Tuesday, 2016-05-17

*** suyash has quit IRC00:05
*** ouchkernel has quit IRC00:09
*** ouchkernel has joined #openstack-swift00:15
*** sams-gleb has joined #openstack-swift00:26
*** NM has joined #openstack-swift00:30
*** hosanai has quit IRC00:30
*** lyrrad has quit IRC00:31
*** bill_az has quit IRC00:33
*** sams-gleb has quit IRC00:33
*** StraubTW has joined #openstack-swift00:34
*** jamielennox|away is now known as jamielennox00:37
*** vint_bra has joined #openstack-swift00:38
*** StraubTW has quit IRC00:39
*** dmorita_ has joined #openstack-swift00:40
*** dmorita has quit IRC00:40
*** diogogmt has joined #openstack-swift00:42
*** gyee has quit IRC00:52
openstackgerritTim Burke proposed openstack/swift: Make get_info requests useful with recheck_*_existence == 0  https://review.openstack.org/31021600:57
*** Jeffrey4l has joined #openstack-swift01:02
openstackgerritMatthew Oliver proposed openstack/swift: Disable fallocate_reserve in SAIO object-servers  https://review.openstack.org/31715801:08
*** wadeholler has quit IRC01:12
*** klrmn has quit IRC01:12
*** wadeholler has joined #openstack-swift01:13
*** wadeholler has quit IRC01:18
*** StraubTW has joined #openstack-swift01:18
*** wadeholler has joined #openstack-swift01:24
*** wadeholler has quit IRC01:26
*** thumpba has quit IRC01:27
*** sams-gleb has joined #openstack-swift01:30
*** sams-gleb has quit IRC01:39
kota_hello01:40
mattoliveraukota_: o/01:42
openstackgerritTim Burke proposed openstack/swift: Use the same key for memcache and env['swift.infocache']  https://review.openstack.org/31056601:44
kota_mattoliverau: o/01:55
*** jamielennox is now known as jamielennox|away01:59
*** klrmn has joined #openstack-swift02:13
*** lakshmiS has quit IRC02:31
*** sams-gleb has joined #openstack-swift02:36
*** sams-gleb has quit IRC02:44
*** sgundur1 has joined #openstack-swift02:45
pdardeaugood evening02:54
kota_pdardeau: good evening02:54
mattoliveraupdardeau: evening02:58
*** StraubTW has quit IRC02:59
*** janonymous has joined #openstack-swift02:59
*** duobei has joined #openstack-swift02:59
pdardeaukota_: o/03:02
pdardeaumattoliverau: o/03:02
openstackgerritOpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925003:03
openstackgerritOpenStack Proposal Bot proposed openstack/swift: Updated from global requirements  https://review.openstack.org/8873603:03
*** itlinux has quit IRC03:09
*** psachin has joined #openstack-swift03:17
*** CaioBrentano has joined #openstack-swift03:17
*** CaioBrentano has quit IRC03:22
notmynamebig change to http://not.mn/swift/swift_community_dashboard.html just finished03:24
notmynamethere's now a section for stuff that needs reviewer follow-up03:24
notmynamethese are patches that are (1) open (2) mergable (2) have had some reviewer comment (3) have the last comment from the owner and (4) have no current negative reviews (WIP, CI failure, negative reviews)03:25
notmynamesorted from oldest to newest (top to bottom)03:26
notmynametimburke: I think you'll like this change :-)03:29
notmynamenext steps will be to add the "community stars" weighting to the sorting03:30
notmynamethen to add some personalization (ie what's important for *you* to review)03:30
*** links has joined #openstack-swift03:31
notmynameFYI I think tomorrow's -infra meeting (1900UTC) may have some golang related stuff. and the TC meeting (2000UTC) will have a discussion on technical merits of using golang03:34
notmynameif you are interested in participating, please feel free03:34
notmynameof course, I'll be attending both03:34
mattoliveraunotmyname: cool03:35
notmynamemattoliverau: meetings or dashboard?03:36
notmyname:-)03:36
mattoliverauall of the above :) but yeah the dashboard.03:36
mahaticdashboard is nice03:36
notmynamemahatic: I hope it's useful03:37
mattoliveraunotmyname: the dashboard didn't scroll for me, so I had to shrink the size. but that might just be me cause I'm kinda blind :)03:37
mahaticthe colors indicate activity?03:37
*** sgundur2 has joined #openstack-swift03:37
notmynamemahatic: no. the colors indicate "you've been to this link" or not03:37
mahaticoic, cool03:37
*** itlinux has joined #openstack-swift03:38
notmynamemahatic: ah, interesting. so there are 3 different layouts it uses depending on how wide your screen is03:38
notmynamemattoliverau: ^03:38
notmynamemattoliverau: the narrowest is decent for phones03:38
*** sgundur2 has quit IRC03:38
notmynamemattoliverau: the middle is what I normally use on my browser03:38
mattoliverauI do have a wide screen.. but I lost number 22 from the list.03:38
notmynamemattoliverau: and the widest is specifically tailored to the big TV on the wall in my office ;-)03:39
*** itlinux has quit IRC03:39
mattoliverauahh ok03:39
notmynameso yeah, the big TV layout intentionally hides all the scrolling stuff03:39
notmynameand has a flatter color scheme that's better for a TV and viewing at a distance03:39
mahaticlooks perfect on my tiny screen :) will check out at work screen03:39
*** sgundur1 has quit IRC03:40
*** vint_bra has quit IRC03:41
*** sams-gleb has joined #openstack-swift03:41
*** rickyrem has quit IRC03:44
brianclinenotmyname: very nice - what comprises the relative score for the 2 left panes? (in the tooltip that pops up)03:44
notmynamebriancline: hard coded to 100% :-)03:44
notmynamebecause I have one html snippet template for all of those individual <li> things. and it takes a relative score parameter03:45
notmynameso on the community weighting one, the baseline is the first one. but the others are just set to 100 right now03:45
notmynameI guess if I sort by the weighted stars I could add that in...hmmm...or actually use the sort key itself for the value03:46
*** NM has quit IRC03:46
brianclineah, makes sense :)03:47
brianclinehaving the unreviewed and follow-up lists are very very helpful03:48
*** sams-gleb has quit IRC03:48
notmynameyeah, with the unreviewed list, I saw that it quickly went from 26 (IIRC) when I first created it down to about 3-6 on a normal basis03:50
notmynameI hope the same for the needs follow-up ones03:50
mahatichmm easy question maybe. now that we have server side copy, the COPY verb is gone (and so is using Destination header in requests i think?). So the only way to copy is use x-copy-from header, am i correct?03:51
mahaticon PUT request03:52
notmynamebriancline: sorry, setting the tooltip values to somethign that's contextually relevant (instead of 100) will take more effort than i want to expend on it tonight03:52
*** duobei has quit IRC03:52
notmynamemahatic: no. the COPY verb still exists03:52
notmynameit was just moved from the proxy->object_controller to middleware03:53
mahaticoh I quickly did a curl -XCOPY and got method not supported. will check03:53
brianclinenotmyname: no worries, it was just a curiosity. wasn't meant to give you more work to do ;)03:54
*** jamielennox|away is now known as jamielennox03:55
mahaticnotmyname: thanks.you're right, my header values were wrong03:55
notmynameah ok :-)03:56
notmynameyeah, I just confirmed locally too. https://gist.github.com/notmyname/47ae97d800e8f85f58907758c8fb837303:56
mahaticcool03:57
*** bill_az has joined #openstack-swift03:58
notmynamewow! if you use symantic for anything at all, either uninstall it or update it immediately. remote ring-0 memory corruption that allows for arbitrary execution by simply going to a web page or reading an email https://bugs.chromium.org/p/project-zero/issues/detail?id=82004:00
*** psachin has quit IRC04:02
brianclineholy cow04:02
notmynamea lot of IT admins are going to have a really bad day04:04
openstackgerritMatthew Oliver proposed openstack/swift: disable_fallocate also disables fallocate_reserve  https://review.openstack.org/31715804:04
mattoliveraunotmyname: ^ thats the other version04:04
notmynamemattoliverau: I like how you just reused the change-id for a completely different patch ;-)04:04
notmynamemattoliverau: yeah, I like that a lot better. thanks04:06
*** sheel has joined #openstack-swift04:06
mattoliveraunotmyname: yeah, I know, that way there is history on the discussions :)04:06
notmynamevery good point04:06
notmyname(and because I just left that comment, this patch won't be on the "needs reviewer followup" section. ha ha!)04:07
mattoliverau\o/04:07
notmynameok, time for me to do other things this evening04:07
notmynamehave a good evening/afternoon (arvo)/night04:07
mattoliveraunotmyname: kk, have fun, and thanks for your dashboard work :)04:08
mattoliveraunotmyname: I see what you did there ;)04:08
mahaticlol, quick review notmyname . yup, nice work on dashboard04:08
*** duobei has joined #openstack-swift04:08
*** dmorita_ has quit IRC04:09
*** lakshmiS has joined #openstack-swift04:12
*** klrmn has quit IRC04:16
*** psachin has joined #openstack-swift04:17
*** duobei has quit IRC04:22
jrichlinotmyname: dashboard looks great!  nice additions04:23
*** ACCESSDEN1ED has joined #openstack-swift04:27
*** delatte has quit IRC04:27
ACCESSDEN1EDHello04:27
*** thumpba has joined #openstack-swift04:27
*** ACCESSDEN1ED has left #openstack-swift04:27
*** apparition has joined #openstack-swift04:29
*** eranrom has quit IRC04:31
*** thumpba has quit IRC04:32
jrichlimattoliverau: I am getting a unittest failure with the last fallocate patch: test_fallocate_reserve (test.unit.common.test_utils.TestUtils)04:37
mattoliveraujrichli: oh thanks, let me check that :)04:37
mattoliveraujrichli: I was being lazy and waiting for zuul to come back :P04:37
*** lakshmiS has quit IRC04:45
*** sams-gleb has joined #openstack-swift04:46
openstackgerritMatthew Oliver proposed openstack/swift: disable_fallocate also disables fallocate_reserve  https://review.openstack.org/31715804:50
mattoliveraujrichli: ^^ there you go. thanks :)04:50
*** SkyRocknRoll has joined #openstack-swift04:52
*** nadeem has joined #openstack-swift04:53
*** sams-gleb has quit IRC04:54
*** SkyRocknRoll has quit IRC04:58
*** SkyRocknRoll has joined #openstack-swift05:02
*** nadeem has quit IRC05:04
*** SkyRocknRoll has quit IRC05:04
*** SkyRocknRoll has joined #openstack-swift05:05
*** nadeem has joined #openstack-swift05:10
openstackgerritMatthew Oliver proposed openstack/swift: disable_fallocate also disables fallocate_reserve  https://review.openstack.org/31715805:13
*** ChubYann has quit IRC05:28
*** ppai has joined #openstack-swift05:33
*** sanchitmalhotra1 has joined #openstack-swift05:38
*** sanchitmalhotra has quit IRC05:40
*** geaaru has joined #openstack-swift05:47
*** duobei has joined #openstack-swift05:49
*** sams-gleb has joined #openstack-swift05:51
*** SkyRocknRoll has quit IRC05:53
*** sams-gleb has quit IRC05:58
*** rcernin has joined #openstack-swift06:02
*** eranrom has joined #openstack-swift06:03
*** Jeffrey4l has quit IRC06:04
*** dmorita has joined #openstack-swift06:09
*** dmorita has quit IRC06:13
*** rcernin has quit IRC06:18
*** rcernin has joined #openstack-swift06:21
*** bill_az has quit IRC06:26
*** thumpba has joined #openstack-swift06:28
*** cdelatte has joined #openstack-swift06:30
*** apparition has quit IRC06:32
*** thumpba has quit IRC06:33
*** ppai has joined #openstack-swift06:33
*** jamielennox is now known as jamielennox|away06:41
*** sams-gleb has joined #openstack-swift06:55
*** sams-gleb has quit IRC06:57
*** duobei has quit IRC06:57
*** sams-gleb has joined #openstack-swift06:57
*** thumpba has joined #openstack-swift07:00
*** sams-gleb has quit IRC07:01
*** sams-gleb has joined #openstack-swift07:01
*** duobei has joined #openstack-swift07:05
*** thumpba has quit IRC07:06
*** ozialien10 has quit IRC07:07
*** rledisez has joined #openstack-swift07:16
*** duobei has quit IRC07:26
*** mariusv has quit IRC07:30
*** mariusv has joined #openstack-swift07:31
*** mariusv has joined #openstack-swift07:31
*** jmccarthy has quit IRC07:40
*** jmccarthy has joined #openstack-swift07:41
*** ozialien10 has joined #openstack-swift07:52
openstackgerritKazuhiro MIYAHARA proposed openstack/swift: WIP: Swift Automated Tiering  https://review.openstack.org/28705707:56
*** ozialien10 has quit IRC07:57
*** ozialien10 has joined #openstack-swift07:58
*** duobei has joined #openstack-swift07:58
*** acoles_ is now known as acoles08:00
openstackgerritMerged openstack/swift: Bypass Account Mgmt Functional Cases When Disabled  https://review.openstack.org/31630308:02
*** admin6 has joined #openstack-swift08:04
*** nadeem has quit IRC08:05
*** dmk0202 has joined #openstack-swift08:12
*** Jeffrey4l has joined #openstack-swift08:13
*** joeljwright has joined #openstack-swift08:13
*** ChanServ sets mode: +v joeljwright08:13
acolesnotmyname: jrichli: fwiw I think it might be possible to have an error raised if the encryption root secret is not changed and also configure test scenarios with a non-default secret.08:18
*** jistr has joined #openstack-swift08:19
acolesthe in process tests could have a config option added. devstack might need a patch, if we were to add crypto to the default pipeline.08:20
acolesnotmyname: were you thinking that crypto would be in default sample pipeline AND also have a requirement to change the default secret option? that could prove frustrating.08:21
*** asettle has joined #openstack-swift08:24
openstackgerritAlistair Coles proposed openstack/swift: Add Encryption to the default pipeline  https://review.openstack.org/31641908:24
acolesjrichli: another attempt to pass in the gate! ^^08:24
acolesjrichli: I actually prefer changing the MAX_HEADER_COUNT in constraints.py because it avoids things *sometimes* breaking if user forgets to set extra_header_count, abd AFAICT there is no cost to increasing the value.08:26
*** NM has joined #openstack-swift08:27
*** NM has quit IRC08:29
*** NM has joined #openstack-swift08:30
*** admin6 has quit IRC08:33
*** mvk has joined #openstack-swift08:33
*** NM has quit IRC08:36
*** NM has joined #openstack-swift08:42
*** janonymous has quit IRC08:50
*** Jeffrey4l has quit IRC08:51
mahaticacoles: +1 on chaning in constraints.py. It is more in our control than depending on user08:57
mahaticchanging*08:57
acolesmahatic: and also, it might work for in process tests ;) since my last attempt failed!08:57
mahaticacoles: oic :)08:58
acolesbut yes, I think it's better than requiring a config option to be set, particularly when the consequence of it not being set is only apparent when large numbers of metadata headers are sent. makes for intermittent errors.08:59
acolesmahatic: just left you a review for disable_encryption - I like the tests!09:00
acolesmaha did you see my probe test patch 316924 yesterday? I'm afraid it may mean that the crypto-override header is no longer needed.09:01
patchbotacoles: https://review.openstack.org/#/c/316924/ - swift (feature/crypto) - crypto - make some probe tests compatible with crypto09:01
acolesmahatic: ^^09:01
*** ouchkernel has quit IRC09:02
*** thumpba has joined #openstack-swift09:02
acolesI was uneasy with the crypto-override header because it could be used by clients, and i don't think we necessarily want that, at least not by default, not until we have a use case for it.09:02
openstackgerritvenkatamahesh proposed openstack/python-swiftclient: Update the home-page with developer documentation  https://review.openstack.org/31732809:03
acolesmahatic: oh i just saw your review. NM, thanks!09:04
mahaticacoles: yup, good call on probe tests, you're smart!09:04
mahaticacoles: the remove byok patch should be good as is i suppose - i.e. w/o override filtering in keymaster09:05
mahaticacoles: also, the test infrastructure you've put in is coming in very handy, thanks!. I wrote a test for COPY and the output was: if I copy an unencrypted object, it is copied encrypted09:07
*** ouchkernel has joined #openstack-swift09:07
*** Shashikant86 has joined #openstack-swift09:07
*** thumpba has quit IRC09:08
acolesmahatic: yeah the first of your "remove byok" patches is probably sufficient now (ie takes out crypt override too)09:09
acolesmahatic: oh nice, so you added copy to the test app pipeline?09:10
*** jordanP has joined #openstack-swift09:10
acolesmahatic: do you think this task can be changed to not require a probe test, as per my comment on trello card? https://trello.com/c/Akvnj76R/10-probetest-put-object-verify-stored-object-is-encrypted-verify-container-listing-metadata-is-consistent-and-encrypted-with-contai09:16
mahaticacoles: sure, I will do that09:21
mahaticacoles: re COPY - my current tests seems off (checking). But, i have a question - the expected output is that when we copy an uncrypted obj, it should be copied unencrypted right?09:22
acolesmahatic: so I think the task reduces to find the diskfile created in this test https://review.openstack.org/#/c/314740/4/test/unit/common/middleware/test_encrypter_decrypter.py@170 and check it is encrypted09:23
patchbotacoles: patch 314740 - swift (feature/crypto) - crypto - don't break conditional GETs on unencrypt...09:23
mahaticacoles: oh sorry about the probe tests reply - i still didnt check the test infra to modify it to examine on disk changes, can i change it once i do that? will ping/post there today09:23
acolesmahatic: re copy, no - when an object is copied then destination object should be encrypted even if source object was not09:25
acolesmahatic: re probe test task, sure, I was just trying to gauge how much work was on trello and I'm hoping that task is easier than it sounds!09:25
acolesmahatic: I am picking up https://trello.com/c/0WQFODPn/84-add-probe-test-to-test-object-async-update-py-that-sets-internal-client-to-use-fast-post-and-then-performs-a-post-and-checks-the09:26
mahaticacoles: yeah i thought that was the tricky part - to find the diskfile created. but maybe it's not. i need to acess the nodes by pasing in acc/cont/obj info09:27
mahaticacoles: oh! (on copy) https://trello.com/c/lBP7wpZa/68-change-keymaster-to-detect-if-a-put-is-really-from-a-copy -> has a comment from jrichli :"In this case, we don't want to encrypt an object that was not previously encrypted"09:29
*** dmk0202 has quit IRC09:31
mahaticdid i read it wrong?09:32
*** mvk has quit IRC09:35
*** Shashikant86 has quit IRC09:36
*** mvk has joined #openstack-swift09:37
acolesmahatic: that's what it says but I think it should be referring to post-as-copy, not a normal copy.09:38
acolesplus in Austin I heard some opinion that even a post-as-copy should encrypt an unencrypted source, I'm not so sure myself.09:39
acolesI changed the trello card title09:39
*** janonymous has joined #openstack-swift09:42
*** NM has quit IRC09:47
*** Shashikant86 has joined #openstack-swift09:52
*** early has quit IRC09:56
*** duobei has quit IRC10:02
*** duobei has joined #openstack-swift10:03
*** Jeffrey4l has joined #openstack-swift10:15
*** early has joined #openstack-swift10:19
*** silor has joined #openstack-swift10:29
*** Trixboxer has quit IRC10:30
*** silor has quit IRC10:36
*** silor has joined #openstack-swift10:37
*** Shashikant86 has quit IRC10:39
*** duobei has quit IRC10:39
*** mmcardle has quit IRC10:40
*** mmcardle has joined #openstack-swift10:40
*** daemontool has quit IRC10:41
*** jamie_h_ has joined #openstack-swift10:43
*** early has quit IRC10:43
*** jamie_h_ has quit IRC10:43
openstackgerritOpenStack Proposal Bot proposed openstack/swift: Updated from global requirements  https://review.openstack.org/8873610:44
*** daemontool has joined #openstack-swift10:45
*** early has joined #openstack-swift10:48
*** jordanP has quit IRC10:49
*** Shashikant86 has joined #openstack-swift10:53
*** jamielennox|away is now known as jamielennox10:53
*** Shashikant86 has quit IRC10:55
*** Jeffrey4l has quit IRC11:02
*** thumpba has joined #openstack-swift11:04
*** thumpba has quit IRC11:08
*** dmorita has joined #openstack-swift11:09
*** dmorita has quit IRC11:14
*** Jeffrey4l has joined #openstack-swift11:14
*** sams-gleb has quit IRC11:16
*** sams-gleb has joined #openstack-swift11:17
*** sanchitmalhotra1 has quit IRC11:24
*** jistr has quit IRC11:24
*** sanchitmalhotra has joined #openstack-swift11:24
*** sams-gleb has quit IRC11:25
*** sanchitmalhotra has quit IRC11:31
*** sanchitmalhotra has joined #openstack-swift11:31
*** diogogmt has quit IRC11:35
*** diogogmt has joined #openstack-swift11:37
*** Jeffrey4l has quit IRC11:37
*** sams-gleb has joined #openstack-swift11:39
*** sanchitmalhotra has quit IRC11:41
*** sanchitmalhotra has joined #openstack-swift11:42
*** Jeffrey4l has joined #openstack-swift11:51
*** mmcardle has quit IRC11:52
*** dmk0202 has joined #openstack-swift11:52
*** diogogmt has quit IRC11:54
*** kei_yama has quit IRC11:58
*** janonymous has quit IRC11:58
*** jistr has joined #openstack-swift12:00
*** duobei has joined #openstack-swift12:07
*** sanchitmalhotra1 has joined #openstack-swift12:09
*** sanchitmalhotra has quit IRC12:09
*** ppai has quit IRC12:14
*** duobei has quit IRC12:15
*** raildo-afk is now known as raildo12:15
*** rickyrem has joined #openstack-swift12:16
*** admin6 has joined #openstack-swift12:17
*** mmcardle has joined #openstack-swift12:23
*** Jeffrey4l has quit IRC12:34
*** NM has joined #openstack-swift12:35
*** ppai has joined #openstack-swift12:37
*** eranrom has quit IRC12:40
*** eranrom has joined #openstack-swift12:41
*** sheel has quit IRC12:45
*** CaioBrentano has joined #openstack-swift12:45
*** ppai has quit IRC12:47
*** vint_bra has joined #openstack-swift13:02
*** hseipp has joined #openstack-swift13:03
*** pauloewerton has joined #openstack-swift13:04
*** vint_bra has quit IRC13:07
pdardeaugood morning13:07
acolespdardeau: hi13:09
*** janonymous has joined #openstack-swift13:10
*** rvasilets__ has joined #openstack-swift13:16
*** jordanP has joined #openstack-swift13:18
*** StraubTW has joined #openstack-swift13:19
pdardeauacoles: o/13:20
*** klamath has joined #openstack-swift13:21
*** klamath has quit IRC13:21
*** klamath has joined #openstack-swift13:21
*** psachin has quit IRC13:24
*** diogogmt has joined #openstack-swift13:26
*** bill_az has joined #openstack-swift13:29
*** openstackgerrit has quit IRC13:32
*** diogogmt has quit IRC13:33
*** openstackgerrit has joined #openstack-swift13:34
*** links has quit IRC13:36
*** sams-gleb has quit IRC13:48
*** psachin has joined #openstack-swift13:48
*** sgundur1 has joined #openstack-swift13:49
*** sams-gleb has joined #openstack-swift13:49
*** sams-gleb has quit IRC13:49
*** sams-gleb has joined #openstack-swift13:49
*** _JZ_ has joined #openstack-swift13:51
*** NM1 has joined #openstack-swift13:53
*** NM has quit IRC13:53
*** jasongalyon has joined #openstack-swift13:56
*** geaaru has quit IRC13:57
*** ametts has joined #openstack-swift13:58
*** geaaru has joined #openstack-swift13:59
openstackgerritAlistair Coles proposed openstack/swift: Send correct size in POST async update for EC object  https://review.openstack.org/31747514:00
*** sgundur- has left #openstack-swift14:01
*** sgundur1 has left #openstack-swift14:01
acolestorgomatic: ^^ as discussed in Austin. I have what I hope is a cleaner approach on feature/crypto, patch 305995, but that's too late for the EC overrides14:02
patchbotacoles: https://review.openstack.org/#/c/305995/ - swift (feature/crypto) - crypto - Update container with correct etag on POST (MERGED)14:02
*** jordanP has quit IRC14:03
*** sgundur- has joined #openstack-swift14:06
*** sgundur1 has joined #openstack-swift14:06
*** sgundur1 has quit IRC14:06
*** sgundur- has left #openstack-swift14:06
*** sgundur- has joined #openstack-swift14:06
*** StraubTW has quit IRC14:07
*** StraubTW has joined #openstack-swift14:07
*** daemontool has quit IRC14:09
*** acoles is now known as acoles_14:11
*** acoles_ is now known as acoles14:12
*** jordanP has joined #openstack-swift14:24
*** ma9 has joined #openstack-swift14:24
*** vint_bra has joined #openstack-swift14:30
*** eeiden has joined #openstack-swift14:30
jrichliacoles: you are right - having encryption means you already needed to modify the pipeline from default.  In that case, we don't need a default root secret in the proxy config.  Just in a test config. is that correct?14:36
jrichliacoles mahatic: re MAX_HEADER_COUNT - I have always prefered changing in constraints.py - so I am fine with that!14:36
jrichliacoles mahatic: I see how it is debatable whether a post-as-copy results in a possible encrypted body - where there was not before.14:37
acolesjrichli: patch 316419 is now passing, so give it a review and maybe we can merge it14:37
patchbotacoles: https://review.openstack.org/#/c/316419/ - swift (feature/crypto) - Add Encryption to the default pipeline14:37
jrichliOne thing to think about - encrypting would make the post-as-copy act differently than fast-post in that same situation.14:38
glangerunning the functional tests take a long, long time now :)14:38
glange{0} test.functional.tests.TestContainerPaths.testContainerListing [293.806183s] ... ok14:38
glangethat one takes 5 minutes, for example14:38
acolesjrichli: ^^ yes that's my concern with post-as-copy, would prefer identical behaviour wrt fast post14:39
sgundur-glange: I just ran functests , results -http://paste.openstack.org/show/497358/14:43
acolesglange: {0} test.functional.tests.TestContainerPaths.testContainerListing [2.233507s]14:43
sgundur-glange: similar time as acoles14:44
acolesglange: sometimes the test setup/teardown is slow if there's lots of garbage in the test account (but we may even have improved that)14:45
glangeah, ok, so it's wrong to blame whatever test happens to catch this cleanup stuff14:45
glangeit took 30 minutes total for all the tests, with 2 failures on a fresh SAIO and brand new virtual server14:46
glangeone of the failures looks like the test just timed out and got killed14:46
glangenot sure though14:47
openstackgerritJanie Richling proposed openstack/swift: disable_fallocate also disables fallocate_reserve  https://review.openstack.org/31715814:52
jrichli^^ just fixed pep814:53
*** diogogmt has joined #openstack-swift14:56
*** Jeffrey4l has joined #openstack-swift15:06
*** sams-gleb has quit IRC15:06
*** sams-gleb has joined #openstack-swift15:07
*** sheel has joined #openstack-swift15:09
*** sams-gleb has quit IRC15:16
*** ma9 has quit IRC15:28
openstackgerritAnne Gentle proposed openstack/swift: Adds migrated API reference files  https://review.openstack.org/31231515:30
notmynamegood morning15:31
notmynameon this day exactly six years ago, Swift was put into production15:31
*** dmk0202 has quit IRC15:33
*** eranrom has quit IRC15:33
*** ppai has joined #openstack-swift15:33
*** ppai has quit IRC15:34
jrichli\o/ happy bday swift!15:34
joeljwright:)15:36
*** gyee has joined #openstack-swift15:36
*** jistr has quit IRC15:38
*** mvk has quit IRC15:39
*** lakshmiS has joined #openstack-swift15:46
*** lyrrad has joined #openstack-swift15:52
*** asettle has quit IRC15:55
*** sams-gleb has joined #openstack-swift16:12
*** acoles is now known as acoles_16:15
*** acoles_ is now known as acoles16:15
*** rledisez has quit IRC16:20
CaioBrentanocongratulation swift!! :)16:20
*** sams-gleb has quit IRC16:21
*** forcebanana has joined #openstack-swift16:22
openstackgerritBrian Ober proposed openstack/swift: Container Object Count Quota Functional Test  https://review.openstack.org/31409916:23
openstackgerritAlistair Coles proposed openstack/swift: crypto: Remove BYOK support  https://review.openstack.org/31485616:24
*** jordanP has quit IRC16:24
acolesjrichli: ^^ another simple-ish review, i just rebased it16:24
CaioBrentanoIf I set weight zero to a device, how long does it take to remove objects and free up space? "min_part_hours" hours?16:25
*** dmorita has joined #openstack-swift16:28
glangeCaioBrentano: it depends upon how fast your replication is - it's something you have to monitor and it's not something you can predict16:29
notmynameCaioBrentano: other way around. min_part_hours should be bigger than your replication cycle time16:30
ober37happy birthday!16:30
CaioBrentanothanks glange16:30
CaioBrentanonotmyname is there a way to measure my replication cycle time?16:31
glangehttp://docs.openstack.org/developer/swift/overview_ring.html <-- that page tells you want min_part_hours is16:31
notmynameCaioBrentano: it's in the logs16:31
glangeit may or may not be related to your actual replication speed16:31
CaioBrentanothanks!16:31
notmynameand there's some statsd metrics too16:33
CaioBrentanohow dangerous is to set a bigger min_part_hours?16:33
*** forcebanana has quit IRC16:34
glangethat controls how often a part can move, bigger means more time between moves16:34
glangebasically changing your rings too often can cause problems - you want to do it a rate that is slow enough to avoid those problems16:35
CaioBrentanoglange I got that! ;) but, based on what notmyname said, I should increase the min_part_hour. the replication cycle time is way bigger than min_part_hour16:35
glangeoh16:36
glangeyeah, I don't know what best practices are16:36
notmynameif you rebalance and deploy updated rings faster than your replication cycle time, you might have unavailability of data (and potentially slightly higher usage).16:37
glangeCaioBrentano: are you changing your ring to add capacity?16:39
CaioBrentanoglange last week I gradually added a new region. today I set weight zero to some devices on the region that I'll have to remove16:41
glangeok16:41
*** klrmn has joined #openstack-swift16:42
CaioBrentanoglange I set weight zero to those devices 5 hours ago... I'm seeing a lot of replications on the logs, but the disk usage space hasn't changed16:45
*** rickyrem has quit IRC16:48
glangewhat do the logs say?16:52
CaioBrentanoabout replication cycle time? 16 hours16:55
glangethere should be messages about what replication is doing and if there are errors16:56
glangeI think it also says things like "Removing X_ objects"16:56
glangereplicaiton might have completed in 16 hours but there might have been failures16:57
glangeand things that need to be fixed/cleared up16:57
*** dmorita has quit IRC16:57
*** dmorita has joined #openstack-swift16:57
*** bill_az has quit IRC16:58
*** bill_az has joined #openstack-swift16:59
*** nadeem has joined #openstack-swift17:00
*** gyee has quit IRC17:00
CaioBrentanoFound no message like "Removing"17:02
CaioBrentanojust tons of "object-replicator: <f+++++++++"17:02
CaioBrentanoand some messages because of this bug (https://bugs.launchpad.net/swift/+bug/1503152)17:04
openstackLaunchpad bug 1503152 in OpenStack Object Storage (swift) "Error syncing handoff partition : no attribute 'intersection" [Undecided,Fix committed] - Assigned to Charles Hsu (charles0126)17:04
*** bill_az has quit IRC17:04
*** gyee has joined #openstack-swift17:08
*** joeljwright has quit IRC17:09
*** eeiden has quit IRC17:10
*** vinsh has quit IRC17:11
*** vinsh has joined #openstack-swift17:11
*** sams-gleb has joined #openstack-swift17:18
*** hseipp has quit IRC17:23
*** wadeholler has joined #openstack-swift17:24
*** Jeffrey4l has quit IRC17:26
*** sams-gleb has quit IRC17:26
*** mmcardle has quit IRC17:28
*** psachin has quit IRC17:36
*** ouchkernel has quit IRC17:36
*** ouchkernel has joined #openstack-swift17:44
*** ChubYann has joined #openstack-swift17:53
*** jrichli has quit IRC17:58
*** lyrrad_ has joined #openstack-swift18:01
*** bill_az has joined #openstack-swift18:01
*** bill_az has quit IRC18:02
*** lyrrad has quit IRC18:02
*** lyrrad_ is now known as lyrrad18:02
*** jrichli has joined #openstack-swift18:02
notmynameno golang topic in the -infra meeting today. but the TC meeting will have a "technical benefits discussion" (https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee)18:21
notmynamethat starts in 1 hour 39 minutes18:21
notmynamein #openstack-meeting18:21
notmynameI'll be there, and if you're interested in the discussion then feel free to join18:21
*** sams-gleb has joined #openstack-swift18:22
notmyname#success exactly 6 years ago today, Swift was put into production18:26
openstackstatusnotmyname: Added success to Success page18:26
redboJun 28 is 7 years since the first commit was made to swift.18:30
*** mmcardle has joined #openstack-swift18:30
*** rcernin has quit IRC18:31
*** sams-gleb has quit IRC18:31
*** eranrom has joined #openstack-swift18:38
*** bill_az has joined #openstack-swift18:43
*** acoles is now known as acoles_18:53
tdasilvaanybody seen functional tests errors on master recently? I'm getting this on a "fresh" new saio: ==============================18:56
tdasilvaFailed 22 tests - output below:18:56
tdasilva==============================18:56
tdasilvatest.functional.tests.TestFile.testFileSizeLimit18:56
tdasilva------------------------------------------------18:56
tdasilvaCaptured traceback:18:57
tdasilva~~~~~~~~~~~~~~~~~~~18:57
tdasilva    Traceback (most recent call last):18:57
tdasilva      File "test/functional/tests.py", line 1871, in testFileSizeLimit18:57
tdasilva        cfg={'set_content_length': i}))18:57
tdasilva      File "test/functional/tests.py", line 1858, in timeout18:57
tdasilva        method(*args, **kwargs)18:57
tdasilva      File "test/functional/swift_test_client.py", line 981, in write18:57
tdasilva        self.conn.make_path(self.path))18:57
tdasilva    test.functional.swift_test_client.ResponseError: 503: 'Service Unavailable' ('PUT' '/v1/AUTH_test/ad56d4c4678a48969aacc38f1c647ccf/c8705e28fd2c476d82f4351ca2ee6378') txid=tx8dbd142e10d9410b908bd-00573b647918:57
tdasilva18:57
tdasilvatest.functional.tests.TestFile.testGetContentType18:57
tdasilva-------------------------------------------------18:57
tdasilvaCaptured traceback:18:57
tdasilva~~~~~~~~~~~~~~~~~~~18:57
tdasilva    Traceback (most recent call last):18:57
tdasilva      File "test/functional/tests.py", line 2036, in testGetContentType18:57
tdasilva        file_item.write_random()18:57
tdasilva      File "test/functional/swift_test_client.py", line 1003, in write_random18:57
tdasilva        if not self.write(data, hdrs=hdrs, parms=parms, cfg=cfg):18:57
tdasilva      File "test/functional/swift_test_client.py", line 981, in write18:57
tdasilva        self.conn.make_path(self.path))18:57
tdasilva    test.functional.swift_test_client.ResponseError: 503: 'Service Unavailable' ('PUT' '/v1/AUTH_test/ad56d4c4678a48969aacc38f1c647ccf/72225189db3c4a45aac5960a4e4cd5a4') txid=tx087838cb4b154fc59bc0f-00573b647918:57
tdasilva18:57
tdasilvatest.functional.tests.TestFile.testGetOnFileThatDoesNotExist18:57
tdasilva------------------------------------------------------------18:57
tdasilvaCaptured traceback:18:57
tdasilva~~~~~~~~~~~~~~~~~~~18:58
tdasilva    Traceback (most recent call last):18:58
tdasilva      File "test/functional/tests.py", line 2047, in testGetOnFileThatDoesNotExist18:58
tdasilva        self.assert_status(404)18:58
tdasilva      File "test/functional/tests.py", line 91, in assert_status18:58
tdasilva        (se18:58
tdasilvacrap18:58
tdasilvahttp://paste.fedoraproject.org/367706/51135914/18:58
* bkeller` golf clap18:58
*** ouchkernel has quit IRC18:59
timburketdasilva: seems similar to the fallocate issues jrichli, mmotiani, and others (mattoliverau, at least?) have seen19:03
*** aj701 has quit IRC19:03
timburkehave you checked for 507s in your logs?19:03
*** aj701 has joined #openstack-swift19:04
*** ouchkernel has joined #openstack-swift19:04
openstackgerritMerged openstack/python-swiftclient: Update the home-page with developer documentation  https://review.openstack.org/31732819:05
redboI'm getting some random failures too, but I haven't looked at it.19:12
tdasilvai'm testing patch 31715819:12
patchbottdasilva: https://review.openstack.org/#/c/317158/ - swift - disable_fallocate also disables fallocate_reserve19:12
*** lcurtis has joined #openstack-swift19:14
*** wadeholler has left #openstack-swift19:20
jrichlitdasilva: the 22 failures are what you got before testing that patch, right?19:25
tdasilvajrichli: yeah, and with the patch, tests pass19:25
jrichligood, good.19:25
*** sams-gleb has joined #openstack-swift19:28
*** jasongalyon has quit IRC19:31
*** jasongalyon has joined #openstack-swift19:31
*** gyee has quit IRC19:33
*** sams-gleb has quit IRC19:35
*** rvasilets__ has quit IRC19:36
*** rvasilets__ has joined #openstack-swift19:43
*** dmorita has quit IRC19:46
notmynameFYI there's a proposal to drop community voting for summit sessions. http://lists.openstack.org/pipermail/community/2016-May/001503.html19:48
*** silor has quit IRC19:48
notmynameso if you feel strongly one way or the other, there's the ML thread in which to do it :-)19:49
tdasilvanotmyname: as a track chair, how much weight do you put on votes when selecting talks?19:55
*** dmorita has joined #openstack-swift19:55
notmynamerelatively little, but I do normally look at it. I'd consider it a 3rd sort key :-)19:55
notmynametopic/content, who's presenting, community votes19:56
*** rcernin has joined #openstack-swift19:58
notmynameTC golang discussion now in -meeting, if you want to participate or lurk20:04
*** jasongalyon has quit IRC20:12
*** mmcardle has quit IRC20:12
*** sheel has quit IRC20:15
*** rickyrem has joined #openstack-swift20:22
*** dmk0202 has joined #openstack-swift20:29
*** sams-gleb has joined #openstack-swift20:32
pdardeauwow, watching the TC golang discussion. remember to show some PTL appreciation today! and maybe buy him a beer20:34
*** gyee has joined #openstack-swift20:40
*** sams-gleb has quit IRC20:41
*** rcernin has quit IRC20:41
notmynamewow. that took a left turn real quick and went where I didn't expect it20:47
redboI do intend to get the go proxy server to the point we can do a fair benchmark comparison.  But I also know "finishing" it would also be a herculean task that probably won't happen.20:47
*** raildo is now known as raildo-afk20:47
notmynameredbo: I know :-)20:48
notmynameredbo: and I also don't (yet) think having an official golang proxy server is a good idea. but let's have that conversation later. after we get the current stuff sussed out :-)20:49
jrichlinotmyname: which turn?  the talk about pluggability? about the dataplane?20:49
notmynamethe "maybe openstack shouldn't have dataplane projects in it" topic20:50
jrichliright20:50
jrichlii mean - not that that's right20:50
jrichlijust, gottcha20:50
notmynamefortunately, I was a little prepared. I had an hour-long phone call with someone yesterday on this topic. I didn't expect the TC meeting to go that way at all, but at least I was slightly prepared, mentally20:51
jrichliyou did good.20:53
jrichli:-)20:53
notmynamethanks20:55
redboI don't know, we don't entirely fit with the openstack vendorpalooza.20:57
*** jasongalyon has joined #openstack-swift20:57
*** joeljwright has joined #openstack-swift21:01
*** ChanServ sets mode: +v joeljwright21:01
*** joeljwright has quit IRC21:02
*** bill_az has quit IRC21:03
*** asettle has joined #openstack-swift21:04
*** bill_az has joined #openstack-swift21:05
*** klamath has quit IRC21:06
*** klamath has joined #openstack-swift21:09
*** dmk0202 has quit IRC21:10
*** pauloewerton has quit IRC21:17
*** joeljwright has joined #openstack-swift21:18
*** ChanServ sets mode: +v joeljwright21:18
*** daemontool has joined #openstack-swift21:20
*** wadeholl_ has joined #openstack-swift21:29
*** CaioBrentano has quit IRC21:30
*** jamielennox is now known as jamielennox|away21:30
*** NM1 has quit IRC21:35
*** lakshmiS has quit IRC21:35
*** sams-gleb has joined #openstack-swift21:38
*** jamielennox|away is now known as jamielennox21:39
*** wadeholl_ has quit IRC21:39
*** ametts has quit IRC21:41
*** geaaru has quit IRC21:42
mattoliveraumorning21:45
redboBut we were here first, we should spin all those non-dataplane projects out to another namespace.21:46
*** sams-gleb has quit IRC21:46
*** joeljwright1 has joined #openstack-swift21:50
pdardeaunotmyname: does http://docs.openstack.org/developer/swift/middleware.html#module-swift.common.middleware.copy satisfy lp bug 1521353?21:50
openstackLaunchpad bug 1521353 in OpenStack Object Storage (swift) "add server-side copy docs" [Wishlist,New] https://launchpad.net/bugs/152135321:50
notmynameredbo: lol21:50
pdardeaumattoliverau: morning!21:51
notmynamepdardeau: certainly looks like it21:51
*** joeljwright1 has quit IRC21:51
*** tqtran has joined #openstack-swift21:52
*** asettle has quit IRC21:53
*** jasongalyon has quit IRC21:53
*** joeljwright has quit IRC21:53
pdardeaunotmyname: updated it with comment and link21:53
notmynamethanks21:54
notmynamepdardeau: can you mark it as closed? if not, I can21:54
*** ametts has joined #openstack-swift21:54
pdardeaunotmyname: i'll try21:54
pdardeaunotmyname: i suspect i don't have authority b/c i don't see a way to mark closed21:55
notmynameok21:55
*** joeljwright has joined #openstack-swift21:58
*** ChanServ sets mode: +v joeljwright21:58
*** daemontool has quit IRC22:03
*** ozialien10 has quit IRC22:04
*** ozialien10 has joined #openstack-swift22:04
*** ametts has quit IRC22:07
*** aj701 has quit IRC22:16
*** nadeem has quit IRC22:18
*** nadeem has joined #openstack-swift22:19
*** aj701 has joined #openstack-swift22:20
*** StraubTW has quit IRC22:30
*** Jeffrey4l has joined #openstack-swift22:35
*** klamath has quit IRC22:41
*** sams-gleb has joined #openstack-swift22:43
*** sams-gleb has quit IRC22:50
notmynamefor those of you in texas: http://openstacktexas.com (and http://openstacktexas.com/call-for-speakers/ is open)22:52
*** diogogmt has quit IRC23:03
*** diogogmt has joined #openstack-swift23:05
*** lakshmiS has joined #openstack-swift23:08
openstackgerritMatthew Oliver proposed openstack/swift: disable_fallocate also disables fallocate_reserve  https://review.openstack.org/31715823:12
*** tmoreira has quit IRC23:17
*** cschwede has quit IRC23:18
*** lcurtis has quit IRC23:19
*** tmoreira has joined #openstack-swift23:21
*** cschwede has joined #openstack-swift23:22
*** kei_yama has joined #openstack-swift23:24
*** klrmn has quit IRC23:24
*** klrmn has joined #openstack-swift23:27
*** ouchkernel has quit IRC23:34
*** vint_bra has quit IRC23:36
*** hosanai has joined #openstack-swift23:45
*** ChanServ sets mode: +v hosanai23:45
*** sams-gleb has joined #openstack-swift23:47
*** joeljwright has quit IRC23:53
*** kevinc has joined #openstack-swift23:53
*** sams-gleb has quit IRC23:56

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!