| *** StraubTW has quit IRC | 00:04 | |
| *** Suyi has quit IRC | 00:23 | |
| *** Jeffrey4l has quit IRC | 00:24 | |
| *** elambert_ has quit IRC | 00:29 | |
| *** vint_bra has joined #openstack-swift | 00:30 | |
| *** vint_bra has quit IRC | 00:30 | |
| *** vint_bra has joined #openstack-swift | 00:32 | |
| *** vint_bra has quit IRC | 00:34 | |
| *** elambert_ has joined #openstack-swift | 00:35 | |
| *** lyrrad has quit IRC | 00:43 | |
| notmyname | patch 326771 is for the 2.8.0 release | 00:45 |
|---|---|---|
| patchbot | notmyname: https://review.openstack.org/#/c/326771/ - releases - swift 2.8.0 release | 00:45 |
| openstackgerrit | John Dickinson proposed openstack/swift: Merge branch 'master' into hummingbird https://review.openstack.org/326772 | 00:46 |
| notmyname | redbo: ^^ | 00:46 |
| *** nadeem has quit IRC | 01:01 | |
| *** tqtran has quit IRC | 01:16 | |
| *** StraubTW has joined #openstack-swift | 01:21 | |
| *** diogogmt has joined #openstack-swift | 01:24 | |
| *** nadeem has joined #openstack-swift | 01:48 | |
| openstackgerrit | Merged openstack/swift: Removed whitespaces from swift-ring-builder manpage https://review.openstack.org/326703 | 01:53 |
| *** elambert_ has quit IRC | 02:09 | |
| *** tanee is now known as tanee_away | 02:16 | |
| *** jrichli has joined #openstack-swift | 02:16 | |
| *** tanee_away is now known as tanee | 02:16 | |
| *** Suyi has joined #openstack-swift | 02:22 | |
| *** klamath_ has joined #openstack-swift | 02:25 | |
| *** klamath has quit IRC | 02:25 | |
| *** nadeem has quit IRC | 02:38 | |
| *** openstackgerrit has quit IRC | 02:47 | |
| *** openstackgerrit has joined #openstack-swift | 02:48 | |
| *** sheel has joined #openstack-swift | 02:49 | |
| *** gyee has quit IRC | 02:50 | |
| *** NM1 has joined #openstack-swift | 03:03 | |
| *** NM1 has quit IRC | 03:03 | |
| *** NM__ has joined #openstack-swift | 03:05 | |
| *** NM__ has quit IRC | 03:09 | |
| *** vint_bra has joined #openstack-swift | 03:09 | |
| *** vint_bra has quit IRC | 03:27 | |
| *** diogogmt has quit IRC | 03:32 | |
| *** links has joined #openstack-swift | 03:35 | |
| *** StraubTW has quit IRC | 03:35 | |
| *** dmorita has quit IRC | 03:36 | |
| *** dmorita has joined #openstack-swift | 03:38 | |
| *** dmorita_ has joined #openstack-swift | 03:39 | |
| *** furlongm has quit IRC | 03:41 | |
| *** furlongm has joined #openstack-swift | 03:42 | |
| *** dmorita has quit IRC | 03:42 | |
| *** vint_bra has joined #openstack-swift | 03:43 | |
| *** dmorita_ has quit IRC | 03:44 | |
| *** vint_bra has quit IRC | 03:46 | |
| *** klamath_ has quit IRC | 03:47 | |
| *** tanee is now known as tanee_away | 04:01 | |
| *** tanee_away is now known as tanee | 04:01 | |
| *** ppai has joined #openstack-swift | 04:05 | |
| openstackgerrit | Merged openstack/swift: Merge branch 'master' into hummingbird https://review.openstack.org/326772 | 04:06 |
| *** elambert_ has joined #openstack-swift | 04:10 | |
| *** jrichli has quit IRC | 04:10 | |
| *** elambert_ has quit IRC | 04:11 | |
| *** psachin has joined #openstack-swift | 04:25 | |
| *** jith_ has joined #openstack-swift | 04:29 | |
| jith_ | Hi all, i have a openstack setup with swift (two replication). I got the following error when i tried to upload 500 GB to swift.. The space available is 506 GB.. Can anyone pls guide?? | 04:32 |
| jith_ | http://paste.openstack.org/show/508826/ | 04:32 |
| *** garthb has joined #openstack-swift | 04:42 | |
| *** klamath has joined #openstack-swift | 04:48 | |
| *** garthb has quit IRC | 04:51 | |
| *** klamath has quit IRC | 04:52 | |
| *** zaitcev has quit IRC | 04:54 | |
| *** dmorita has joined #openstack-swift | 04:56 | |
| *** _JZ_ has quit IRC | 04:57 | |
| *** ChubYann has quit IRC | 04:59 | |
| *** dmorita has quit IRC | 05:01 | |
| *** nadeem has joined #openstack-swift | 05:08 | |
| mattoliverau | jith_: do you have any swift logs handy, that'll help. As for space available, what do you mean? You only have 506GB left on the cluster and your trying to put a 500GB file up or trying to put a 250GB file up (as 2x250 == 500). | 05:19 |
| jith_ | mattoliverau: no no i have 506 +506 GB free | 05:30 |
| mattoliverau | jith_: well if your trying to put a 506GB file up that will be stored as 2 replica's then you do't have enough space. Because yours asking swift to store 2x of your file in the cluster. | 05:37 |
| *** Suyi has quit IRC | 05:48 | |
| jith_ | mattoliverau: i have enough space | 05:59 |
| jith_ | i m trying to put 500 gb in 506 disk | 05:59 |
| jith_ | means my swift replication is with two partition... one partition is having 506gb and another having 506gb | 06:00 |
| jith_ | i could see it when i do df -h | 06:00 |
| jith_ | can u say anything on seeing this log...http://paste.openstack.org/show/508840/ | 06:01 |
| *** ppai has quit IRC | 06:03 | |
| *** rcernin has joined #openstack-swift | 06:12 | |
| *** lifeless has quit IRC | 06:13 | |
| *** klrmn has quit IRC | 06:15 | |
| *** nadeem has quit IRC | 06:35 | |
| *** ppai has joined #openstack-swift | 06:36 | |
| *** openstackgerrit has quit IRC | 06:48 | |
| *** openstackgerrit has joined #openstack-swift | 06:48 | |
| *** klamath has joined #openstack-swift | 06:49 | |
| *** klamath has quit IRC | 06:55 | |
| *** blair has joined #openstack-swift | 06:57 | |
| *** tesseract has joined #openstack-swift | 07:00 | |
| *** tanee is now known as tanee_away | 07:03 | |
| *** tanee_away is now known as tanee | 07:03 | |
| *** rledisez has joined #openstack-swift | 07:20 | |
| *** hseipp has joined #openstack-swift | 07:21 | |
| *** geaaru has joined #openstack-swift | 07:29 | |
| *** dmorita has joined #openstack-swift | 07:36 | |
| *** dmorita has quit IRC | 07:41 | |
| *** openstackgerrit has quit IRC | 07:48 | |
| *** openstackgerrit has joined #openstack-swift | 07:49 | |
| *** Jeffrey4l has joined #openstack-swift | 07:50 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c https://review.openstack.org/318441 | 08:10 |
| *** takashi has joined #openstack-swift | 08:14 | |
| openstackgerrit | Or Ozeri proposed openstack/swift: raise expirer 412 error https://review.openstack.org/326903 | 08:27 |
| openstackgerrit | Or Ozeri proposed openstack/swift: Currently, the expirer daemon treats 412 (precondition failed) as successful DELETEs. https://review.openstack.org/326903 | 08:35 |
| *** dmk0202 has joined #openstack-swift | 08:37 | |
| *** ozeri has joined #openstack-swift | 08:37 | |
| openstackgerrit | Or Ozeri proposed openstack/swift: Raise 412 response on expirer https://review.openstack.org/326903 | 08:39 |
| *** acoles_ is now known as acoles | 08:44 | |
| *** kei_yama has quit IRC | 08:45 | |
| acoles | zaitcev: if I am fired then nobody has told me yet ;) | 08:46 |
| *** openstackgerrit has quit IRC | 08:48 | |
| *** openstackgerrit has joined #openstack-swift | 08:48 | |
| *** jmccarthy1 has joined #openstack-swift | 08:52 | |
| *** jmccarthy1 has quit IRC | 09:01 | |
| *** mvk has quit IRC | 09:03 | |
| *** hseipp has quit IRC | 09:11 | |
| mahatic_ | acoles: hello. thanks for the TODO cleanups. On thsi trello mentioned https://trello.com/c/B4lmJw7Q/116-add-test-for-non-zero-offset-in-test-create-decryption-context, jrichli mentions an existing test | 09:19 |
| mahatic_ | on this trello* | 09:20 |
| *** asettle has joined #openstack-swift | 09:33 | |
| *** mvk has joined #openstack-swift | 09:34 | |
| mahatic_ | acoles: was wondering the same, if we need another test | 09:35 |
| *** jmccarthy1 has joined #openstack-swift | 09:37 | |
| *** hosanai has quit IRC | 09:38 | |
| kmARC | acoles we soon would start pushing some code and documentation on secure deletion | 09:39 |
| kmARC | should we use feature-crypto (I don't think so) or open a branch from that on our own (feature-secure-deletion) | 09:40 |
| *** takashi has quit IRC | 09:40 | |
| mathiasb | kmARC, acoles: the secure deletion code basically has a drop-in replacement for the keymaster (in separate files), so to activate it, you would only replace keymaster with the new secdel_keymaster in proxy.conf | 09:51 |
| mathiasb | (if that helps to decide the naming one way or another) | 09:51 |
| acoles | kmARC: mathiasb the planned next step is for the current content of feature/crypto* to be proposed for merging to master (via a sanitised set of patches on a feature/crypto-review) branch. (* almost, we have one more patch to merge) | 09:52 |
| acoles | kmARC: my suggestion would be to wait for that to happen, then propose enhancements as patches on master. notmyname may have different advice. | 09:53 |
| kmARC | okay it sounds sensible | 09:54 |
| acoles | kmARC: actually, mys suggestion is to *review* what we propose to feature/crypto-review when that happens and ensure it does not preclude your enhancements | 09:54 |
| acoles | ...and then propose to master :) | 09:54 |
| mathiasb | acoles: what's your estimate for the timeline for the merging of feature/crypto to master? | 09:55 |
| kmARC | sure. we all read the documentation there. I even saw comments from Chrisian there | 09:55 |
| mathiasb | I don't think anyone of us has spotted anything that clashes with what we would be proposing | 09:55 |
| acoles | mathiasb: I hope we have feature/crypto-review ready this week, hard to say how long that will take to be reviewed and merged to master, again I would *hope* that could happen in <2 weeks. | 10:00 |
| acoles | but my estimates have been proven grossly optimistic before! | 10:01 |
| acoles | mahatic_: hi, left a comment on trello. it would be nice to have a direct test that sets offset >0 and checks the context iv. | 10:01 |
| acoles | mahatic_: but good to know there is some coverage - so I didn't check, do we have coverage report showing for these lines https://github.com/openstack/swift/blob/79d401033d432d85616ac34922f079fd043b0976/swift/common/middleware/crypto_utils.py#L75-L82 | 10:02 |
| mathiasb | acoles: got it, thanks! | 10:02 |
| acoles | phew! just completed reading the TC minutes, need a coffee... | 10:03 |
| openstackgerrit | Christian Schwede proposed openstack/swift: Make test_ringbuilder less brittle https://review.openstack.org/326966 | 10:07 |
| openstackgerrit | Christian Schwede proposed openstack/swift: Rebalance with min_part_seconds_left > 0 https://review.openstack.org/326967 | 10:08 |
| mahatic_ | acoles: oic. thanks. yeah, I don't think there is any coverage for those lines you mention. | 10:08 |
| mahatic_ | acoles: TC minutes huh, that has been my "morning news" for the past couple of days | 10:09 |
| *** jamie_h has joined #openstack-swift | 10:12 | |
| openstackgerrit | Merged openstack/swift: crypto - remove stale TODOs https://review.openstack.org/326654 | 10:13 |
| acoles | cschwede: sorry, I had meant to delete my comment about the threading bug on patch 323950, I realised after I had written it that the subprocess code was no longer being used because you had a much cleaner approach. | 10:16 |
| patchbot | acoles: https://review.openstack.org/#/c/323950/ - swift - Refactor locale tests and unicode issue | 10:16 |
| cschwede | acoles: ah ok - thx a lot for coming back on this! yes, that was my 2nd idea behind that patch - cleanup the code a bit | 10:17 |
| *** jmccarthy1 has quit IRC | 10:26 | |
| *** lifeless has joined #openstack-swift | 10:35 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Add simple multiple server type test https://review.openstack.org/326399 | 10:46 |
| *** lifeless has quit IRC | 10:47 | |
| *** lifeless_ has joined #openstack-swift | 10:47 | |
| *** tesseract has quit IRC | 10:50 | |
| *** tesseract has joined #openstack-swift | 10:51 | |
| *** d0ugal has quit IRC | 10:59 | |
| *** d0ugal has joined #openstack-swift | 11:00 | |
| *** takashi has joined #openstack-swift | 11:08 | |
| *** d0ugal has quit IRC | 11:09 | |
| *** takashi has quit IRC | 11:23 | |
| *** diogogmt has joined #openstack-swift | 11:27 | |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift: crypto - Test with non-zero offset https://review.openstack.org/327028 | 11:32 |
| *** d0ugal has joined #openstack-swift | 11:32 | |
| mahatic_ | acoles: jrichli: ^ | 11:33 |
| *** bikmak has joined #openstack-swift | 11:34 | |
| *** dmorita has joined #openstack-swift | 11:37 | |
| *** jmccarthy has quit IRC | 11:37 | |
| *** jmccarthy has joined #openstack-swift | 11:38 | |
| acoles | mahatic_: thanks | 11:41 |
| *** dmorita has quit IRC | 11:42 | |
| *** diogogmt has quit IRC | 12:05 | |
| jith_ | hi all, how to enable seperate logging for swift.. in my setup all swift logs are moving to syslog.. how can i seperate it to /var/log/swift.log | 12:18 |
| *** ppai has quit IRC | 12:38 | |
| *** takashi has joined #openstack-swift | 12:51 | |
| *** klamath has joined #openstack-swift | 12:55 | |
| *** klamath has quit IRC | 13:00 | |
| *** links has quit IRC | 13:03 | |
| portante | jith_: what do you have for an /etc/rsyslog.conf and /etc/rsyslog.d/*.conf setup? | 13:18 |
| *** links has joined #openstack-swift | 13:18 | |
| jith_ | portante: thanks for the reply.. | 13:20 |
| jith_ | nothing specific to swift | 13:21 |
| jith_ | http://paste.openstack.org/show/508931/ | 13:21 |
| portante | jith_: seems like you duplicate logs in a number of places, okay | 13:23 |
| *** klamath has joined #openstack-swift | 13:24 | |
| jith_ | portante.. no i m getting swift logs in /var/log/syslog file alone | 13:24 |
| jith_ | i want those in a seperate file like /var/log/swift.log | 13:24 |
| *** sheel has quit IRC | 13:25 | |
| portante | so in your various /etc/*.conf files for swift, do you have log_facility = LOG_LOCAL0 enabled | 13:26 |
| jith_ | https://github.com/openstack/swift/blob/master/etc/swift-rsyslog.conf-sample ... is this the sample rsyslog.conf file.. should i add these contents to rsyslog.conf | 13:27 |
| jith_ | oh i never noticed.. will check now | 13:27 |
| portante | yes to the sample swift-rsyslog.conf-sample. You can add it to /etc/rsyslog.d/ as swift-rsyslog.conf and it will get included ahead of your other rules | 13:28 |
| jith_ | no i havnt enabled the log_facility | 13:29 |
| portante | then you don't have to change your /etc/rsyslog.conf | 13:29 |
| jith_ | what is log_address | 13:29 |
| portante | then I believe you can just use that /etc/rsyslog.conf as is | 13:29 |
| portante | since you have swift logs flowing through rsyslog | 13:29 |
| jith_ | yes.. but i want those swift logs in a seperate file | 13:30 |
| portante | okay, so uncomment the appropriate lines to get them in /var/log/swift.log or by individual services | 13:30 |
| jith_ | thanks really | 13:31 |
| portante | does that make sense? | 13:31 |
| jith_ | log_address = /dev/log ?? | 13:31 |
| portante | I don't think you need to change any swift configuration file | 13:31 |
| jith_ | should i change it to log_address=/var/log/swift.log | 13:31 |
| portante | if you have swift logs in /var/log/messages or /var/log/syslog, then swift is already sending logs there | 13:32 |
| jith_ | then i just have to put that swift-rsyslog.conf in syslog.d?? | 13:32 |
| portante | yes | 13:32 |
| portante | and then pick one of the three behaviors your want | 13:32 |
| portante | all swift logs in one file? uncomment line 2, comment out lines 11-15 | 13:32 |
| portante | all swift logs by major service category? leave as is | 13:33 |
| portante | all swift logs by individual daemon? comment out lines 11-15, uncomment lines 18-31 | 13:33 |
| portante | but leave line 35 as is | 13:33 |
| portante | ah, one modification, I think line 2 should be "local0.* /var/log/swift/all.log", not "local.* ..." | 13:34 |
| jith_ | thanks a lott | 13:36 |
| *** ozeri has quit IRC | 14:00 | |
| *** ametts has joined #openstack-swift | 14:02 | |
| *** psachin has quit IRC | 14:23 | |
| *** links has quit IRC | 14:29 | |
| *** jrichli has joined #openstack-swift | 14:29 | |
| *** StraubTW has joined #openstack-swift | 14:34 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - Don't encrypt update override etags for empty object https://review.openstack.org/327144 | 14:39 |
| *** _JZ_ has joined #openstack-swift | 14:39 | |
| notmyname | wait, is portante back!? | 14:40 |
| acoles | jrichli: ^^ follow up on the anomaly you spotted yesterday | 14:40 |
| openstackgerrit | Merged openstack/swift: crypto - purge crypto sysmeta from responses https://review.openstack.org/320579 | 14:40 |
| *** StraubTW has quit IRC | 14:40 | |
| * portante ducks | 14:40 | |
| jrichli | acoles: thanks, i will take a look | 14:40 |
| openstackgerrit | Merged openstack/swift: crypto - refactor footers callback https://review.openstack.org/324313 | 14:40 |
| *** pauloewerton has joined #openstack-swift | 14:42 | |
| *** StraubTW has joined #openstack-swift | 14:42 | |
| *** flaper87 has left #openstack-swift | 14:42 | |
| notmyname | portante: did you get answers to what you were asking about (via twitter)? | 14:43 |
| notmyname | acoles: I'll look over the crypto-review patch chain email this morning | 14:51 |
| notmyname | acoles: sorry i ddin't get to it yesterday. seems like there was something else going on... ;-) | 14:52 |
| acoles | notmyname: ok np. I merged the final essential patch today so close to ready. | 14:52 |
| acoles | notmyname: :) I wasn't expecting it to be top of your list. | 14:53 |
| *** thumpba has joined #openstack-swift | 14:54 | |
| portante | notmyname: I have not checked yet | 14:55 |
| notmyname | portante: ok :-) | 14:55 |
| *** links has joined #openstack-swift | 14:57 | |
| acoles | timburke: thanks for the locale tip | 14:58 |
| portante | just checked, as I guessed, this is because of the python GIL | 14:58 |
| portante | combined with the fact that linux does not do try async IO at its core | 14:58 |
| notmyname | back after a while | 15:00 |
| *** zul_ has joined #openstack-swift | 15:05 | |
| *** zul has quit IRC | 15:05 | |
| *** zul_ has quit IRC | 15:06 | |
| *** zul has joined #openstack-swift | 15:07 | |
| *** sheel has joined #openstack-swift | 15:07 | |
| *** tmoreira has quit IRC | 15:12 | |
| *** klrmn has joined #openstack-swift | 15:14 | |
| timburke | acoles: happy to help. that was the biggest problem i hit with my gibberizer; i couldn't figure out how to tell python to still use utf8 :-/ | 15:19 |
| *** tmoreira has joined #openstack-swift | 15:24 | |
| acoles | timburke: I'm still curious as to why the test works with just LC_ALL=eo, no need for a utf8 qualifier on that locale name | 15:26 |
| *** arch-nemesis has joined #openstack-swift | 15:26 | |
| timburke | acoles: in test_locale? we're no longer trying to send this to python's print, are we? we're just verifying that we get a unicode object out | 15:28 |
| timburke | (maybe also worth noting that we declare the encoding for the file at the top: `# coding: utf-8`) | 15:29 |
| *** dmk0202 has quit IRC | 15:29 | |
| acoles | timburke: yes, test_locale - I added a print to the tests and it was happy. But as you say there is an encoding declared. | 15:30 |
| timburke | ah. hrm, not sure then... maybe the python interpreter was still working under the assumption of en/utf-8, but swift's gettext_ had eo? | 15:31 |
| timburke | the declared encoding was more about why we could have crazy raw bytes in the unicode strings instead of \uwhatever | 15:32 |
| *** dmorita has joined #openstack-swift | 15:37 | |
| *** jrichli has quit IRC | 15:40 | |
| *** dmorita has quit IRC | 15:41 | |
| *** dmorita has joined #openstack-swift | 15:45 | |
| *** bikmak has quit IRC | 15:47 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Improve locale tests and fix unicode issue https://review.openstack.org/323950 | 15:47 |
| *** zaitcev has joined #openstack-swift | 15:51 | |
| *** ChanServ sets mode: +v zaitcev | 15:51 | |
| cschwede | acoles: timburke: thanks for the tests & tips! i just uploaded a new patchset, saw your comments too late… | 15:58 |
| *** bikmak has joined #openstack-swift | 15:58 | |
| *** chlong has quit IRC | 15:58 | |
| cschwede | acoles: last patchset includes the old test again; i think there was still an issue | 15:58 |
| *** jrichli has joined #openstack-swift | 15:58 | |
| *** Jeffrey4l has quit IRC | 15:59 | |
| cschwede | and now it works both with and without using utf-8 lang settings | 16:00 |
| timburke | that makes me nervous. what if someone legitimately isn't using utf-8? what if their terminal expects utf-16, or latin1, or god knows what? | 16:02 |
| cschwede | timburke: yes, that was my thinking as well; i don’t mind removing that, in which case one has to explicilty set the encoding itself (like you proposed) | 16:03 |
| timburke | locales make my head hurt :-( | 16:04 |
| acoles | timburke: that makes two of us then | 16:05 |
| cschwede | timburke: acoles: make it four - my colleague and me are just super annoyed by debugging locale issues right now | 16:06 |
| timburke | i fear it will only get worse if we ever try again to add py3 support... | 16:07 |
| *** gyee has joined #openstack-swift | 16:15 | |
| *** links has quit IRC | 16:15 | |
| cschwede | timburke: well, let’s go back to patchset #2, the one where acoles already put his +2 on it | 16:18 |
| *** jith_ has quit IRC | 16:19 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Refactor locale tests and fix unicode issue https://review.openstack.org/323950 | 16:20 |
| openstackgerrit | Tim Burke proposed openstack/swift: Tighten up some sysmeta purging https://review.openstack.org/327221 | 16:22 |
| *** lyrrad has joined #openstack-swift | 16:22 | |
| timburke | acoles: ^^^ a few follow-ups to patch 320579 that i hadn't gotten around to posting | 16:23 |
| patchbot | timburke: https://review.openstack.org/#/c/320579/ - swift (feature/crypto) - crypto - purge crypto sysmeta from responses (MERGED) | 16:23 |
| *** klrmn has quit IRC | 16:23 | |
| *** suyash has joined #openstack-swift | 16:26 | |
| *** takashi has quit IRC | 16:26 | |
| *** daemontool has joined #openstack-swift | 16:27 | |
| jrichli | timburke: did you mean to put that last patch on master? | 16:27 |
| jrichli | i mean - to propose patch 327221 to master instead of feature/crypto | 16:28 |
| patchbot | jrichli: https://review.openstack.org/#/c/327221/ - swift (feature/crypto) - Tighten up some sysmeta purging | 16:28 |
| jrichli | oh, nevermind | 16:28 |
| acoles | timburke: ack, thanks | 16:28 |
| jrichli | i guess its gonna be one of those days for me | 16:28 |
| timburke | jrichli: had me worried for a minute :-) | 16:29 |
| timburke | first for needing to fix my patch, then for needing to fix patchbot | 16:29 |
| acoles | jrichli: just blame it on being a thursday | 16:29 |
| jrichli | acoles: lol, yes. exactly. | 16:30 |
| timburke | i never could get the hang of thursdays... | 16:30 |
| *** tesseract has quit IRC | 16:31 | |
| *** rledisez has quit IRC | 16:32 | |
| *** d0ugal has quit IRC | 16:33 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Add --get <key> option to `swift capabilities` / `swift info` https://review.openstack.org/276043 | 16:35 |
| *** nadeem has joined #openstack-swift | 16:39 | |
| *** rcernin has quit IRC | 16:42 | |
| *** nadeem has quit IRC | 16:46 | |
| *** nadeem has joined #openstack-swift | 16:46 | |
| *** d0ugal has joined #openstack-swift | 16:49 | |
| *** mvk has quit IRC | 16:50 | |
| *** ChubYann has joined #openstack-swift | 16:52 | |
| *** asettle has quit IRC | 16:56 | |
| *** daemontool has quit IRC | 16:57 | |
| acoles | timburke: this is a good addition https://review.openstack.org/#/c/327221/1/test/unit/common/middleware/test_decrypter.py@95 | 17:00 |
| patchbot | acoles: patch 327221 - swift (feature/crypto) - Tighten up some sysmeta purging | 17:00 |
| timburke | acoles: yeah, otherwise when i took out the purging in GET/HEAD, i'd just see a body mismatch in test_encrypter_decrypter | 17:01 |
| acoles | timburke: do you mind if I add assertNotIn for the the other crypto- headers too? | 17:02 |
| acoles | the rest of your changes are good too :) but I wanted to ask about that ^^ | 17:02 |
| timburke | not at all. i was just going for a quick pass/fail sort of a thing rather than full coverage | 17:03 |
| timburke | might even be worth pulling it out to a _assert_no_crypto_sysmeta and adding it to other tests as appropriate | 17:04 |
| *** thumpba has quit IRC | 17:04 | |
| *** thumpba has joined #openstack-swift | 17:05 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Tighten up some sysmeta purging https://review.openstack.org/327221 | 17:07 |
| acoles | looks like there is a pep8 fail there anyway (2 blank lines) so I will at least fill one of them up | 17:07 |
| acoles | timburke: if you and jenkins are happy with that then please +A | 17:08 |
| *** acoles is now known as acoles_ | 17:11 | |
| *** klrmn has joined #openstack-swift | 17:16 | |
| *** thumpba has quit IRC | 17:20 | |
| *** thumpba has joined #openstack-swift | 17:21 | |
| *** thumpba has quit IRC | 17:23 | |
| *** thumpba has joined #openstack-swift | 17:25 | |
| *** asettle has joined #openstack-swift | 17:26 | |
| *** bikmak has quit IRC | 17:27 | |
| *** geaaru has quit IRC | 17:49 | |
| *** asettle has quit IRC | 17:49 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Tighten up some sysmeta purging https://review.openstack.org/327221 | 17:58 |
| notmyname | good morning again | 18:04 |
| *** jamie_h has quit IRC | 18:07 | |
| -openstackstatus- NOTICE: Zuul has been restarted to correct an error condition. Events since 17:30 may have been missed; please 'recheck' your changes if they were uploaded since then, or have "NOT_REGISTERED" errors. | 18:11 | |
| *** silor has joined #openstack-swift | 18:13 | |
| notmyname | this one goes out to pandemicsyn https://twitter.com/sigje/status/740588201547419648 | 18:20 |
| *** welldannit has joined #openstack-swift | 18:24 | |
| *** silor1 has joined #openstack-swift | 18:26 | |
| *** silor has quit IRC | 18:27 | |
| *** silor1 is now known as silor | 18:27 | |
| *** links has joined #openstack-swift | 18:32 | |
| *** lcurtis has joined #openstack-swift | 18:38 | |
| zaitcev | haha nuclear option by ttx | 18:40 |
| *** diogogmt has joined #openstack-swift | 18:42 | |
| ahale | kinda feel thats the only realistic option | 18:43 |
| notmyname | zaitcev: the phrasing of his comment was interesting. TC oversight (IMO) is not the carrot. it's the stick by which you get other good things from the foundation. seems that his phrasing implies that the TC oversight is the benefit instead of the cost | 18:44 |
| notmyname | ahale: no, I don't think so | 18:44 |
| zaitcev | I'm going Adm. Farragut regardless. | 18:45 |
| ahale | im not a fan of special casing really, theres enough friction around over that already with swift | 18:45 |
| zaitcev | "Damn the TC, full steam ahead!" | 18:45 |
| notmyname | zaitcev: I don't get that reference | 18:46 |
| notmyname | zaitcev: ah | 18:46 |
| zaitcev | https://en.wikipedia.org/wiki/David_Farragut | 18:46 |
| notmyname | ahale: yeah, I agree. I've worked hard to remove that stigma, and I do believe that swift doesn't have special-casing today. however, should the TC explicitly say "swift is special, golang is ok", then it will make future internal debates harder | 18:48 |
| notmyname | however, that's the path forward that the TC wants to pursue, so that's what I'll work on. it's a lot better than arbitrarily splitting the repo into different parts | 18:48 |
| notmyname | and to zaitcev's point, yeah, I think we should still be full steam ahead on using golang, as we discussed in austin | 18:49 |
| *** pgbridge has joined #openstack-swift | 18:49 | |
| *** arch-nemesis has quit IRC | 18:50 | |
| pdardeau | when did TC first learn of the serious interest in golang? hopefully, not within the last 30-60 days | 18:52 |
| notmyname | a year ago http://lists.openstack.org/pipermail/openstack-dev/2015-April/063019.html | 18:56 |
| ahale | the first summit presentation was vancouver i think? | 18:59 |
| ahale | and yeah - you made the insanity of an arbituary split pretty clear in the email thread you sent :) | 19:00 |
| notmyname | first time "hummingbird" was used in this channel was http://eavesdrop.openstack.org/irclogs/%23openstack-swift/%23openstack-swift.2014-11-24.log.html#t2014-11-24T18:07:30 (november 24, 2014) | 19:02 |
| notmyname | /lunch | 19:03 |
| *** arch-nemesis has joined #openstack-swift | 19:05 | |
| *** rcernin has joined #openstack-swift | 19:25 | |
| *** diogogmt has quit IRC | 19:28 | |
| *** links has quit IRC | 19:31 | |
| *** gyee has quit IRC | 19:40 | |
| * notmyname hopes people have been thinking about ranged copy | 19:48 | |
| timburke | oh yeah, that... | 19:49 |
| notmyname | timburke: think real hard, real fast ;-) | 19:49 |
| timburke | i definitely want to keep it, at least for single ranges. i'm inclined to say we should de-MIME multiple ranges, but i think i'd be ok with just disallowing that (with a 400) | 19:52 |
| timburke | storing the MIME doc is stupid, though | 19:52 |
| notmyname | my initial reaction is that sounds reasonable | 19:52 |
| torgomatic | storing the MIME doc is obviously dumb; storing the concatenated ranges is dumb because, due to partial satisfiability, you don't even know what the indices of the bytes are | 19:55 |
| torgomatic | it's all dumb | 19:55 |
| torgomatic | 400 Bad Idea | 19:55 |
| torgomatic | HTTP/1.1 400 What Were You Thinking | 19:56 |
| *** StraubTW has left #openstack-swift | 19:58 | |
| *** asettle has joined #openstack-swift | 19:59 | |
| *** asettle has quit IRC | 19:59 | |
| timburke | torgomatic: what do you mean by partial satisfiability? | 20:05 |
| torgomatic | timburke: imagine a 1000-byte file... I say Range: bytes=0-100,800-1200,1450-2210,650-750 | 20:06 |
| timburke | torgomatic: nope, 416 | 20:08 |
| cschwede | i hope apps check if this kind of request makes sense at all… | 20:08 |
| torgomatic | timburke: probably the pathological-range detector firing; try sorting by first byte | 20:08 |
| timburke | heh, i hadn't actually tried it; i was declaring what *should* be the response | 20:09 |
| *** jeblair has quit IRC | 20:09 | |
| *** jeblair has joined #openstack-swift | 20:10 | |
| *** asettle has joined #openstack-swift | 20:10 | |
| timburke | bah :-( i suppose we could add some X-Ensure-All-Ranges-Satisfiable header or something | 20:13 |
| *** mvk has joined #openstack-swift | 20:19 | |
| notmyname | redbo: nadeem: dfg: it would be really great if one of you gophers could work on getting hummingbird processes running under swift-init. I know clayg had some thoughts on it (but he's on vacation for another week) | 20:22 |
| torgomatic | timburke: even so, stuff like bytes 900-1100 of a 1000-byte object counts as satisfiable, but you don't get all 200 bytes. | 20:22 |
| torgomatic | er, 201 bytes, I suppose | 20:22 |
| notmyname | torgomatic: timburke: single-range ok and 4xx on multirage seems like a sane thing | 20:23 |
| *** thumpba has quit IRC | 20:27 | |
| timburke | torgomatic: i'm inclined to say that in this hypothetical, special, do-what-i-said mode, that would also 416 | 20:27 |
| openstackgerrit | Merged openstack/swift-bench: Put py34 first in the env order of tox https://review.openstack.org/260527 | 20:28 |
| *** thumpba has joined #openstack-swift | 20:29 | |
| *** silor has quit IRC | 20:33 | |
| *** thumpba has quit IRC | 20:33 | |
| *** sheel has quit IRC | 20:35 | |
| notmyname | 2.8.0 request has been approved https://review.openstack.org/#/c/326771/ | 20:40 |
| patchbot | notmyname: patch 326771 - releases - swift 2.8.0 release | 20:40 |
| *** thumpba has joined #openstack-swift | 20:43 | |
| notmyname | team meeting in 15 minutes | 20:46 |
| *** openstackgerrit has quit IRC | 20:48 | |
| *** openstackgerrit has joined #openstack-swift | 20:48 | |
| *** gyee has joined #openstack-swift | 20:55 | |
| *** thumpba has quit IRC | 20:56 | |
| *** acoles_ is now known as acoles | 20:57 | |
| kota_ | morning | 20:57 |
| notmyname | hello kota_ | 20:57 |
| kota_ | hello, notmynme | 20:58 |
| *** joeljwright has joined #openstack-swift | 20:59 | |
| *** ChanServ sets mode: +v joeljwright | 20:59 | |
| notmyname | joeljwright: hello. I saw you'd left comments on the versions in the sidebar patch. thanks | 20:59 |
| notmyname | joeljwright: are you planning on pushing a patch set over mine (please feel free), or are you just trying stuff locally and waiting for me to update it? | 21:00 |
| joeljwright | notmyname: no worries, I thought it was worth following up | 21:00 |
| notmyname | yeah, definitely. thanks | 21:00 |
| joeljwright | I can push a version if you like | 21:00 |
| notmyname | yep :-) | 21:00 |
| *** _JZ_ has quit IRC | 21:00 | |
| notmyname | team meeting time in #openstack-meeting | 21:00 |
| joeljwright | need to work out how to define the variable, so it won't be until tomorrow probably | 21:00 |
| notmyname | ok | 21:01 |
| joeljwright | sorry - it's late | 21:01 |
| *** _JZ_ has joined #openstack-swift | 21:02 | |
| *** asettle has quit IRC | 21:03 | |
| *** jrichli_ has joined #openstack-swift | 21:04 | |
| *** jrichli_ has quit IRC | 21:05 | |
| *** jrichli_ has joined #openstack-swift | 21:05 | |
| *** chlong has joined #openstack-swift | 21:05 | |
| *** jrichli_ has quit IRC | 21:05 | |
| *** jrichli_ has joined #openstack-swift | 21:06 | |
| *** jrichli__ has joined #openstack-swift | 21:06 | |
| *** jrichli has quit IRC | 21:06 | |
| *** ametts has quit IRC | 21:14 | |
| *** mvk_ has joined #openstack-swift | 21:25 | |
| *** mvk has quit IRC | 21:28 | |
| *** lifeless_ is now known as lifeless | 21:30 | |
| *** dfg has joined #openstack-swift | 21:31 | |
| *** CaioBrentano has quit IRC | 21:32 | |
| *** dfg is now known as dfg_ | 21:34 | |
| *** pauloewerton has quit IRC | 21:48 | |
| openstackgerrit | Merged openstack/swift: [Docs] Document prevention of disk full scenarios https://review.openstack.org/315610 | 21:51 |
| *** rcernin is now known as rcernin|off | 21:54 | |
| *** ouchkernel has quit IRC | 21:57 | |
| onovy | notmyname: np, let's continue here | 22:01 |
| *** dfg_ is now known as dfg | 22:02 | |
| *** jrichli__ has quit IRC | 22:02 | |
| *** jrichli_ has quit IRC | 22:02 | |
| *** ouchkernel has joined #openstack-swift | 22:02 | |
| onovy | oct 2015 is really far in past... :( | 22:03 |
| *** asettle has joined #openstack-swift | 22:04 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 22:05 |
| *** klamath has quit IRC | 22:05 | |
| *** klamath has joined #openstack-swift | 22:05 | |
| notmyname | onovy: heh, sorry. was in the wrong channel | 22:06 |
| onovy | notmyname: i'm really sad about this package and thinking about abbadonning it, taking simple few-line patch with psutils and patch it in debian during packing. | 22:06 |
| onovy | /package/patch/ | 22:06 |
| onovy | peter already give it up btw | 22:07 |
| *** dfg is now known as dfg_ | 22:07 | |
| notmyname | I totally get your frustration, and it's completely justified | 22:08 |
| notmyname | reviews are a very very slow process. so slow it can discourage people from even wanting to participate | 22:08 |
| *** asettle has quit IRC | 22:09 | |
| *** notmorgan has joined #openstack-swift | 22:10 | |
| * notmorgan comes to lurk. and make not<tab> harder (j/k, mostly to lurk) | 22:10 | |
| notmyname | torgomatic: you'd probably like this http://jvns.ca/blog/2016/06/07/strace-y/ | 22:10 |
| notmorgan | notmyname: so, i wanted to touch on something i talked with you yesterday on. | 22:10 |
| notmyname | sure | 22:11 |
| notmorgan | notmyname: when you have a few minutes :). just a process thing around hummingbird. | 22:11 |
| onovy | nevermind, gn | 22:11 |
| notmorgan | and the install process/running it. | 22:11 |
| notmorgan | and feel free to tell me to comeback tomorrow | 22:11 |
| notmorgan | i have plenty of other things to do besides come and bother you | 22:12 |
| notmyname | now's good | 22:12 |
| notmorgan | ok | 22:12 |
| notmyname | and in here is good because some other people here definitely know more than me about it | 22:12 |
| notmorgan | so as you said, the process for hummingbird is pip install swift and then compile/load-in-from-package-or-whatever the hummingbird binary | 22:12 |
| notmorgan | correct? | 22:12 |
| notmorgan | s/pip/install the python bits/ | 22:13 |
| notmyname | compile the golang code, then copy that binary to whatever machine you're runnign it on, then run it | 22:13 |
| notmorgan | "get a binary somehow" | 22:13 |
| notmyname | right | 22:14 |
| notmyname | so you'll have the python code that's run as normal (eventually it's effectively a `python module.py`) | 22:14 |
| notmyname | and another binary that's executable,e g `./hummingbird` | 22:15 |
| notmorgan | now from a strict code management standpoint, does it make sense to combine the golang stuff in the same repo as the python? -- this has no bearing on offiical/non-officialness of things. | 22:15 |
| notmorgan | just a pure "with golang and python" or "any lang that isn't python native" does the same repo make sense | 22:15 |
| * notmorgan admits ignorance on this part of golang things. | 22:15 | |
| notmorgan | which is why i am asking | 22:15 |
| notmyname | right. I get where you're coming from. in this case I think it does, for a few reasons | 22:15 |
| notmorgan | cool :). i get to learn things today :) | 22:16 |
| notmyname | first, from a community perspective, as my other email(s) said, it's very closely tied together and it's the same people working on it. it's very beneficial to keep those together and release them at the same time etc | 22:16 |
| notmyname | a corollary to that is that it's not actually too different from the situation on the master branch today | 22:17 |
| notmyname | that is, we have a proxy server process and an object server process, both written in python, but they're normally distributed in separate packages and have different binaries | 22:17 |
| notmorgan | right. i just see this as almost two deliverables (hell i see nova compute vs nova-api in the same way, but thats off in the weeds) | 22:17 |
| notmyname | just so happens that today they are implemented in the same language and tomorrow they won't be | 22:18 |
| notmorgan | regardless of impl. object-server and api-proxy | 22:18 |
| * notmorgan wouldn't advocate changing, just had some folks ask about how this actually works. | 22:18 | |
| notmorgan | so when building the sdist for python we're just ignoreing the golang in this case. | 22:18 |
| notmorgan | and when building golang bits we're ignoring python bits. | 22:18 |
| notmyname | second reason...wait, i think I just gave both reasons listed under "first..." | 22:18 |
| notmorgan | since they're in the same repo. | 22:18 |
| notmyname | right | 22:18 |
| notmorgan | hehe i think you did. | 22:18 |
| notmorgan | i almost wish things were a bit more split in the land of API vs "non-API" agent things in openstack. | 22:19 |
| notmorgan | packaging wise. | 22:19 |
| notmorgan | but splitting hairs. cool i got what i wanted to know :) | 22:20 |
| notmorgan | thnx! | 22:20 |
| notmyname | great | 22:20 |
| notmyname | any time | 22:20 |
| * notmorgan exits channel to try and keep not<tab> to be easier here. | 22:20 | |
| notmorgan | ;) | 22:20 |
| *** notmorgan has left #openstack-swift | 22:20 | |
| *** acoles is now known as acoles_ | 22:20 | |
| *** rcernin|off has quit IRC | 22:21 | |
| *** kmARC has quit IRC | 22:25 | |
| openstackgerrit | David Goetz proposed openstack/swift: go: have replicateDevice sleep a little between loops https://review.openstack.org/327401 | 22:31 |
| *** Jeffrey4l has joined #openstack-swift | 22:47 | |
| *** joeljwright has quit IRC | 22:57 | |
| *** arch-nemesis has quit IRC | 23:00 | |
| *** pgbridge has quit IRC | 23:26 | |
| *** nadeem has quit IRC | 23:27 | |
| *** lcurtis has quit IRC | 23:27 | |
| *** kei_yama has joined #openstack-swift | 23:36 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!