf has quit IRC08:06
*** mvk has joined #openstack-swift08:10
acolesgood morning08:12
*** oshritf has joined #openstack-swift08:14
*** frekoya_ has joined #openstack-swift08:22
*** wes_dillingham has quit IRC08:27
*** mabrams has quit IRC08:31
*** frekoya_ has quit IRC08:31
kota_good morning acoles08:33
*** wes_dillingham has joined #openstack-swift08:33
*** frekoya_ has joined #openstack-swift08:37
*** mvk has quit IRC08:38
*** lnxnut has joined #openstack-swift08:39
openstackgerritAlistair Coles proposed openstack/swift feature/deep: minor follow up on cleave unit tests  https://review.openstack.org/50910908:40
acoleskota_: o/08:40
kota_o/08:41
*** spectr has joined #openstack-swift08:41
*** frekoya_ has quit IRC08:43
*** mahatic has joined #openstack-swift08:45
openstackgerritAlistair Coles proposed openstack/swift feature/deep: make broker get_shard_range_rows private  https://review.openstack.org/50911108:47
openstackgerritAlistair Coles proposed openstack/swift feature/deep: make broker get_shard_range_rows private  https://review.openstack.org/50911108:48
*** lnxnut has quit IRC08:48
*** mvk has joined #openstack-swift08:54
*** abhitechie has quit IRC09:07
*** abhinavtechie has joined #openstack-swift09:07
openstackgerritHCLTech-SSW proposed openstack/swift master: Add ability to undelete an account.  https://review.openstack.org/50780809:13
*** mvk has quit IRC09:20
*** kei_yama has quit IRC09:20
*** mvk has joined #openstack-swift09:34
*** abhinavtechie has quit IRC09:44
*** geaaru has joined #openstack-swift09:46
*** lnxnut has joined #openstack-swift09:46
*** oshritf has quit IRC09:54
*** lnxnut has quit IRC09:56
*** mingyu has quit IRC10:18
*** wes_dillingham has quit IRC10:46
*** cshastri has quit IRC10:50
*** lnxnut has joined #openstack-swift10:53
*** lnxnut has quit IRC11:02
openstackgerritKota Tsuyuzaki proposed openstack/swift master: Follow up for symlink implementaion  https://review.openstack.org/50914911:06
*** spectr has quit IRC11:07
*** mingyu has joined #openstack-swift11:09
kota_tdasilva, timburke: I'm starting to look up the symlink patch and I've concerned on some point of view on the docs that may be wrong or not enough for the spec. I'd like to get your help on the english terms and docs for that if you could have time.11:11
*** abhitechie has joined #openstack-swift11:11
kota_acoles: maybe you also can help me because I think you had some discussion on symlink at past several mid-cycles ;)11:12
acoleskota_: I can try11:12
kota_acoles: thank you so much!11:13
acoleskota_: what's the question?11:13
kota_let me check my comments on https://review.openstack.org/#/c/232162/59/swift/common/middleware/symlink.py11:15
patchbotpatch 232162 - swift - Symlink implementation.11:15
*** abhitechie has quit IRC11:15
*** abhitechie has joined #openstack-swift11:15
kota_I had some comments on the patch to correct (or get understand correctly) but one of questions that could be issue is the behavior on OPTIONS11:17
kota_acoles: the docs just says, "the request will not be redirected to the target object."11:17
kota_however, the linked (i.e. target actual) object could be in a container different cors setting rather than the symlink object.11:18
kota_i.e. an actual verb (e.g. GET/HEAD) after pre-flight OPTIONS can be failed with 4xx even the client got allowed the methods by OPTIONS request11:21
kota_sorry, not yet tested, just I thought when reading the docs11:22
kota_and not sure if it's by design.11:22
*** abhitechie has quit IRC11:22
kota_acoles: I don't hurry so much for the answer right now, sorry. Just I'd like to notify you to make the patch in your rader.11:25
kota_acoles: And also, I know I have some homework from you for composite ring patches since ptg :/ sorry.11:26
acoleskota_: in terms of the English at line 73-74, I read that as the OPTIONS request will not be redirected, so it does look like you analysis is correct11:28
*** abhitechie has joined #openstack-swift11:29
acolesI'm not familiar with the design decision though11:29
kota_acoles: ok, thanks.11:31
* kota_ is feeling to need clear head and eyes for deeper consideration, so being off from office11:32
acoleskota_: sorry my machine froze on me for a while then11:33
acoleskota_: ok, have a good night11:33
kota_acoles: oh i see.11:33
kota_acoles: thanks for taking time it :D11:33
acoleslines 53-56 I think it is trying to explain why you should not POST to a symlink11:34
kota_yeah, and it's the reason why the response will be 3xx.11:35
acolesthe POST cannot be guaranteed to be redirected to the target because at the time the POST arrives at an object server the current .data file may not yet be the symlink (if the symlink PUT failed to that server). so because POST redirection cannot be guaranteed, it is never attempted. At least, that's the reasoning I recall.11:36
kota_ok... so the basic idea, the client SHOULD NOT POST to a symlink as you said.11:39
kota_i like it11:39
kota_probably, the reason why I felt a little bit complicated is from the previous sentence11:39
kota_ POST request to a symlink must be accompanied of the query parameter11:40
kota_``?symlink=post`` to have the new metadata applied to the symlink object11:40
kota_itself11:40
acoleskota_: the symlink=post param is new to me :)11:40
kota_whooaa!?11:41
kota_anyway11:41
acoleskota_: it has been a long time since I reviewed this, maybe i just forgot11:41
kota_reading from top to bottom, it looks like as POST is supported.. but actually, it seems like the POST result symlink will be never applied anytime on normal GET/HEAD11:42
kota_POST result to symlink11:42
kota_normal GET/HEAD to symlink (i.e. GET/HEAD the target via symlink)11:43
kota_so that, the important information seems that you said above, don't POST to symlink, if you did it, Swift will notify you that it's symlink via 3xx.11:44
kota_imo11:44
acoleskota_: my understanding is that the POST metadata will always be applied to the symlink (not target) but without the symlink=post param the response code is changed to a redirect11:44
kota_acoles: I think you're true.11:45
acolesso then a GET or HEAD will not get that POSTed metadata because it will be redirected to the target11:45
acolesunless it has symlink=get11:46
kota_yes11:46
acoleskota_: I do remember we filled several whiteboards in several hackathons convincing ourselves that clients must redirect the POST to the target, not swift :)11:47
acoles(I remember the whiteboard in Tokyo summit!!!)11:47
kota_then, general question is, is current docs11:47
kota_oh, sorry, i pushed enter key before I filled my sentence11:47
acoleshehe I do that all the11:48
acolestime :D11:48
kota_general question is, does the current docs describe the spec in straightforward?11:48
kota_i didn't feel like it, but not sure because I'm not native.11:49
acoleskota_: hmm, I am not sure if this line 51 is true "A POST request to a symlink must be accompanied of the query parameter11:50
acoles``?symlink=post`` to have the new metadata applied to the symlink object11:50
acolesitself."11:50
kota_i think, it may not be symlink object. because it is not guaranteed (it may be overwritten at the same time)?11:51
*** NM has joined #openstack-swift11:51
acolesI will leave a comment on gerrit11:53
kota_acoles: much helpful, awesome!11:53
acolesOIC!11:54
*** abhitechie has quit IRC11:59
*** lnxnut has joined #openstack-swift11:59
acolesyeah that's weird, POST to a symlink without a symlink header will change it to not be a symlink12:00
acoleskota_: tdasilva will have all the answers :)12:00
kota_yeah we need tdasilva!12:04
*** lnxnut has quit IRC12:09
*** spectr has joined #openstack-swift12:24
*** catintheroof has joined #openstack-swift12:36
*** catintheroof has quit IRC12:36
*** catintheroof has joined #openstack-swift12:36
*** gkadam has quit IRC12:38
*** catintheroof has quit IRC12:41
*** mat128 has joined #openstack-swift12:46
*** catintheroof has joined #openstack-swift12:54
*** lnxnut has joined #openstack-swift13:07
*** chlong has joined #openstack-swift13:10
*** chlong_ has joined #openstack-swift13:10
*** lnxnut has quit IRC13:17
*** SkyRocknRoll has quit IRC13:57
*** spectr has quit IRC13:58
*** spectr has joined #openstack-swift13:59
*** lnxnut has joined #openstack-swift14:14
*** itlinux has joined #openstack-swift14:23
*** lnxnut has quit IRC14:24
*** itlinux has quit IRC14:26
*** chlong has quit IRC14:31
*** chlong_ has quit IRC14:31
*** chlong has joined #openstack-swift14:32
*** honga has quit IRC14:50
*** abhitechie has joined #openstack-swift15:04
*** rcernin has quit IRC15:12
*** pcaruana has quit IRC15:20
*** lnxnut has joined #openstack-swift15:21
*** itlinux has joined #openstack-swift15:29
*** lnxnut has quit IRC15:31
*** itlinux has quit IRC15:31
*** abhitechie has quit IRC15:31
*** chlong has quit IRC15:36
*** mvk has quit IRC15:40
*** chsc has joined #openstack-swift15:43
*** chsc has joined #openstack-swift15:43
*** itlinux has joined #openstack-swift15:44
*** psachin has quit IRC15:45
*** chsc has quit IRC15:50
notmynamegood morning15:57
*** gyee has joined #openstack-swift16:02
timburkegood morning16:11
timburkekota_: acoles: iirc, the ?symlink=post param was purely intended as an escape hatch so you'd get back a 204 instead of a 30716:14
*** silor has joined #openstack-swift16:15
acolestimburke: yep that's how it looked to me16:17
acolestimburke: fyi I've been working on unit test for _misplaced_objects16:19
*** geaaru has quit IRC16:20
acoleshmmm, zuul does not seem happy16:21
timburkenot it does not. i've been peeking in -infra a bit, seems like we may be better off ignoring gate results for today16:22
timburkedon't be afraid of patch chains if it prevents conflicts -- then once things are actually functional, we can just issue a bunch of rechecks and wait, rather than spend time chasing down merge conflicts16:23
acolestimburke: ok. I won't be able to push what I have because I locally rebased https://review.openstack.org/508635 on feature/deep to pick up some test infrastructure, hoping that 508635 would have merged by end of my day16:23
patchbotpatch 508635 - swift (feature/deep) - Fix up misplaced-objects handling16:23
*** SkyRocknRoll has joined #openstack-swift16:24
acolesshould have rebased it then merged but I wanted to get it in the pipe16:24
timburkethat's fine -- push it up anyway, i'll see what the state of the gate is like and +A / recheck as it makes sense before heading out today16:25
acolestimburke: ok, let' see how far I get before end of my day, still have some work to do16:25
*** lnxnut has joined #openstack-swift16:29
*** SkyRocknRoll has quit IRC16:29
*** silor has quit IRC16:34
*** lnxnut has quit IRC16:36
*** tesseract has quit IRC16:37
*** silor has joined #openstack-swift16:39
*** vint_bra has joined #openstack-swift17:01
*** chsc has joined #openstack-swift17:02
*** chsc has joined #openstack-swift17:02
*** gabor_antal has quit IRC17:30
*** gabor_antal_ has joined #openstack-swift17:31
*** lnxnut has joined #openstack-swift17:34
*** itlinux has quit IRC17:41
*** itlinux has joined #openstack-swift17:42
*** lnxnut has quit IRC17:43
openstackgerritPete Zaitcev proposed openstack/swift master: PUT+POST and its development test  https://review.openstack.org/42791117:44
claygzaitcev: This is the 3-phase version (PUT+POST+POST) !!! whoot!17:45
zaitcevclayg: yes, I only need to find and add torgomatic's patch that allows linkat(), run functests, and fix-up unit tests. Then it should be good to go, unless you guys find something glaringly broken.17:47
*** itlinux has quit IRC17:48
zaitcevHonestly it was a surprise to me that object server examines the data that's stored in it. Some KISS principle that is.17:48
*** itlinux has joined #openstack-swift17:54
*** itlinux has quit IRC17:58
*** gabor_antal_ has quit IRC18:06
*** gabor_antal has joined #openstack-swift18:07
claygawesome!  I've put it on my todo - I'll try to give it a glance before the meeting tomorrow - but probably won't get in a proper review until the end of the week.18:11
*** gabor_antal has quit IRC18:15
*** gabor_antal has joined #openstack-swift18:16
*** itlinux has joined #openstack-swift18:21
*** vint_bra has quit IRC18:28
*** mingyu has quit IRC18:33
*** vint_bra has joined #openstack-swift18:36
*** lnxnut has joined #openstack-swift18:41
acolesmattoliverau: mahatic kota_ fwiw I'm going to try to be around at 0700 UTC tomorrow18:47
*** lnxnut has quit IRC18:50
*** mvk has joined #openstack-swift18:56
*** ChubYann has joined #openstack-swift19:00
*** vint_bra has quit IRC19:01
acolestimburke: IDK where https://review.openstack.org/#/c/508635/ has gone - last seen ~ 2hrs starting gate jobs but I can't see it on zuul status19:01
patchbotpatch 508635 - swift (feature/deep) - Fix up misplaced-objects handling19:01
*** NM has quit IRC19:02
acolesso I'm pushing my tests which will rebase 50863519:02
timburkeyeah, so don't worry about it -- i'll kick things off again after the rollback19:02
timburkeperfect!19:02
acolesoh right, is zuul rolling back?19:02
timburkehttp://lists.openstack.org/pipermail/openstack-dev/2017-October/123049.html19:03
*** NM has joined #openstack-swift19:05
*** skudlik has quit IRC19:06
*** vint_bra has joined #openstack-swift19:08
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Fix up misplaced-objects handling  https://review.openstack.org/50863519:09
openstackgerritAlistair Coles proposed openstack/swift feature/deep: unit tests for _misplaced_objects  https://review.openstack.org/50925619:09
acolestimburke: ok I'm done for today ^^19:09
timburkeacoles: have a good night!19:10
acolesgood night19:10
*** NM has quit IRC19:15
*** NM has joined #openstack-swift19:15
*** catintheroof has quit IRC19:32
*** catintheroof has joined #openstack-swift19:32
*** chlong_ has joined #openstack-swift19:43
*** silor has quit IRC19:44
*** chlong has joined #openstack-swift19:44
*** NM has quit IRC19:44
*** chlong has quit IRC19:46
*** lnxnut has joined #openstack-swift19:48
*** NM has joined #openstack-swift19:52
*** lnxnut has quit IRC19:59
*** NM has quit IRC20:04
*** NM has joined #openstack-swift20:08
*** catintheroof has quit IRC20:10
*** catintheroof has joined #openstack-swift20:11
*** itlinux has quit IRC20:22
*** itlinux has joined #openstack-swift20:32
*** catintheroof has quit IRC20:42
*** catintheroof has joined #openstack-swift20:42
*** itlinux has quit IRC20:57
*** chlong_ has quit IRC21:03
*** itlinux has joined #openstack-swift21:03
*** NM has quit IRC21:14
*** gyee has quit IRC21:34
openstackgerritMerged openstack/swift feature/deep: Fix up misplaced-objects handling  https://review.openstack.org/50863521:44
*** itlinux has quit IRC21:53
*** lnxnut has joined #openstack-swift21:55
mattoliveraumorning22:02
*** lnxnut has quit IRC22:07
timburkemattoliverau: o/22:10
mattoliverauDay light savings means I start earlier now.. but also means I need more coffee to get going :P22:11
mattoliverautimburke: o/22:11
*** vint_bra has quit IRC22:27
openstackgerritTim Burke proposed openstack/swift master: Clean up SLO tests and docs  https://review.openstack.org/50930522:43
openstackgerritTim Burke proposed openstack/swift master: Let clients request heartbeats during SLO PUTs  https://review.openstack.org/50930622:43
*** mingyu has joined #openstack-swift22:49
*** mingyu has quit IRC22:53
*** catintheroof has quit IRC23:01
*** lnxnut has joined #openstack-swift23:04
*** wes_dillingham has joined #openstack-swift23:06
*** lnxnut has quit IRC23:14
*** itlinux has joined #openstack-swift23:21
*** kei_yama has joined #openstack-swift23:22
*** vint_bra has joined #openstack-swift23:30
*** itlinux has quit IRC23:32
*** chsc has quit IRC23:34
*** abhitechie has joined #openstack-swift23:49

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!