| ho | mattoliverau: morning! | 00:01 |
|---|---|---|
| openstackgerrit | Merged openstack/swift: remove pbr from requirements.txt. It's not a run-time dependency https://review.openstack.org/245266 | 00:03 |
| notmyname | wheee | 00:04 |
| *** sgundur has quit IRC | 00:07 | |
| *** jrichli has joined #openstack-swift | 00:10 | |
| *** tsg has quit IRC | 00:10 | |
| *** openstackgerrit has quit IRC | 00:16 | |
| *** openstackgerrit has joined #openstack-swift | 00:16 | |
| *** blmartin has quit IRC | 00:37 | |
| ktsuyuzaki | good morning, I'm back from business travel week. now starting to catch up recent conversations. | 00:46 |
| ktsuyuzaki | oh, my nick has been changed... | 00:47 |
| *** ktsuyuzaki is now known as kota_ | 00:47 | |
| *** m_kazuhiro has joined #openstack-swift | 00:56 | |
| mattoliverau | kota_: morning | 00:58 |
| ho | kota_: morning! | 00:59 |
| kota_ | mattoliverau: ho: morning! | 01:00 |
| *** NM has joined #openstack-swift | 01:02 | |
| *** NM has quit IRC | 01:02 | |
| *** jerrygb has quit IRC | 01:09 | |
| *** jerrygb has joined #openstack-swift | 01:10 | |
| *** jerrygb has quit IRC | 01:14 | |
| ho | tdasilva: around? | 01:33 |
| *** klrmn1 has quit IRC | 01:54 | |
| *** tsg has joined #openstack-swift | 01:57 | |
| *** 17SAD4YI2 has joined #openstack-swift | 02:00 | |
| *** Vinsh_ has joined #openstack-swift | 02:00 | |
| *** 17SAD4YI2 has quit IRC | 02:01 | |
| *** haomaiwa_ has joined #openstack-swift | 02:01 | |
| *** Vinsh has quit IRC | 02:04 | |
| openstackgerrit | Merged openstack/swift: Unit tests for account/backend.py https://review.openstack.org/240613 | 02:25 |
| *** david-ly_ has quit IRC | 02:36 | |
| *** Vinsh_ has quit IRC | 02:42 | |
| *** jerrygb has joined #openstack-swift | 02:44 | |
| *** sanchitmalhotra has joined #openstack-swift | 02:56 | |
| tsg | kota_: ping | 02:58 |
| kota_ | tsg: pong | 02:58 |
| *** diazjf has joined #openstack-swift | 02:59 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** diazjf has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-swift | 03:01 | |
| *** venkat_p has joined #openstack-swift | 03:02 | |
| *** jrichli has quit IRC | 03:04 | |
| janonymous | morning o/ | 03:08 |
| *** jerrygb has quit IRC | 03:11 | |
| *** proteusguy_ has quit IRC | 03:12 | |
| *** jerrygb has joined #openstack-swift | 03:13 | |
| mattoliverau | janonymous: morning | 03:20 |
| *** Vinsh has joined #openstack-swift | 03:21 | |
| *** Vinsh has quit IRC | 03:26 | |
| *** pcaruana has quit IRC | 03:27 | |
| *** klrmn1 has joined #openstack-swift | 03:28 | |
| *** Vinsh has joined #openstack-swift | 03:34 | |
| ho | janonymous: morning! | 03:38 |
| *** proteusguy has joined #openstack-swift | 03:49 | |
| *** sgundur has joined #openstack-swift | 03:57 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwa_ has joined #openstack-swift | 04:01 | |
| *** dimasot has quit IRC | 04:03 | |
| *** david-lyle has joined #openstack-swift | 04:08 | |
| janonymous | o/ matt ho | 04:12 |
| notmyname | good evening | 04:15 |
| *** sgundur has quit IRC | 04:17 | |
| *** m_kazuhiro has quit IRC | 04:19 | |
| janonymous | hi John | 04:21 |
| *** m_kazuhiro has joined #openstack-swift | 04:21 | |
| ho | notmyname: hello- | 04:21 |
| *** trifon has joined #openstack-swift | 04:23 | |
| *** m_kazuhiro has quit IRC | 04:24 | |
| *** jerrygb has quit IRC | 04:24 | |
| *** klrmn1 has quit IRC | 04:26 | |
| *** silor has joined #openstack-swift | 04:27 | |
| notmyname | did you ever think we should rename functests to funtests? | 04:27 |
| mattoliverau | or nofuntests | 04:31 |
| notmyname | heh | 04:31 |
| notmyname | tox, putting the fun in functests | 04:31 |
| notmyname | jenkins, taking the fun out of functests | 04:32 |
| mattoliverau | lol | 04:32 |
| notmyname | pbr, taking the fun out of everything ;-) | 04:32 |
| mattoliverau | but at least it isn't a runtime dependency anymore ;) | 04:35 |
| *** ppai has joined #openstack-swift | 04:35 | |
| notmyname | well it never was, but now we don't lie about that | 04:35 |
| mattoliverau | lol | 04:35 |
| *** dimasot has joined #openstack-swift | 04:39 | |
| *** trifon has quit IRC | 04:47 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 04:50 | |
| *** sanchitmalhotra has quit IRC | 04:52 | |
| *** mahatic has joined #openstack-swift | 04:54 | |
| *** haomaiwa_ has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-swift | 05:01 | |
| *** eranrom has quit IRC | 05:02 | |
| *** diazjf has joined #openstack-swift | 05:03 | |
| mahatic | good morning | 05:04 |
| *** diazjf1 has joined #openstack-swift | 05:05 | |
| *** diazjf has quit IRC | 05:08 | |
| mattoliverau | mahatic: morning | 05:15 |
| mahatic | mattoliverau: hello | 05:15 |
| *** rcernin has joined #openstack-swift | 05:15 | |
| *** haomaiwang has quit IRC | 05:19 | |
| *** haomaiwa_ has joined #openstack-swift | 05:27 | |
| *** venkat_p has quit IRC | 05:33 | |
| *** rcernin has quit IRC | 05:34 | |
| *** mahatic has quit IRC | 05:38 | |
| mattoliverau | notmyname: if I set the concurrency of testr to 1 then it seems the func tests on my SAIO works (well hasn't failed yet and it's been going a while). So might a concurrency thing, ie. clean up between tests and how testr actaully splits up the work. | 05:40 |
| notmyname | oh, interesting | 05:40 |
| mattoliverau | so we might be able to set it to concurrency of 1 if we run out of time. | 05:41 |
| mattoliverau | i.e can't get it working concurrently before the deadline | 05:41 |
| mattoliverau | or maybe we need to be sure we are using different container names in different tests in the func tests.. if we aren't already, I haven't looked, just thinking out loud | 05:43 |
| *** mahatic has joined #openstack-swift | 05:43 | |
| *** pcaruana has joined #openstack-swift | 05:49 | |
| notmyname | yeah, that's a good thought | 05:49 |
| *** venkat_p has joined #openstack-swift | 05:51 | |
| notmyname | the swift functests aren't the slowest test in the gate, so running with concurrency as 1 won't slow anything down | 05:52 |
| mattoliverau | we are using uuid.uuid().hex to create container names in func test Envs. but are we reusing account names, we do keep clearing containers, but depends on how it's been called concurrently, I assume it wouldn't be per test in a class. so maybe it's per file.. I think I need to go read some testr docs or even code :) | 05:53 |
| notmyname | mattoliverau: so I glanced at the reverse listings patch, and the only thing that jumped out (totally not a full review) is that you were taking user input and doing math on it | 05:55 |
| * notmyname goes to find the link | 05:55 | |
| mattoliverau | notmyname: true. we may have a work around for now. Did you want to try using a concurrency or 1? | 05:55 |
| notmyname | mattoliverau: with testr, I'm not sure what I ran with. I don't really care as long as it works ;-) | 05:56 |
| mattoliverau | notmyname: yeah, I think I know what your talking about, that was Tim's magic cause there were some bounds issue, I need to look back in patch sets. | 05:56 |
| *** trifon has joined #openstack-swift | 05:57 | |
| notmyname | mattoliverau: with the reverse listings patch, https://review.openstack.org/#/c/120709/14/swift/container/backend.py in line 593 takes user input conversts to byte, adds 1, then converts to char. which fails if the user input is %ff | 05:57 |
| notmyname | but I don't know if that's actually possible. to send %ff. I got an error in my one quick test locally trying to do that | 05:58 |
| mattoliverau | notmyname: as a hack you can add --parallel --concurrency=1 to wrap-testr.sh (might be able to just use concurrency=). Or pass it in as default. or even just run: python setup.py test --testr-args="--subunit --parallel --concurrency=1" | subunit2pyunit | 06:00 |
| *** haomaiwa_ has quit IRC | 06:01 | |
| notmyname | mattoliverau: push a change over ;-) | 06:01 |
| *** ppai has quit IRC | 06:01 | |
| mattoliverau | notmyname: or I could do that ;P | 06:01 |
| *** haomaiwa_ has joined #openstack-swift | 06:02 | |
| mattoliverau | I'll wait to make sure it passes on my SAIO first :) | 06:02 |
| mattoliverau | it should be almost finished | 06:02 |
| *** sanchitmalhotra has joined #openstack-swift | 06:16 | |
| *** diazjf1 has quit IRC | 06:18 | |
| *** sanchitmalhotra1 has quit IRC | 06:18 | |
| *** rcernin has joined #openstack-swift | 06:18 | |
| *** ppai has joined #openstack-swift | 06:26 | |
| *** pcaruana has quit IRC | 06:28 | |
| *** mahatic has quit IRC | 06:31 | |
| openstackgerrit | Catherine Northcott proposed openstack/swift: Add support for storage policies to have more than one name https://review.openstack.org/241978 | 06:35 |
| *** dimasot has quit IRC | 06:35 | |
| notmyname | Zyric1: :-) | 06:35 |
| *** dimasot has joined #openstack-swift | 06:36 | |
| notmyname | Zyric1: just glanced over it. what I saw looks good. thanks. I'll take a more detailed look tomorrow | 06:38 |
| notmyname | Zyric1: also, I *love* the fact that we'll now have maori in the tests :-) | 06:39 |
| Zyric1 | Thanks notmyname! Sorry it took all day, I had some family engagements and helping my brother with his exam studies :) | 06:39 |
| notmyname | no worries. | 06:40 |
| * notmyname checks out for the night | 06:41 | |
| *** Zyric1 has quit IRC | 06:42 | |
| *** takashi_ has joined #openstack-swift | 06:48 | |
| *** takashi has quit IRC | 06:50 | |
| *** eranrom has joined #openstack-swift | 06:51 | |
| *** takashi_ is now known as takashi | 06:51 | |
| *** SkyRocknRoll has joined #openstack-swift | 06:55 | |
| *** haomaiwa_ has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:01 | |
| eranrom | Greetings, is ho around? | 07:07 |
| *** venkat_p has quit IRC | 07:16 | |
| *** venkat_p has joined #openstack-swift | 07:28 | |
| *** jerrygb has joined #openstack-swift | 07:37 | |
| *** jerrygb has quit IRC | 07:42 | |
| *** mac_ifie_ has quit IRC | 07:44 | |
| *** hseipp has joined #openstack-swift | 07:44 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwa_ has joined #openstack-swift | 08:02 | |
| *** arnox has joined #openstack-swift | 08:05 | |
| *** mahatic has joined #openstack-swift | 08:12 | |
| *** jerrygb has joined #openstack-swift | 08:12 | |
| *** wuhg has joined #openstack-swift | 08:13 | |
| *** jerrygb has quit IRC | 08:17 | |
| ho | eranrom: yeah | 08:42 |
| *** geaaru has joined #openstack-swift | 08:43 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 08:47 | |
| *** sanchitmalhotra has quit IRC | 08:49 | |
| *** zaitcev has quit IRC | 08:50 | |
| *** jordanP has joined #openstack-swift | 08:59 | |
| *** haomaiwa_ has quit IRC | 09:01 | |
| *** haomaiwa_ has joined #openstack-swift | 09:01 | |
| *** ppai has quit IRC | 09:09 | |
| *** mattoliverau has quit IRC | 09:10 | |
| *** matt6434 has joined #openstack-swift | 09:11 | |
| *** openstackgerrit has quit IRC | 09:16 | |
| *** openstackgerrit has joined #openstack-swift | 09:16 | |
| ho | eranrom: i have to leave here now. i will get in touch with you tomorrow same time :-) | 09:17 |
| *** ho has quit IRC | 09:17 | |
| *** ppai has joined #openstack-swift | 09:23 | |
| *** jistr has joined #openstack-swift | 09:44 | |
| *** sanchitmalhotra has joined #openstack-swift | 09:52 | |
| *** acoles_ is now known as acoles | 09:52 | |
| *** sanchitmalhotra1 has quit IRC | 09:55 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 09:57 | |
| *** sanchitmalhotra has quit IRC | 09:59 | |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** haomaiwa_ has joined #openstack-swift | 10:02 | |
| *** admin6 has joined #openstack-swift | 10:03 | |
| *** aix has joined #openstack-swift | 10:05 | |
| acoles | mahatic: hrou https://review.openstack.org/214438 has a +1 from jrichli, are either of you mid-way through a review? would be great to merge it this week. | 10:07 |
| * acoles is trying to reduce the crypto patch chain | 10:07 | |
| acoles | notmyname: sorry! (re dashboard) | 10:09 |
| *** mahatic has quit IRC | 10:12 | |
| *** mahatic has joined #openstack-swift | 10:16 | |
| mahatic | acoles: yes actually, will be done reviewing shortly. thanks for the effort | 10:17 |
| *** kei_yama has quit IRC | 10:17 | |
| acoles | mahatic: great, thank you. no rush - better to do a good review and find all my bugs :) | 10:17 |
| mahatic | acoles: heh, I hope to find bugs too ;) | 10:19 |
| *** silor has quit IRC | 10:20 | |
| *** Guest31166 has quit IRC | 10:23 | |
| *** sanchitmalhotra has joined #openstack-swift | 10:23 | |
| *** sanchitmalhotra1 has quit IRC | 10:26 | |
| *** dimasot has quit IRC | 10:42 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 10:52 | |
| *** matt6434 has quit IRC | 10:54 | |
| *** sanchitmalhotra has quit IRC | 10:54 | |
| *** matt6434 has joined #openstack-swift | 10:55 | |
| openstackgerrit | venkatesh proposed openstack/swift: remove unused parameters in the method https://review.openstack.org/243401 | 10:56 |
| *** haomaiwa_ has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| *** sanchitmalhotra has joined #openstack-swift | 11:03 | |
| *** SkyRocknRoll has quit IRC | 11:05 | |
| *** sanchitmalhotra1 has quit IRC | 11:06 | |
| *** rcernin is now known as rcernin|lunch | 11:13 | |
| *** aix has quit IRC | 11:14 | |
| *** aix has joined #openstack-swift | 11:15 | |
| *** SkyRocknRoll has joined #openstack-swift | 11:21 | |
| *** ppai has quit IRC | 11:23 | |
| *** venkat_p has quit IRC | 11:31 | |
| *** ppai has joined #openstack-swift | 11:34 | |
| *** sanchitmalhotra has quit IRC | 11:34 | |
| *** daemontool has joined #openstack-swift | 11:38 | |
| *** daemontool has quit IRC | 11:40 | |
| *** daemontool has joined #openstack-swift | 11:40 | |
| *** ppai has quit IRC | 11:42 | |
| *** silor has joined #openstack-swift | 11:52 | |
| *** ppai has joined #openstack-swift | 11:57 | |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwa_ has joined #openstack-swift | 12:02 | |
| *** daemontool has quit IRC | 12:08 | |
| *** daemontool has joined #openstack-swift | 12:09 | |
| *** silor1 has joined #openstack-swift | 12:14 | |
| *** silor has quit IRC | 12:15 | |
| *** silor1 is now known as silor | 12:15 | |
| *** jmccarthy has quit IRC | 12:19 | |
| *** jmccarthy has joined #openstack-swift | 12:20 | |
| *** mahatic has quit IRC | 12:24 | |
| *** mahatic has joined #openstack-swift | 12:30 | |
| *** alejandrito has joined #openstack-swift | 12:34 | |
| *** rcernin|lunch is now known as rcernin | 12:36 | |
| *** janonymous__ has joined #openstack-swift | 12:38 | |
| *** wolsen has quit IRC | 12:41 | |
| *** ppai has quit IRC | 12:43 | |
| *** dewsday has joined #openstack-swift | 12:54 | |
| *** dewsday has left #openstack-swift | 12:54 | |
| *** MVenesio has joined #openstack-swift | 12:56 | |
| *** haomaiwa_ has quit IRC | 13:01 | |
| *** haomaiwang has joined #openstack-swift | 13:01 | |
| openstackgerrit | Merged openstack/swift: Remove _keystone_identity method https://review.openstack.org/204050 | 13:02 |
| *** silor has quit IRC | 13:04 | |
| *** marcusvrn_ has quit IRC | 13:05 | |
| *** haomaiwang has quit IRC | 13:11 | |
| *** wolsen has joined #openstack-swift | 13:14 | |
| *** NM has joined #openstack-swift | 13:21 | |
| *** daemontool has quit IRC | 13:22 | |
| *** SkyRocknRoll has quit IRC | 13:23 | |
| *** NM has quit IRC | 13:34 | |
| *** daemontool has joined #openstack-swift | 13:39 | |
| *** haomaiwang has joined #openstack-swift | 13:49 | |
| *** daemontool has quit IRC | 13:56 | |
| *** daemontool has joined #openstack-swift | 13:57 | |
| *** daemontool has quit IRC | 13:58 | |
| *** daemontool has joined #openstack-swift | 13:59 | |
| *** haomaiwang has quit IRC | 14:01 | |
| *** haomaiwa_ has joined #openstack-swift | 14:01 | |
| *** rcernin has quit IRC | 14:03 | |
| *** eranrom has quit IRC | 14:06 | |
| hrou | Hey acoles ! I started taking a peek at that one late last night; I'll be done by tomr morning my time. | 14:14 |
| acoles | hrou: great thanks! | 14:14 |
| *** jerrygb has joined #openstack-swift | 14:23 | |
| *** felixcantournet has joined #openstack-swift | 14:30 | |
| *** willsama has joined #openstack-swift | 14:36 | |
| *** tsg has quit IRC | 14:39 | |
| *** NM has joined #openstack-swift | 14:40 | |
| *** Fuz1on has quit IRC | 14:43 | |
| *** goodygum_ is now known as noqa_v_qoovnie | 14:44 | |
| gmmaha | good morning | 14:48 |
| *** minwoob has joined #openstack-swift | 14:56 | |
| *** jerrygb has quit IRC | 14:56 | |
| *** jerrygb has joined #openstack-swift | 14:57 | |
| *** daddyjoseph97 has joined #openstack-swift | 15:00 | |
| jordanP | guys, is that on purge that the jenkins job "gate-tempest-dsvm-full" in Swift also run the volume and compute tests ? | 15:00 |
| jordanP | *on purpose | 15:00 |
| jordanP | what about running only image and object tests ? | 15:00 |
| *** haomaiwa_ has quit IRC | 15:01 | |
| jordanP | I can take care of that if you want | 15:01 |
| *** haomaiwa_ has joined #openstack-swift | 15:01 | |
| *** trifon has quit IRC | 15:03 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add more functional tests for reverse listings https://review.openstack.org/249262 | 15:03 |
| *** breitz has quit IRC | 15:08 | |
| *** breitz has joined #openstack-swift | 15:08 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add container and account reverse listings https://review.openstack.org/120709 | 15:08 |
| *** pdardeau has joined #openstack-swift | 15:09 | |
| *** daddyjoseph97 has quit IRC | 15:10 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add more functional tests for reverse listings https://review.openstack.org/249262 | 15:12 |
| *** jerrygb has quit IRC | 15:15 | |
| *** jerrygb has joined #openstack-swift | 15:16 | |
| *** sgundur has joined #openstack-swift | 15:18 | |
| *** tsg has joined #openstack-swift | 15:25 | |
| gmmaha | matt6434: notmyname: found the same last night with functests & concurrency.. | 15:25 |
| *** jerrygb has quit IRC | 15:33 | |
| openstackgerrit | Merged openstack/swift-bench: update hacking in test-requirements https://review.openstack.org/243346 | 15:38 |
| *** haomaiwa_ has quit IRC | 15:49 | |
| *** jrichli has joined #openstack-swift | 16:02 | |
| *** geaaru has quit IRC | 16:05 | |
| *** NM has quit IRC | 16:07 | |
| *** sgundur has left #openstack-swift | 16:10 | |
| *** sgundur has joined #openstack-swift | 16:10 | |
| notmyname | good morning | 16:13 |
| *** rcernin has joined #openstack-swift | 16:16 | |
| pchng | morning ! | 16:16 |
| *** wbhuber_ has joined #openstack-swift | 16:22 | |
| *** NM has joined #openstack-swift | 16:24 | |
| *** wbhuber_ has quit IRC | 16:26 | |
| peluse | morning | 16:26 |
| *** wbhuber_ has joined #openstack-swift | 16:27 | |
| *** klrmn1 has joined #openstack-swift | 16:27 | |
| notmyname | oh yay. reverse listings was approved! | 16:27 |
| *** wbhuber_ has quit IRC | 16:31 | |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Update mailmap https://review.openstack.org/249311 | 16:31 |
| acoles | notmyname: i'm assuming your script for AUTHORS works on swiftclient too ;) ^^ | 16:32 |
| notmyname | acoles: I've got one for that too :-) | 16:32 |
| *** sgundur has quit IRC | 16:32 | |
| acoles | notmyname: this time i was going to do it myself but then noticed it was alphabetically sorted - thats too hard for a human ;) | 16:33 |
| notmyname | acoles: just looks like 3 missing/incorrect names right now. not counting the hp.com changes | 16:34 |
| peluse | notmyname: just added a note to the testr patch, it works for me if I add the options "--parallel --concurrency=2". How many CPUs are in your SAIO VM? | 16:35 |
| peluse | =1 I mean, not 2 | 16:35 |
| notmyname | peluse: heresy, I know, to say this to Intel, but my SAIO MV has only 1 vCPU | 16:36 |
| *** sgundur has joined #openstack-swift | 16:36 | |
| peluse | notmyname: oh wow, mine is like 4 :) | 16:36 |
| peluse | notmyname: that probably explains it then | 16:36 |
| peluse | well, mostly | 16:37 |
| *** jrichli has left #openstack-swift | 16:38 | |
| notmyname | I think we should update the patch to hard code that, if it seems to fix it for people | 16:38 |
| *** SkyRocknRoll has joined #openstack-swift | 16:40 | |
| *** aix has quit IRC | 16:40 | |
| peluse | notmyname: yeah, I'll add another note if you didn't already and also reqeust a comment to not change it until someone figures out how to make it work with >1 | 16:41 |
| notmyname | yeah | 16:41 |
| notmyname | peluse: I haven't added anything to it. can you push a patch over it? | 16:41 |
| peluse | yeah sure | 16:41 |
| notmyname | thanks | 16:42 |
| *** pdardeau has quit IRC | 16:42 | |
| *** pdardeau has joined #openstack-swift | 16:43 | |
| *** daemontool has quit IRC | 16:43 | |
| openstackgerrit | paul luse proposed openstack/swift: Modify functional tests to use testr https://review.openstack.org/214206 | 16:44 |
| *** daemontool has joined #openstack-swift | 16:44 | |
| acoles | peluse: notmyname i have only just started looking at the testr change but fwiw swiftclient uses concurrency=1 | 16:45 |
| peluse | bah, new editor looks like it added whitespace | 16:45 |
| *** klrmn1 has quit IRC | 16:47 | |
| *** jrichli has joined #openstack-swift | 16:48 | |
| openstackgerrit | paul luse proposed openstack/swift: Modify functional tests to use testr https://review.openstack.org/214206 | 16:48 |
| *** pdardeau has quit IRC | 16:48 | |
| peluse | phew :) | 16:49 |
| openstackgerrit | Merged openstack/swift: remove Python 2.6-specific logging workaround https://review.openstack.org/245324 | 16:49 |
| *** daemontool has quit IRC | 16:49 | |
| *** pdardeau has joined #openstack-swift | 16:50 | |
| *** jrichli has quit IRC | 16:51 | |
| *** jrichli has joined #openstack-swift | 16:51 | |
| *** admin6 has quit IRC | 16:52 | |
| *** pdardeau has quit IRC | 16:53 | |
| *** diazjf has joined #openstack-swift | 16:53 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Reverse-listings follow-up https://review.openstack.org/248924 | 16:54 |
| timburke | acoles: ^ added the test case, but to be clear: it's just to guard against our future selves, right? test still passed without needing backend changes | 16:55 |
| acoles | timburke: right. i was fiddling with the marker resetting in the original patch and noticed that i could make that change without any test failing, then same with your patch. | 16:57 |
| acoles | timburke: the pastebin change i mean | 16:58 |
| timburke | cool. just making sure :) | 16:58 |
| jrichli | apologies for the enter/leave messages. bear with me for a sec | 17:01 |
| *** jrichli has quit IRC | 17:01 | |
| *** pdardeau has joined #openstack-swift | 17:08 | |
| *** arnox has quit IRC | 17:08 | |
| *** jrichli has joined #openstack-swift | 17:10 | |
| pchng | welcome back jrichli :) | 17:10 |
| notmyname | jrichli: setting up a bouncer? | 17:10 |
| jrichli | YAY! notmyname acoles - yes, bouncer setup done! Thanks to pchng | 17:11 |
| notmyname | nice | 17:11 |
| *** pdardeau_ has joined #openstack-swift | 17:11 | |
| notmyname | are you using znc or did you chose a different one? | 17:11 |
| jrichli | znc | 17:11 |
| pchng | notmyname: znc | 17:11 |
| acoles | jrichli: i thought you already had one? | 17:11 |
| timburke | i find mine quite handy. thanks for setting it up long ago notmyname! | 17:11 |
| notmyname | acoles: no, previously she was just always online ;-) | 17:12 |
| jrichli | acoles: I did things the old fashioned way of reading scrollback on the web | 17:12 |
| acoles | of course! :) | 17:12 |
| hrou | I love mine : ) | 17:15 |
| * jrichli is gonna go to lunch now. feel free to send a private message while away ;-) | 17:16 | |
| hrou | pchng, did you setup znc in softlayer btw ? | 17:17 |
| *** blmartin has joined #openstack-swift | 17:18 | |
| pchng | hrou: Originally I was going to, but I ended up just getting a VM internal to IBM | 17:18 |
| *** klrmn1 has joined #openstack-swift | 17:18 | |
| pchng | hrou: Let me know if you want access :) if you are not already using a bouncer | 17:18 |
| *** mahatic has quit IRC | 17:19 | |
| hrou | pchng, haha me too; Its a bit annoying because you need to log into the VPN if not on site :) FYI I believe wbhuber set it up in softlayer implying there's external access as well : ) | 17:19 |
| *** david-ly_ has joined #openstack-swift | 17:19 | |
| pchng | ah okay, that's great to have that option. but yeah, having it in softlayer for external access is nicer at times | 17:20 |
| pchng | hrou: And i think the VM they gave me is running a GUI/desktop, so I should probably disable that... | 17:20 |
| pchng | (i think i requested ubuntu server ed. but that might have gotten lost) | 17:21 |
| *** zaitcev has joined #openstack-swift | 17:22 | |
| *** ChanServ sets mode: +v zaitcev | 17:22 | |
| *** david-ly_ has quit IRC | 17:23 | |
| *** david-l__ has joined #openstack-swift | 17:23 | |
| *** david-lyle has quit IRC | 17:23 | |
| clayg | heyoh! | 17:26 |
| acoles | clayg: hi! | 17:27 |
| *** david-l__ has quit IRC | 17:30 | |
| *** proteusguy has quit IRC | 17:31 | |
| clayg | hi | 17:33 |
| *** jmccarthy has quit IRC | 17:35 | |
| *** jmccarthy has joined #openstack-swift | 17:35 | |
| *** jordanP has quit IRC | 17:38 | |
| *** CaioBrentano has joined #openstack-swift | 17:40 | |
| *** sayali has quit IRC | 17:42 | |
| *** proteusguy has joined #openstack-swift | 17:44 | |
| *** david-lyle has joined #openstack-swift | 17:44 | |
| *** pdardeau_ has quit IRC | 17:44 | |
| *** jmccarthy has quit IRC | 17:46 | |
| *** pdardeau_ has joined #openstack-swift | 17:48 | |
| *** NM has quit IRC | 17:48 | |
| *** zehicle has quit IRC | 17:48 | |
| openstackgerrit | Merged openstack/swift: Add container and account reverse listings https://review.openstack.org/120709 | 17:54 |
| notmyname | FYI next week I want to finish up a release for swiftclient (b/c of py26) and consider one for swift | 17:56 |
| *** pdardeau has quit IRC | 17:56 | |
| *** aix has joined #openstack-swift | 17:57 | |
| *** pdardeau has joined #openstack-swift | 17:58 | |
| *** pdardeau has quit IRC | 18:00 | |
| *** hseipp has quit IRC | 18:01 | |
| clayg | notmyname: did ring fixes merge? | 18:01 |
| *** eranrom has joined #openstack-swift | 18:01 | |
| *** pdardeau has joined #openstack-swift | 18:01 | |
| notmyname | clayg: patch 241571? not yet | 18:02 |
| patchbot | notmyname: https://review.openstack.org/#/c/241571/ - Put part-replicas where they go | 18:02 |
| notmyname | but has one +2 already (still) | 18:02 |
| *** david-lyle has quit IRC | 18:03 | |
| notmyname | dfg_: hurricanerix: glange: did you have a chance to run your rings against that patch? | 18:03 |
| *** pdardeau has quit IRC | 18:03 | |
| *** jistr has quit IRC | 18:04 | |
| *** pdardeau has joined #openstack-swift | 18:06 | |
| clayg | notmyname: hurricanerix told me he was looking at it (yesterday) | 18:06 |
| notmyname | great :-) | 18:06 |
| *** pdardeau has quit IRC | 18:06 | |
| *** jmccarthy has joined #openstack-swift | 18:06 | |
| *** david-lyle has joined #openstack-swift | 18:11 | |
| *** wasmum has joined #openstack-swift | 18:11 | |
| clayg | notmyname: wow, so we merged reverse listings - that's like a new api feature - that's sorta cool! | 18:13 |
| clayg | timburke: nice work! | 18:13 |
| notmyname | yeah! | 18:13 |
| clayg | probably mattoliver and cschwede too | 18:13 |
| clayg | and acoles | 18:13 |
| clayg | high fives all around | 18:14 |
| timburke | yeah, i love it! so who wants to go look at patch 234391 and second mattoliverau's +2? | 18:15 |
| patchbot | timburke: https://review.openstack.org/#/c/234391/ - Have versioned_writes use reverse listings | 18:15 |
| notmyname | matt, actually :-) | 18:15 |
| openstackgerrit | Travis Nguyen proposed openstack/swift: skip object versioning tests when allow_versioned_writes = False https://review.openstack.org/249354 | 18:25 |
| notmyname | timburke: ^ | 18:29 |
| acoles | clayg: swift_bytes is a pain :/ | 18:30 |
| acoles | clayg: can you think of any reasons why it would be a bad idea to make the object server aware of swift_bytes and update the container server? (right now the object server is blissfully unaware) I am in two minds which way to go with https://review.openstack.org/#/c/248219/ but feel like I want to make swift_bytes be a piece of sysmeta on the object server | 18:32 |
| acoles | torgomatic: you may have an opinion too ^^ - I always associate slo with torgomatic :) | 18:33 |
| timburke | acoles: are we trying to make the container server blissfully unaware instead? if so, i think will cause double-counting in the usage stats | 18:36 |
| *** sgundur has quit IRC | 18:38 | |
| acoles | timburke: no. what i meant was that currently SLO tacks swift_bytes onto content-type and the obj server passes it on to the container server, unaware of the swift_bytes param. But if fast-post is going to update content-type then SLO won't be able to append swift_bytes to POSTed content-type (cos it won't know what the size is during a POST) so I think the object server needs to get involved. | 18:40 |
| acoles | timburke: but yeah the container server always needs to get the value | 18:41 |
| acoles | timburke: and even with post-as-copy its ugly handling the swift_bytes during a POST (thats what I am trying to fix with patch 248219) | 18:41 |
| patchbot | acoles: https://review.openstack.org/#/c/248219/ - Fix listing of SLO manifest bytes after POST or COPY | 18:41 |
| *** pdardeau has joined #openstack-swift | 18:42 | |
| torgomatic | I generally want things to be as isolated as possible, otherwise I can't hold them all in my head | 18:44 |
| torgomatic | others with larger brains may have differing opinions | 18:44 |
| *** jmccarthy has quit IRC | 18:46 | |
| *** jmccarthy has joined #openstack-swift | 18:47 | |
| *** CaioBren_ has joined #openstack-swift | 18:47 | |
| *** CaioBrentano has quit IRC | 18:51 | |
| *** pdardeau has quit IRC | 18:52 | |
| *** pdardeau has joined #openstack-swift | 18:54 | |
| *** pdardeau has quit IRC | 18:55 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Stop staticweb revealing container existence to unauth'd requests https://review.openstack.org/248867 | 18:56 |
| acoles | torgomatic: me too. i keep thinking of that blog you linked about the cost of complexity. | 18:58 |
| *** pdardeau has joined #openstack-swift | 19:00 | |
| torgomatic | acoles: that's the only real reason I have for wanting to keep things out of the object server when possible. if it's necessary, though, I guess we should do it. | 19:00 |
| *** jerrygb has joined #openstack-swift | 19:01 | |
| *** wuhg has quit IRC | 19:01 | |
| clayg | acoles: when I was thinking about ith jrichli I had convinced myself that the slo middleware should set some sysmeta on the PUT request - and that the object server should know how to pass that through to the container update (and save in async's) | 19:01 |
| acoles | clayg: yeah thats where my thinking is heading | 19:02 |
| clayg | acoles: ideally the object server would just sort of blindly allow middleware to add sysmeta to the object updates - so it wouldn't really be swift_bytes specific from the object server POV - then it's for the middlware and the container server to negotiate where the information lives and what is done with it | 19:02 |
| clayg | acoles: i got stumped trying to think of how to do it in rolling upgrade | 19:02 |
| acoles | clayg: of course we have to deal with existing manifests that don't have that sysmeta | 19:03 |
| clayg | it seems like maybe the middleware will have to do both for awhile - and only at some point later when it's sure everyone is upgraded it can stop messing with content-type | 19:03 |
| clayg | acoles: interesting.... | 19:03 |
| *** pdardeau1 has joined #openstack-swift | 19:04 | |
| *** pdardeau1 has quit IRC | 19:04 | |
| clayg | acoles: what does the object server do currently? can I ever see the swift_bytes in the content-type on a GET response for a manifest? | 19:04 |
| acoles | clayg: the obj server has swift_bytes appended to its content_type, but the obj controller strips it off in GET path, so no client doesn't ever see it | 19:05 |
| *** SkyRocknRoll has quit IRC | 19:09 | |
| jrichli | acoles clayg torgomatic: I dont know if this could be in any way related (needing a new solution), but there is another internal param used within the content-type field: "boundary" for sloranges | 19:10 |
| *** alejandrito has quit IRC | 19:10 | |
| openstackgerrit | Merged openstack/python-swiftclient: Update mailmap https://review.openstack.org/249311 | 19:10 |
| torgomatic | jrichli: is that on 206 responses? | 19:10 |
| acoles | clayg: i think the obj server has to store swift_bytes somehow (either asit does now or in sysmeta) so that when the manifest is copied the swift_bytes can be included in update to the destination container server | 19:11 |
| jrichli | togomatic: I don't know. I don't even know if its hidden from the user in the same way. I just know slorange is causing issues with encryption because of this param, and we need to handle it. pchng was working it | 19:11 |
| torgomatic | jrichli: okay. that sounds to me like it might be the multi-range GET response stuff from RFC 7233, but I'm not 100% certain of that. | 19:12 |
| acoles | torgomatic: jrichli so on that subject, mahatic and I were looking at this line yesterday https://github.com/openstack/swift/blob/master/swift/common/swob.py#L1187 | 19:13 |
| *** zehicle has joined #openstack-swift | 19:13 | |
| *** zehicle has quit IRC | 19:13 | |
| acoles | torgomatic: and I question the use of the content_type property there - its getter strips off all the params, like charset, but also our crypto-meta :/ | 19:14 |
| *** zehicle has joined #openstack-swift | 19:14 | |
| *** zehicle has quit IRC | 19:14 | |
| *** zehicle has joined #openstack-swift | 19:14 | |
| *** zehicle has quit IRC | 19:15 | |
| torgomatic | acoles: yeah, that sounds like something to fix in the property class there | 19:15 |
| acoles | torgomatic: do you know, is that deliberate? ie is there a reason to not include charset an dother params? | 19:15 |
| torgomatic | acoles: probably because webob did it? | 19:15 |
| *** zehicle has joined #openstack-swift | 19:15 | |
| *** zehicle has quit IRC | 19:15 | |
| acoles | torgomatic: the property getter does seem to conform to the webob api - i was thinking that that ine should just fetch the content-type header and not use the property attr | 19:16 |
| *** zehicle has joined #openstack-swift | 19:16 | |
| acoles | s/ine/line/ | 19:16 |
| torgomatic | there's a couple weird API decisions in swob that are there because they mimic webob | 19:16 |
| *** zehicle has quit IRC | 19:16 | |
| *** zehicle has joined #openstack-swift | 19:16 | |
| torgomatic | I know there's one where ranges are (start_byte, end_byte + 1) instead of (start_byte, end_byte) because webob | 19:16 |
| torgomatic | this is probably another one | 19:16 |
| acoles | looks that way | 19:16 |
| *** pdardeau has quit IRC | 19:16 | |
| *** zehicle has quit IRC | 19:16 | |
| *** zehicle has joined #openstack-swift | 19:17 | |
| *** zehicle has quit IRC | 19:17 | |
| *** zehicle has joined #openstack-swift | 19:18 | |
| *** zehicle has quit IRC | 19:18 | |
| *** zehicle has joined #openstack-swift | 19:18 | |
| *** zehicle has quit IRC | 19:18 | |
| acoles | torgomatic: ok maybe that could be changed on master | 19:19 |
| *** zehicle has joined #openstack-swift | 19:19 | |
| pchng | acoles: Is this the content-type stripping you're referring to? https://github.com/openstack/swift/blob/master/swift/common/swob.py#L403 | 19:19 |
| *** zehicle has quit IRC | 19:19 | |
| torgomatic | acoles: yep, http://docs.webob.org/en/1.1/modules/webob.html says that the content_type property strips params | 19:19 |
| acoles | pchng: yes | 19:19 |
| *** zehicle has joined #openstack-swift | 19:20 | |
| *** sgundur has joined #openstack-swift | 19:20 | |
| *** zehicle has quit IRC | 19:20 | |
| acoles | torgomatic: yep, and use the charset property if you really want to get that, and if you want other params, well...tough :) | 19:20 |
| *** zehicle has joined #openstack-swift | 19:20 | |
| torgomatic | heh, webob doesn't even have a charset property :) | 19:20 |
| pchng | acoles, torgomatic: I was looking into some functests (GETs for SLOs that had ranged manifests) that were filing, possibly because of this stripping. | 19:20 |
| *** zehicle has quit IRC | 19:20 | |
| torgomatic | oh no, wait, it's in 1.2 | 19:21 |
| *** zehicle has joined #openstack-swift | 19:21 | |
| torgomatic | or 1.1 and I can't read. ANYWAY | 19:21 |
| *** zehicle has quit IRC | 19:21 | |
| acoles | heh, yeah its there | 19:21 |
| *** zehicle has joined #openstack-swift | 19:21 | |
| pchng | acoles, torgomatic: essentially it looks like the content-type should have had "multipart/byteranges; boundary=<something>" but only "multipart/byteranges" was showing up | 19:22 |
| *** zehicle has quit IRC | 19:22 | |
| acoles | pchng: failing on master? | 19:22 |
| torgomatic | point being that _prepare_for_ranges could use self.headers['Content-Type'] and remove any boundary that's in there, then put that in each part's header | 19:22 |
| torgomatic | (assuming I'm understanding the bug) | 19:22 |
| *** zehicle has joined #openstack-swift | 19:22 | |
| *** zehicle has quit IRC | 19:22 | |
| pchng | acoles: No, on feature/crypto. I think after a merge from master, maybe last week? Details in trello here: https://trello.com/c/rPjrIefY/61-fix-functests-failures-in-slo-ranged-gets | 19:23 |
| *** zehicle has joined #openstack-swift | 19:23 | |
| *** zehicle has quit IRC | 19:23 | |
| *** zehicle has joined #openstack-swift | 19:23 | |
| jrichli | pchng acoles: slorange and crypto combo has always failed those tests | 19:24 |
| *** zehicle has quit IRC | 19:24 | |
| *** zehicle has joined #openstack-swift | 19:24 | |
| openstackgerrit | John Dickinson proposed openstack/swift: Add support for storage policies to have more than one name https://review.openstack.org/241978 | 19:24 |
| acoles | torgomatic: yep. actually i hadn't considered that the source content-type might have a boundary | 19:24 |
| *** zehicle has quit IRC | 19:24 | |
| acoles | pchng: oic | 19:24 |
| clayg | acoles: so is this week a holiday for you guys? | 19:25 |
| *** zehicle has joined #openstack-swift | 19:25 | |
| pchng | jrichli: I think you said in trello that after the merge, these two more were failing, the ones specifically related to GETs for SLO that had ranged manifests. But yes I think the other SLO ones were failing before that | 19:25 |
| clayg | acoles: do you have any other fall holiday between halloween and christmas? | 19:25 |
| *** zehicle has quit IRC | 19:25 | |
| jrichli | pchng: oh, really? I'll take your word for it :-) | 19:25 |
| acoles | clayg: nope. we have nothing to give thanks for, we were left behind !) | 19:25 |
| *** zehicle has joined #openstack-swift | 19:26 | |
| acoles | clayg: and nope, in fact halloween isn't so big here either | 19:26 |
| *** zehicle has quit IRC | 19:26 | |
| jrichli | acoles: what about the cake you get at work all the time? | 19:26 |
| acoles | oh yeah | 19:26 |
| *** zehicle has joined #openstack-swift | 19:26 | |
| *** zehicle has quit IRC | 19:27 | |
| acoles | clayg: thanks for working on that staticweb thing btw. looks like i missed something there. | 19:27 |
| clayg | acoles: well luckily for you and me both cschwede has his eyes open | 19:27 |
| acoles | pchng: hmmm, could be related. my take away is "read the docstring before using the content_type property on swob.Response" | 19:28 |
| clayg | this is the worst procrastination ever - I want to start reviewing concurrent-gets - but I keep finding other reviews that'll be like "quick" so I wanna knock those out first :'( | 19:29 |
| *** zehicle has joined #openstack-swift | 19:29 | |
| *** zehicle has quit IRC | 19:29 | |
| acoles | i know what you mean, had to overcome that today to get stuck into reverse listings | 19:30 |
| pchng | acoles: thanks, that will help with my investigation - still learning about SLO/DLO, etc. | 19:30 |
| acoles | clayg: but after doing a 'longer' review i rewarded myself with changing a mailmap file :) | 19:31 |
| *** zehicle has joined #openstack-swift | 19:32 | |
| *** zehicle has quit IRC | 19:32 | |
| *** zehicle has joined #openstack-swift | 19:33 | |
| *** zehicle has quit IRC | 19:33 | |
| timburke | and thank you for the longer review acoles! | 19:33 |
| clayg | timburke: acoles is the best | 19:33 |
| *** zehicle has joined #openstack-swift | 19:33 | |
| acoles | timburke: i felt i was getting out of practice ;) | 19:33 |
| *** zehicle has quit IRC | 19:34 | |
| *** zehicle has joined #openstack-swift | 19:34 | |
| *** zehicle has quit IRC | 19:34 | |
| *** zehicle has joined #openstack-swift | 19:35 | |
| *** zehicle has quit IRC | 19:35 | |
| *** zehicle has joined #openstack-swift | 19:35 | |
| *** zehicle has quit IRC | 19:36 | |
| *** zehicle has joined #openstack-swift | 19:36 | |
| *** zehicle has quit IRC | 19:36 | |
| *** zehicle has joined #openstack-swift | 19:37 | |
| *** zehicle has quit IRC | 19:37 | |
| *** zehicle has joined #openstack-swift | 19:37 | |
| acoles | good night, or in honour of patch 120709 i should say "night good" | 19:38 |
| *** zehicle has quit IRC | 19:38 | |
| patchbot | acoles: https://review.openstack.org/#/c/120709/ - Add container and account reverse listings | 19:38 |
| *** zehicle has joined #openstack-swift | 19:38 | |
| jrichli | acoles : cute. good night! :-) | 19:38 |
| *** zehicle has quit IRC | 19:38 | |
| *** zehicle has joined #openstack-swift | 19:39 | |
| *** zehicle has quit IRC | 19:39 | |
| acoles | notmyname: have a great thanksgiving (and anyone else who is heading off soon) | 19:39 |
| *** zehicle has joined #openstack-swift | 19:39 | |
| notmyname | thanks | 19:39 |
| *** zehicle has quit IRC | 19:40 | |
| *** acoles is now known as acoles_ | 19:40 | |
| *** zehicle has joined #openstack-swift | 19:40 | |
| notmyname | zehicle: you're spamming the channel with join/part messages | 19:40 |
| *** zehicle has quit IRC | 19:40 | |
| clayg | notmyname: filter that noise! | 19:40 |
| *** zehicle has joined #openstack-swift | 19:41 | |
| *** zehicle has quit IRC | 19:41 | |
| *** ChanServ sets mode: +o notmyname | 19:41 | |
| *** zehicle has joined #openstack-swift | 19:41 | |
| *** zehicle was kicked by notmyname (zehicle) | 19:41 | |
| *** jmccarthy has quit IRC | 19:41 | |
| notmyname | clayg: done. seems to have worked | 19:45 |
| *** notmyname sets mode: -o notmyname | 19:45 | |
| clayg | notmyname: oh i just ment on your client - but that's coo | 19:50 |
| clayg | my favorite part of reviewing concurrent-gets is the meta-hacking on dfg's slowdown | 20:00 |
| *** CaioBren_ has quit IRC | 20:02 | |
| openstackgerrit | Merged openstack/swift: Reverse-listings follow-up https://review.openstack.org/248924 | 20:03 |
| *** sgundur has left #openstack-swift | 20:04 | |
| *** CaioBrentano has joined #openstack-swift | 20:05 | |
| *** sgundur has joined #openstack-swift | 20:06 | |
| *** zhill has joined #openstack-swift | 20:08 | |
| *** CaioBrentano has quit IRC | 20:09 | |
| *** changbl has quit IRC | 20:11 | |
| hurricanerix | notmyname yeah, i am playing around with it this week. (and by this week, i mean today and tomorrow) :) | 20:19 |
| notmyname | :-) | 20:19 |
| *** haypo has joined #openstack-swift | 20:20 | |
| haypo | tsg: hi. any progress on https://review.openstack.org/#/c/247936/ "Add liberasurecode to CI slaves for Swift/PyECLib" (and https://review.openstack.org/#/c/244287/ "Bump PyECLib requirement to >=1.0.7")? | 20:21 |
| tsg | haypo: the latest status is on the review pages | 20:22 |
| tsg | haypo: we are waiting on the F22 and F23 packages to appear in updates | 20:22 |
| tsg | haypo: the rest is all taken care of | 20:22 |
| tsg | haypo: zaitcev has this pending - https://bodhi.fedoraproject.org/updates/FEDORA-2015-3889c3fcba | 20:23 |
| tsg | haypo: may be you can help with testing and add to karma. https://bodhi.fedoraproject.org/updates/FEDORA-2015-3889c3fcba and https://bodhi.fedoraproject.org/updates/FEDORA-2015-1ab5c917e2 both | 20:23 |
| zaitcev | tsg, haypo: no idea how a request to pending could be pending for so long | 20:23 |
| zaitcev | er. request to testing | 20:24 |
| zaitcev | so while this is going on, I am unable to build pyeclib | 20:24 |
| tsg | zaitcev: can you please file an issue so Kevin also sees the error | 20:25 |
| haypo | zaitcev: sorry, what is pending? | 20:25 |
| zaitcev | haypo: Once the build system has produced the build, e.g. http://koji.fedoraproject.org/koji/buildinfo?buildID=701091, a magic process occurs that copies the resulting RPMs into the master of mirrors and updates repo (a number of gigantic XML files *.xml.gz). Then dnf/yum are able to see that package even exists, and whem mirrors fetch it they can download. | 20:29 |
| zaitcev | A certain ticket is filed to initiate the magic. Buld system files automatically for (none)->testing, while I have to file testing->stable. While the magic hasn't completed, the ticket is "pending". | 20:30 |
| zaitcev | Should usually take a few minutes, but it's 21 hours now. | 20:30 |
| *** changbl has joined #openstack-swift | 20:31 | |
| haypo | zaitcev: oh | 20:32 |
| zaitcev | Infra says about my build "it's in the current push that I started 3 hours ago" | 20:33 |
| zaitcev | Time for coffee it appears | 20:33 |
| openstackgerrit | Tim Burke proposed openstack/swift: Have versioned_writes use reverse listings https://review.openstack.org/234391 | 20:41 |
| *** eranrom has quit IRC | 20:44 | |
| haypo | zaitcev, tsg : i tested liberasurecode 1.1 package: it works, PyEClib tests pass with the package. but i failed to post a comment on https://bodhi.fedoraproject.org/updates/FEDORA-2015-3889c3fcba | 20:45 |
| zaitcev | Yeah, it's locked while the aforementioned magic is active. | 20:45 |
| haypo | nevermind, i succeeded to vote +1 | 20:46 |
| zaitcev | haypo: oooooh | 20:46 |
| haypo | zaitcev: i see four +1 votes on https://bodhi.fedoraproject.org/updates/FEDORA-2015-3889c3fcba but the Karma is at 2? | 20:47 |
| *** zhill has quit IRC | 20:48 | |
| haypo | Patch Set 5: | 20:48 |
| haypo | (oops) | 20:48 |
| haypo | tsg: i also pinged Clark Boylan on #openstack-infra :) | 20:49 |
| *** zhill has joined #openstack-swift | 20:49 | |
| *** matt6434 is now known as mattoliverau | 20:52 | |
| *** ChanServ sets mode: +v mattoliverau | 20:52 | |
| tsg | haypo: I talked to clarkb there yesterday - we are holding off until the package is available in F22 (https://bodhi.fedoraproject.org/updates/FEDORA-2015-1ab5c917e2) | 20:54 |
| tsg | zaitcev: might need you help with the F22 update as well | 20:54 |
| zaitcev | tsg: oh god | 20:54 |
| tsg | zaitcev: the devstack folks need it | 20:54 |
| *** NM has joined #openstack-swift | 20:54 | |
| haypo | zaitcev: we all rely on you | 20:54 |
| tsg | zaitcev: :) sorry, doesn't have to be an update - we can simply push the existing 1.0.9 pkg to F22 - see the activity on the F22 ticket | 20:55 |
| janonymous__ | Hi , a quick question do we need to mention upgraded tox version for py3 changes in tox file: http://paste.openstack.org/show/479917/ also? | 20:56 |
| haypo | zaitcev: ah, tsg told me that i have to log in to vote. otherwise, my vote doesn't count | 20:57 |
| haypo | zaitcev: would a +1 vote change something? | 20:57 |
| zaitcev | it changes the move to stable so you don't need to wait for 3 days in testing | 20:58 |
| *** CaioBrentano has joined #openstack-swift | 21:00 | |
| *** tsg has quit IRC | 21:07 | |
| haypo | zaitcev: i got an error from https://admin.fedoraproject.org/pkgdb/package/liberasurecode/ before and now i got an error when i try to create a fedora account :-/ | 21:07 |
| *** tim_ has joined #openstack-swift | 21:09 | |
| *** zhill has quit IRC | 21:10 | |
| openstackgerrit | Janie Richling proposed openstack/swift: WIP - Support for http footers - Replication and EC https://review.openstack.org/237393 | 21:10 |
| *** tsg has joined #openstack-swift | 21:11 | |
| janonymous__ | *Away | 21:12 |
| *** janonymous__ has quit IRC | 21:12 | |
| haypo | zaitcev, tsg : ooook, i succeded to vote +1 and increase the karma to +3: https://bodhi.fedoraproject.org/updates/FEDORA-2015-3889c3fcba | 21:14 |
| zaitcev | tsg wanted https://bodhi.fedoraproject.org/updates/FEDORA-2015-1ab5c917e2 I think | 21:15 |
| tsg | haypo: yes, if you follow the same steps on F22 and update https://bodhi.fedoraproject.org/updates/FEDORA-2015-1ab5c917e2 that will be great | 21:15 |
| haypo | tsg: i don't have Fedora 22 | 21:16 |
| zaitcev | So Haikel submitted liberasurecode-1.0.9.fc22 to stable. | 21:16 |
| zaitcev | 7 hours ag | 21:16 |
| *** djc has joined #openstack-swift | 21:17 | |
| djc | how do you find out what the partition power of a swift cluster is? Is there a config file that contains that information. | 21:18 |
| tsg | zaitcev: I see it now - thanks. Is the package expected to land in stable by tonight? | 21:18 |
| zaitcev | tsg: I would expect so. | 21:19 |
| zaitcev | djc: run swift-ring-build xxxxx.builder without any other arguments | 21:19 |
| haypo | zaitcev, tsg : why not 1.1 for fc22? | 21:19 |
| tsg | zaitcev: ok .. I will let the infra guys know so patch 238926 can go in | 21:19 |
| patchbot | tsg: https://review.openstack.org/#/c/238926/ - EC: Use ec_type=liberasurecode_rs_vand in unit tests | 21:19 |
| tsg | haypo: we simply need "a version" available on F22 for now | 21:20 |
| tsg | haypo: update is the next step | 21:20 |
| haypo | tsg: ah | 21:20 |
| *** minwoob has quit IRC | 21:25 | |
| *** badari has joined #openstack-swift | 21:26 | |
| *** blmartin has quit IRC | 21:26 | |
| *** djc has quit IRC | 21:28 | |
| haypo | hey, i'm looking for reviews on my python3 patches. https://review.openstack.org/#/c/237039/ https://review.openstack.org/#/c/237005/ https://review.openstack.org/#/c/237019/ already got +2 from samuel merritt who looks to be a good reviewer since he usually complains on most of my changes :-D | 21:28 |
| *** zhill has joined #openstack-swift | 21:28 | |
| haypo | see https://review.openstack.org/#/q/owner:%22Victor+Stinner%22+status:open+project:openstack/swift,n,z for all my pending swift patches for py3 | 21:29 |
| notmyname | lifeless: were you looking for me for something? | 21:31 |
| haypo | zaitcev: ^^ | 21:32 |
| zaitcev | haypo: I must focus on fast-post | 21:33 |
| haypo | zaitcev: what is fast-post? | 21:33 |
| zaitcev | haypo: I think it's this https://review.openstack.org/135380 | 21:34 |
| *** NM has quit IRC | 21:37 | |
| *** tim_ has quit IRC | 21:38 | |
| *** NM has joined #openstack-swift | 21:40 | |
| *** blmartin has joined #openstack-swift | 21:42 | |
| *** wbhuber_ has joined #openstack-swift | 21:42 | |
| *** blmartin has quit IRC | 21:43 | |
| *** wbhuber_ has quit IRC | 21:55 | |
| lifeless | notmyname: I wanted to cc you on a cross project spec I've got | 21:55 |
| lifeless | notmyname: about actually testing backwards compat of clients (and libraries, but for you, clients are the key) | 21:55 |
| notmyname | ok, thanks. I saw it | 21:56 |
| notmyname | (saw that it exists) | 21:56 |
| notmyname | lifeless: I too have problems searching for people in gerrit. often the person doesn't show up and I have to search my email hoping I can find an email address | 21:57 |
| mattoliverau | Morning | 21:58 |
| *** haypo has left #openstack-swift | 22:00 | |
| *** MVenesio has quit IRC | 22:04 | |
| *** zhill has quit IRC | 22:04 | |
| *** felixcantournet has quit IRC | 22:05 | |
| clayg | sigh - and things were going so well :'( | 22:05 |
| *** zhill has joined #openstack-swift | 22:05 | |
| clayg | mattoliverau: I'm seeing a 404 at the end of a GET again? | 22:05 |
| clayg | mattoliverau: I would have swore you fixed that in some previous rev | 22:05 |
| mattoliverau | Yeah, that sounds familiar | 22:07 |
| clayg | mattoliverau: like if your concurrency == replica count shouldn't it be inflight < concurrency? otherwise you can have replica count requests in flight and then still only wait concurrency_timeout instead of node_timeout | 22:09 |
| clayg | like if inflight == replicas - I think you wait node timeout | 22:10 |
| clayg | (don't trust me on any of this - i'm just thinking outload) | 22:10 |
| clayg | I mean you an trust me on the 404's - i'm seeing the 404's and maybe they're justified or not - maybe i'm doing something stupid - maybe it's a bug - who knows! | 22:10 |
| clayg | i'll just try chaing the <= to < (what could go wrong) | 22:10 |
| mattoliverau | Lol | 22:11 |
| mattoliverau | You could be right, this conversation seems familiar. I need to go stare at the code again | 22:13 |
| clayg | well... it got slower :'( | 22:13 |
| clayg | I think pile.waitall(timeout) is waiting up to the timeout and if two happen to finish it's fine with that - it's not like - well idk | 22:14 |
| clayg | crap | 22:14 |
| *** jerrygb has quit IRC | 22:14 | |
| clayg | ok, yeah waitall definately waits up to timeout | 22:14 |
| mattoliverau | The other major change in that patchset is we now wait conn_timeout by default rather then node timeout. But if your using 0 concurrency_timeout then your bypassing that | 22:15 |
| *** jerrygb has joined #openstack-swift | 22:15 | |
| clayg | mattoliverau: yeah make everyone faster by default - seems good to me - if it works like a boss | 22:16 |
| clayg | mattoliverau: i'm not sure I buy the concurrent_requests t/f always use replica_count thing | 22:16 |
| mattoliverau | Lol, yeah if | 22:16 |
| mattoliverau | We could t/f/# | 22:17 |
| clayg | oh no - now I made it faster | 22:26 |
| clayg | mattoliverau: it was dfg that thought it was a potential for mis-config yeah? | 22:27 |
| clayg | mattoliverau: like I'm not really exactly sure what someone would expect if they configured concurrency for 4 on a three replica ring - i guess it'd dig for the handoff sooner - which could be what you want if you're running hummingbird and PUT's get 503's to handoffs more often because of device limit checks | 22:28 |
| clayg | the real problem is just that it's not configurable per-policy/ring unless you just make it t/f and turn it into replica | 22:29 |
| lifeless | notmyname: I found out today - it is case sensitive | 22:29 |
| lifeless | notmyname: "john d" -> you are not there. "John D" -> you are there. | 22:29 |
| notmyname | we should probably agree between hummingbird and python object servers that a response to "hey I'm still up just dealing with other stuff" probably should be something other than 503/507 | 22:29 |
| notmyname | lifeless: interesting | 22:29 |
| clayg | I think the crazy handoff node count has mathy stuff like "replica / 2" in the dsl | 22:29 |
| clayg | notmyname: everyone says that (I said that) but it's so trivial to fix it's not even worth discussing - it's like what color to paint the bikeshead | 22:30 |
| notmyname | clayg: blue! no, yellow! | 22:31 |
| clayg | notmyname: when something like that gets implemented in python I'm not going to +2 it w/o a decent error code and a test | 22:31 |
| clayg | then if you can run that test on hummingbird it's like 10 seconds to get it passing - yay! | 22:32 |
| notmyname | clayg: yeah, I get your point, but that doesn't mean actually changing it (ie code/tests) shouldn't happen. since we're spread all over, it's not jsut like "hey redbo do you hate code 4xx for this?" | 22:32 |
| notmyname | I just think we need to spend 5 minutes talking about it, and that's about it | 22:32 |
| clayg | notmyname: let's do it in the review | 22:32 |
| notmyname | :-) | 22:33 |
| clayg | mattoliverau: neway - I think i'm cool to just go with a boolean for now where truthy means replica - we can probably expand on that later if needed without breaking backwards compat - so I'm good with that I htink | 22:34 |
| zaitcev | tsg: "hguemar's liberasurecode-1.0.9-3.fc22 bodhi update completed push to stable" | 22:36 |
| tsg | zaitcev: awesome! thanks for the note. I will let -infra folks know so we can get the package into CI | 22:37 |
| *** travisn has quit IRC | 22:38 | |
| tsg | zaitcev, should the F22 package appear in "updates" for "updates-testing"? | 22:40 |
| *** zhill has quit IRC | 22:41 | |
| *** sjmc7 has joined #openstack-swift | 22:42 | |
| openstackgerrit | Travis Nguyen proposed openstack/swift: skip object versioning tests when allow_versioned_writes = False https://review.openstack.org/249354 | 22:51 |
| notmyname | you know what would be cool? reverse listings in swiftclient | 22:51 |
| mattoliverau | clayg: it could be a good parameter to add at the policy level one day | 22:54 |
| *** wuhg has joined #openstack-swift | 22:57 | |
| *** rohit_ has joined #openstack-swift | 22:58 | |
| openstackgerrit | Steve McLellan proposed openstack/swift: WIP Oslo.messaging middleware https://review.openstack.org/249471 | 23:01 |
| *** sgundur has quit IRC | 23:02 | |
| *** NM has quit IRC | 23:04 | |
| *** km has joined #openstack-swift | 23:09 | |
| *** km is now known as Guest59208 | 23:09 | |
| *** diazjf has quit IRC | 23:11 | |
| *** TravT has joined #openstack-swift | 23:27 | |
| *** kei_yama has joined #openstack-swift | 23:29 | |
| *** mac_ified has joined #openstack-swift | 23:29 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 23:31 |
| *** willsama has quit IRC | 23:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 23:32 |
| *** ho has joined #openstack-swift | 23:34 | |
| clayg | man, swift-bench is failing to pip install because of some TypeError: must be type, not classobj in setuptools :'( | 23:37 |
| clayg | notmyname: i said that in the review | 23:37 |
| clayg | notmyname: well in the review of the reverse listings | 23:37 |
| notmyname | :-) | 23:37 |
| clayg | notmyname: it doesn't have --limit either | 23:38 |
| notmyname | yeah, I mostly said it to write it down somewhere | 23:38 |
| clayg | or --path | 23:38 |
| clayg | notmyname: and now it's in irc logs forever! | 23:38 |
| notmyname | timburke has been griping about that for a while | 23:38 |
| notmyname | who cares about --path ;-) | 23:38 |
| clayg | notmyname: fair | 23:38 |
| timburke | clayg: what, | sort -r isn't good enough for you? :P | 23:41 |
| * notmyname steps away for a bit | 23:41 | |
| *** rcernin has quit IRC | 23:42 | |
| mattoliverau | lol | 23:52 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!