Monday, 2016-06-20

jrichlinotmyname: I guess I picked a bad name for the etherpad.  by swift_md5_encryption, I meant the encryption of the  md5 / etag.  :-)00:46
jrichlinotmyname: the encryption algorithm is still AES-CTR.  There is some detail on the etherpad about md5 because the nature of the message being encrypted is a factor when discussing vulnerabilities.00:50
kota_morning01:29
kota_mattoliverau: welcome back to online :-)01:29
*** m_kazuhiro has joined #openstack-swift01:39
*** zaitcev has joined #openstack-swift01:42
*** ChanServ sets mode: +v zaitcev01:42
mattoliveraukota_: morning and thanks :)01:51
*** yarkot1 has joined #openstack-swift02:30
notmynamejrichli: ah, ok. but aren't we using md5 in the new crypto methods to derive the iv or something?02:48
jrichliYes, you will see an md5 calculated in _get_derived_iv in the case that the path is not long enough to use for an iv.  but since iv's are just normal values in the clear, it's ok to use here.02:54
jrichlithe problem with the iv is just that it will not necessary be unique.02:55
jrichlinotmyname ^^02:57
-openstackstatus- NOTICE: static.openstack.org (which hosts logs.openstack.org) is currently migrating due to a hardware failure. It should be back up shortly.03:11
*** sheel has joined #openstack-swift03:15
mahatic_good morning03:55
mahatic_jrichli: thanks for the writeup, it's informative03:56
*** links has joined #openstack-swift03:56
mahatic_jrichli: question - you mention of p1 and p1 in the etherpad, they are diff etag values of diff objects put at the same path, correct?04:01
mahatic_p1 and p2*04:01
kota_log.openstack.org has stalled? I got timeouts with my broweser when looking at the gate job logs.04:07
mattoliveraumahatic_: morning04:16
mattoliveraukota_: they're migrating the server due to a hardware failure04:17
kota_mattoliverau: got it, thanks!04:18
mahatic_mattoliverau: hello! welcome back04:24
*** takashi has joined #openstack-swift04:24
jrichlimahatic_ yes, the fact that the same keystream (k) is used is essentially what captures the fact that the path was the same.04:25
jrichli... since the keystream is determined by the iv and key, which are both derived from the path.04:27
*** klrmn has quit IRC04:29
* jrichli is gonna sleep now. g'night!04:29
mahatic_jrichli: got it, thanks. good night04:30
mattoliveraujrichli: night04:30
*** takashi has quit IRC04:47
*** ChubYann has quit IRC04:57
*** klamath has joined #openstack-swift05:00
*** klamath_ has quit IRC05:00
*** zaitcev has quit IRC05:06
*** links has quit IRC05:10
*** klamath has quit IRC05:11
*** nadeem has joined #openstack-swift05:17
-openstackstatus- NOTICE: static.openstack.org (which hosts logs.openstack.org and tarballs.openstack.org among others) is currently being rebuilt. As jobs can not upload logs they are failing with POST_FAILURE. This should be resolved soon. Please do not recheck until then.05:23
*** ChanServ changes topic to "static.openstack.org (which hosts logs.openstack.org and tarballs.openstack.org among others) is currently being rebuilt. As jobs can not upload logs they are failing with POST_FAILURE. This should be resolved soon. Please do not recheck until then."05:23
*** links has joined #openstack-swift05:25
*** ppai has joined #openstack-swift06:00
*** klamath has joined #openstack-swift06:12
*** rcernin has joined #openstack-swift06:13
*** klamath has quit IRC06:17
*** pcaruana has joined #openstack-swift06:24
*** nadeem has quit IRC06:25
*** SkyRocknRoll has joined #openstack-swift06:34
*** jmccarthy has quit IRC06:40
*** jmccarthy has joined #openstack-swift06:41
*** hseipp has joined #openstack-swift06:42
*** links has quit IRC07:08
*** admin6 has joined #openstack-swift07:12
*** rledisez has joined #openstack-swift07:14
*** klamath has joined #openstack-swift07:14
*** Jeffrey4l_ has quit IRC07:16
*** hseipp has quit IRC07:16
*** klamath has quit IRC07:18
*** links has joined #openstack-swift07:20
*** ozeri has joined #openstack-swift07:27
*** Jeffrey4l has joined #openstack-swift07:42
*** hseipp has joined #openstack-swift07:45
*** hseipp has quit IRC07:49
*** geaaru has joined #openstack-swift07:51
*** mvk_ has quit IRC08:00
*** mvk_ has joined #openstack-swift08:27
*** permalac has joined #openstack-swift08:33
*** clayg has quit IRC08:37
*** kei_yama has quit IRC08:40
*** ntt has quit IRC08:48
*** mvk_ has quit IRC08:56
*** acoles_ is now known as acoles09:01
*** mvk has joined #openstack-swift09:02
*** stantonnet_ has quit IRC09:03
acolesmattoliverau: welcome back, hope you're coping in the new place, sounds awful ("the beach house") :P09:05
*** ouchkernel has joined #openstack-swift09:07
*** SkyRocknRoll has quit IRC09:21
*** SkyRocknRoll has joined #openstack-swift09:46
*** shakamunyi has quit IRC09:48
*** shakamunyi has joined #openstack-swift09:56
admin6Hi there, In my object-replicator logs, I have constant failures such as : Jun 20 11:54:59 STACO2 object-replicator: 154354 successes, 376 failures   /   Jun 20 11:59:59 STACO2 object-replicator: 162334 successes, 376 failures  /  Jun 20 12:04:59 STACO2 object-replicator: 169894 successes, 376 failures.  Do you know where I could find more information about these failures?10:09
openstackgerritDavanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c  https://review.openstack.org/31844110:10
*** hosanai has quit IRC10:12
*** rledisez1 has joined #openstack-swift10:24
*** rledisez has quit IRC10:26
openstackgerritMahati Chamarthy proposed openstack/swift: remove empty db hash and suffix directories  https://review.openstack.org/33160110:28
*** m_kazuhiro has quit IRC10:29
*** rcernin is now known as rcernin|lunch10:37
*** ChanServ changes topic to "Time to review feature/crypto-review https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/crypto-review | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Hackathon: https://www.eventbrite.com/e/openstack-swift-july-2016-hackathon-tickets-25913773795"10:39
-openstackstatus- NOTICE: static.openstack.org is back up. If you have POST_FAILURE and are missing logs from your CI jobs, please leave a 'recheck'.10:39
*** nadeem has joined #openstack-swift10:44
*** nadeem has quit IRC10:48
*** nadeem has joined #openstack-swift10:49
*** hseipp has joined #openstack-swift11:19
*** daemontool has joined #openstack-swift11:22
*** ppai has quit IRC11:32
*** nadeem has quit IRC11:35
*** ppai has joined #openstack-swift11:45
*** rcernin|lunch is now known as rcernin11:53
*** ozeri has quit IRC11:54
*** raildo-afk is now known as raildo11:55
*** daemontool_ has joined #openstack-swift11:59
*** daemontool has quit IRC12:02
*** ozeri has joined #openstack-swift12:07
*** mlanner has quit IRC12:13
*** furlongm has quit IRC12:13
*** furlongm has joined #openstack-swift12:13
*** daemontool has joined #openstack-swift12:13
*** daemontool_ has quit IRC12:14
*** ouchkernel has quit IRC12:14
*** ekarlso has quit IRC12:15
*** links has quit IRC12:16
*** kaleta has quit IRC12:16
*** kaleta has joined #openstack-swift12:17
*** jlvillal has quit IRC12:18
*** ouchkernel has joined #openstack-swift12:18
*** timur_ has quit IRC12:18
*** hoonetorg has quit IRC12:19
*** timur has joined #openstack-swift12:19
*** jlvillal has joined #openstack-swift12:19
*** Jeffrey4l has quit IRC12:20
*** mlanner has joined #openstack-swift12:21
*** ppai has quit IRC12:26
*** links has joined #openstack-swift12:26
*** dosaboy has quit IRC12:26
*** Jeffrey4l has joined #openstack-swift12:27
*** pauloewerton has joined #openstack-swift12:28
*** dosaboy has joined #openstack-swift12:28
*** mvk has quit IRC12:28
*** hoonetorg has joined #openstack-swift12:29
*** ppai has joined #openstack-swift12:37
*** ekarlso has joined #openstack-swift12:59
*** permalac_ has joined #openstack-swift12:59
*** permalac has quit IRC12:59
*** links has quit IRC13:00
-openstackstatus- NOTICE: OpenID login from review.o.o is experiencing difficulties, possibly due to transatlantic network performance issues. Things are being investigated13:00
*** ChanServ changes topic to "OpenID login from review.o.o is experiencing difficulties, possibly due to transatlantic network performance issues. Things are being investigated"13:00
*** mvk has joined #openstack-swift13:11
*** ChanServ changes topic to "Time to review feature/crypto-review https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/crypto-review | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Hackathon: https://www.eventbrite.com/e/openstack-swift-july-2016-hackathon-tickets-25913773795"13:13
-openstackstatus- NOTICE: OpenID logins are back to normal13:13
*** klamath has joined #openstack-swift13:14
admin6Hi there, do you know what means Erasure coding logs : "object-reconstructor: Unexpected entity in data dir: u'/srv/node/s01z1ecd07/objects-1/auditor_status_ALL.json". Should I worry about them?13:14
*** joeljwright has joined #openstack-swift13:16
*** ChanServ sets mode: +v joeljwright13:16
mahatic_admin6: erasure coding is a storage policy supported in swift - http://docs.openstack.org/developer/swift/overview_erasure_code.html13:21
admin6mahatic_: Thanks, and the reconstructor is ckecking that all fragment are correct. But my worries is about what the "unexpected entity" means13:24
*** SkyRocknRoll has quit IRC13:26
mahatic_admin6: which swift release are you using? You shouldn't be worrying about that warning. It is fixed in a recent change - https://review.openstack.org/#/c/315334/13:29
patchbotmahatic_: patch 315334 - swift - Stop complaining about auditor_status files (MERGED)13:29
*** permalac__ has joined #openstack-swift13:30
admin6II’m using swift 2.7.013:30
*** permalac_ has quit IRC13:30
admin6mahatic_: thanks for your answer :-)13:32
*** hseipp has quit IRC13:33
admin6mahatic_: Btw, do you know where I could find more information about these other constant failures I have for my replica 3 storage policy : Jun 20 12:04:59 STACO2 object-replicator: 169894 successes, 376 failures.  I can’t find any more information about "what" fails.13:36
*** cdelatte has joined #openstack-swift13:51
mahatic_admin6: I think it indicates how many times the replication has failed , which will be taken care by replicator (i.e. replication will happen again with that same info). But you can confirm this with folks from the US, who have more expertise. They will be shortly online.13:56
mahatic_admin6: sorry I need to step away for now13:56
admin6mahatic_: thanks a lot13:56
*** zaitcev has joined #openstack-swift14:00
*** ChanServ sets mode: +v zaitcev14:00
raildonotmyname: ping, can you take a look on this patch? https://review.openstack.org/#/c/313659/8 I need some swift core's eyes here :)14:03
patchbotraildo: patch 313659 - openstack-infra/project-config - Run Swift functional tests in Identity v3-only14:03
raildonotmyname: I followed the nova approach that was already merged here: https://review.openstack.org/#/c/313588/14:03
patchbotraildo: patch 313588 - openstack-infra/project-config - Run novaclient functional tests in Identity v3-only (MERGED)14:03
*** sheel has quit IRC14:05
*** manous has joined #openstack-swift14:09
*** daemontool has quit IRC14:10
*** mwheckmann has joined #openstack-swift14:13
cschwedeadmin6: you don’t need to worry; it’s just a warning and has been fixed meanwhile: https://github.com/openstack/swift/commit/ad16e2c14:28
cschwedeadmin6: that is included in the 2.8 release14:29
*** hseipp has joined #openstack-swift14:29
*** hseipp has quit IRC14:32
admin6cschwede: ok, thanks14:36
*** zul_ is now known as zul14:40
*** ozeri has quit IRC14:56
*** ppai has quit IRC15:01
*** manous has quit IRC15:07
*** arch-nemesis has joined #openstack-swift15:07
timburkegood morning15:12
*** adu has joined #openstack-swift15:16
*** klrmn has joined #openstack-swift15:25
*** bapalm has joined #openstack-swift15:31
*** _JZ_ has joined #openstack-swift15:34
notmynamegood morning15:44
joeljwrightgood morning15:45
notmynamejoeljwright: it's not morning for you!15:45
joeljwrightnotmyname: true15:45
*** pcaruana has quit IRC15:45
joeljwrightnotmyname: you may want to comment on this - https://review.openstack.org/#/c/324024/15:46
patchbotjoeljwright: patch 324024 - oslosphinx - fix other versions sidebar links15:46
notmynamejoeljwright: yes. yes I do15:46
joeljwrightI implemented the taggerdate sorting suggestion, but I don't like it and I'm concerned this could end up going on forever15:46
notmynamejoeljwright: in openstack?! no way!15:47
joeljwright:D15:47
notmynameyeah, let's try to get the "fix what we have" before trying to make it perfect15:47
joeljwrightindeed15:47
joeljwrightit would be nice to drop the git tag lookup from this patch entirely15:48
*** tongli has joined #openstack-swift15:48
joeljwrightbecause the more I mess with it the more it looks like each project will have to provide their own tag lookup command15:48
notmynameeww15:49
*** gyee has joined #openstack-swift15:50
joeljwrightwith taggerdate and nova you get a mixture of 12.x.x.x, 13.x.x.x and 2015.x.x versions15:50
joeljwrightit's a mess15:50
*** dmorita has joined #openstack-swift15:50
joeljwrightswift just suffers from getting lists like [2.7.0, 2.6.0, 2.3.2, 2.5.0]15:51
joeljwrightwhich IMO looks revolting15:51
*** dmorita has quit IRC15:52
*** dmorita has joined #openstack-swift15:52
*** rcernin has quit IRC15:55
*** zul has quit IRC16:00
*** dmorita has quit IRC16:00
*** dmorita has joined #openstack-swift16:00
*** SkyRocknRoll has joined #openstack-swift16:03
notmynameI'll have to look at that (and understand why it's not ordered)16:07
timburkeis there a swift 2.3.2? i know there was a 2.2.2...16:09
notmynamegood point. IIRC, there hasn't been a x.y.2 in a long time16:10
timburkelooks like there was a 2.3.0rc2...16:10
timburkebut that's not the same thing16:10
notmynamethere was for swiftclient16:10
timburkeah, yeah, makes sense. and it must've been on a stable branch, explaining why it was tagged last august while 2.5.0 was tagged last july16:12
timburkejoeljwright: maybe we just need a way to pick up on which release cycle the tag was for? although that may run up against issues for other projects; see http://lists.openstack.org/pipermail/openstack-dev/2016-June/096895.html and their tagging shenanigans16:14
*** nadeem has joined #openstack-swift16:14
*** daemontool has joined #openstack-swift16:14
notmynameactually, that's not an "other projects" issue. we have the same thing16:14
notmyname(however, swift doesn't have the YYYY.X versions pattern)16:15
notmynamefor more detail, it only a problem for us historically. not any more. jsut weirdness with the old milestone-proposed branch and a commit there being tagged16:22
*** adu has quit IRC16:24
acolesnotmyname: I made some progress understanding the func test failure we were looking at on Friday - on master the proxy server raises a 499 client disconnect when each iteration of the test PUT is terminated with no data sent, and the objects servers also raise a 499 pretty much immediately. On the review branch the object servers are not raising the 499 and then 60 secs later raise 408 read timeout, and i think that durin16:26
acolesg that 60 sec period the fallocate bytes are perhaps pinned, hence the 507 on subsequent attempts to fallocate.16:26
acolesnotmyname: EC on master exhibits similar behaviour (i.e. obj server waits then 408s after 60 secs) but not every time ?!?16:27
notmynameinteresting16:27
acolesnotmyname: so far I have failed in seeing where the difference is in the two PUT paths16:28
*** lyrrad has joined #openstack-swift16:29
notmynameI had a thought this morning, too, but I haven't had a chance to investigate yet. the patch that was failing for me raises the extra headers default from 32 to 132. that's the sort of thing that could account for the extra bytes that are causing my fallocate()s to fail16:29
*** klrmn has quit IRC16:29
notmynamebut your discovery about the differences of the object disconnect seems closer to the root of any problem16:29
acolesnotmyname: BTW, the mode of the fallocate means that the tmp file size is not increased by the fallocate, so once the file is closed, with no bytes written, it will have zero size...which is I think how the fallocate bytes are "freed". so at least that makes more sense.16:30
joeljwrighttimburke: yeah, I was thinking about how to do it (and I may have mixed up the version numbers a bit)16:30
notmynameso it's like the mime handling code patch adds in a wait? but the patch isn't doing anything with the replicated mime PUT path yet, right?16:30
notmynameacoles: oh, interesting16:31
acolesnotmyname: I'm working on review 328204 branch where we have not yet increased the default max headers16:31
notmynameah ok16:31
joeljwrighttimburke: the real problem is that we want to fix the links to previous versions, but reviewers are expanding the discussion16:31
*** mvk has quit IRC16:33
acolesyeah, this isn't using the MIMEPutter - my hunch is that whatever causes the object server to get a read error and 499 is not happening, but afaict the proxy putter is closing the HTTP connection without sending bytes, just like on master, its just the obj server sits there tryihng to read for 60 secs :/16:33
notmynamehmmm16:34
acolesnotmyname: so, I'm still digging, that was just an update.16:34
notmynamethanks16:34
acolesjrichli: thanks for the etherpad write up16:34
*** rledisez1 has quit IRC16:35
jrichliacoles: I heard cca wrote some updates.  i still need to look at them.  been busy this am16:35
acolesnotmyname: England match in Euros today so i don't plan to be hear past 9pm like Friday ;)16:36
*** Suyash has joined #openstack-swift16:36
acolesjrichli: yes, I likewise haven't studied them much yet due to ^^16:36
acoles^^ being the func test issue, not the footie16:36
notmynameacoles: hope they do well! I watches the ARG v VEN game this weekend (and missed the 7-0 blowout where MEX got destroyed)16:37
acolesnotmyname: ouch did I really type 'hear' ?!? :(16:37
acolesnotmyname: heh, copa is getting little coverage over here, I missed that result16:38
notmynameI get all the copa games at home, but I can't watch the euro ones16:39
jrichliacoles: understood.  sounds like you are making good progress on that issue.  :-)16:41
tdasilvanotmyname: us x arg should be fun to watch16:43
acolesI am not planning to push any revised patches to crypto-review today, I'll wait to see if any more comments accumulate.16:45
*** dmorita has quit IRC16:47
notmynametdasilva: "fun". right. USA has several players out (on cards!) and just in general can't do important things like passing. or playing together as a team.16:49
notmynametdasilva: but as a fan of ARG and a citizen if USA, I'll be happy whoever wins *cough*ARG*cough*16:49
tdasilvanotmyname: well, then I guess it will be a "fun" night for messi16:49
*** dmorita has joined #openstack-swift16:50
notmynameacoles: ok, thanks. I'll keep going through the patch chain today16:50
*** rcernin has joined #openstack-swift16:53
notmynametdasilva: but against ven, arg didn't have a very good defense. I'll be impressed if usa gets one goal, but I'm completely expecting arg to easily win16:53
*** joeljwright has quit IRC16:54
*** dmorita has quit IRC16:58
*** dmorita has joined #openstack-swift16:59
*** ChubYann has joined #openstack-swift17:05
*** rcernin has quit IRC17:18
*** chsc has joined #openstack-swift17:20
*** chsc has joined #openstack-swift17:20
*** tqtran has joined #openstack-swift17:29
*** tongli has quit IRC17:32
*** klrmn has joined #openstack-swift17:38
*** mvk has joined #openstack-swift17:38
*** ametts has joined #openstack-swift17:39
*** nadeem has quit IRC17:49
*** mvk_ has joined #openstack-swift17:50
*** rcernin has joined #openstack-swift17:52
*** mvk has quit IRC17:53
*** mvk has joined #openstack-swift17:54
*** mkrcmari__ has joined #openstack-swift17:55
*** mvk_ has quit IRC17:57
*** mvk has quit IRC17:57
*** mkrcmari__ has quit IRC18:01
*** daemontool has quit IRC18:06
*** mkrcmari__ has joined #openstack-swift18:06
*** geaaru has quit IRC18:08
*** shakamunyi has quit IRC18:17
*** SkyRocknRoll has quit IRC18:21
*** zul has joined #openstack-swift18:53
*** zul has quit IRC18:54
*** zul has joined #openstack-swift18:54
*** acoles is now known as acoles_19:11
*** Jeffrey4l_ has joined #openstack-swift19:32
openstackgerritMohit Motiani proposed openstack/python-swiftclient: Insufficent info about read and write ACL syntax  https://review.openstack.org/32473919:33
*** Jeffrey4l has quit IRC19:33
*** gyee has quit IRC19:45
*** nadeem has joined #openstack-swift20:06
*** mvk_ has joined #openstack-swift20:12
*** mvk_ has quit IRC20:13
*** mvk_ has joined #openstack-swift20:14
*** mkrcmari__ has quit IRC20:15
*** zul has quit IRC20:18
*** zul has joined #openstack-swift20:22
*** awelleck has joined #openstack-swift20:26
*** nadeem has quit IRC20:31
*** rcernin has quit IRC20:31
*** nadeem has joined #openstack-swift20:31
*** gyee has joined #openstack-swift20:33
*** daemontool has joined #openstack-swift20:38
*** zul has quit IRC20:39
*** ouchkernel has quit IRC20:44
*** ouchkernel has joined #openstack-swift20:49
*** daemontool has quit IRC20:49
*** zul has joined #openstack-swift20:56
*** Anticime1 is now known as Anticimex20:57
*** raildo is now known as raildo-afk21:02
*** pauloewerton has quit IRC21:10
notmynameacoles_: FWIW, I briefly chatted with torgomatic_ about the functest failures. he's got an idea, and is also working on a diff for the Putters21:11
*** mwheckmann has quit IRC21:12
*** zul has quit IRC21:14
*** zul has joined #openstack-swift21:16
*** zul has quit IRC21:25
torgomatic_note that the idea and the diff are two completely different things; don't expect any of my futzing with the putters to fix the functests21:26
notmyname;-)21:26
notmynametorgomatic_: I've just filed it all away in the "sam will save us" bucket21:27
torgomatic_notmyname: that's one of those metal buckets with the handles and the holes in the bottom, right?21:27
*** zul has joined #openstack-swift21:31
*** ouchkernel has quit IRC21:58
*** nadeem has quit IRC22:02
*** zul has quit IRC22:03
*** ouchkernel has joined #openstack-swift22:05
*** clayg has joined #openstack-swift22:06
*** ChanServ sets mode: +v clayg22:06
*** clayg has quit IRC22:06
*** clayg has joined #openstack-swift22:06
*** ChanServ sets mode: +v clayg22:06
*** awelleck has quit IRC22:11
mattoliverauMorning22:25
*** ametts has quit IRC22:27
*** nadeem has joined #openstack-swift22:28
*** nadeem has quit IRC22:29
*** nadeem has joined #openstack-swift22:30
*** nadeem has quit IRC22:46
*** gyee has quit IRC22:58
*** kei_yama has joined #openstack-swift22:58
*** chsc has quit IRC23:43
*** chlong has joined #openstack-swift23:51
*** chlong is now known as chlong|rhce_trng23:52

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