| *** sanchitmalhotra1 has joined #openstack-swift | 00:01 | |
| *** changbl has joined #openstack-swift | 00:03 | |
| *** sanchitmalhotra has quit IRC | 00:03 | |
| *** jrichli has joined #openstack-swift | 00:11 | |
| *** sanchitmalhotra has joined #openstack-swift | 00:14 | |
| *** sanchitmalhotra1 has quit IRC | 00:16 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 00:20 | |
| *** sanchitmalhotra has quit IRC | 00:22 | |
| *** sanchitmalhotra has joined #openstack-swift | 00:41 | |
| *** sanchitmalhotra1 has quit IRC | 00:43 | |
| *** peluse1 has joined #openstack-swift | 00:46 | |
| *** asettle has quit IRC | 00:54 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 00:58 | |
| *** sanchitmalhotra has quit IRC | 01:00 | |
| *** sanchitmalhotra has joined #openstack-swift | 01:01 | |
| *** sanchitmalhotra1 has quit IRC | 01:02 | |
| *** nexusz99 has joined #openstack-swift | 01:03 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 01:10 | |
| *** sanchitmalhotra has quit IRC | 01:12 | |
| *** sanchitmalhotra has joined #openstack-swift | 01:17 | |
| *** sanchitmalhotra2 has joined #openstack-swift | 01:19 | |
| *** sanchitmalhotra1 has quit IRC | 01:19 | |
| *** sanchitmalhotra has quit IRC | 01:21 | |
| *** sanchitmalhotra has joined #openstack-swift | 01:40 | |
| *** sanchitmalhotra2 has quit IRC | 01:42 | |
| *** rodrigod` has joined #openstack-swift | 01:54 | |
| *** rodrigod` is now known as rodrigods | 01:57 | |
| *** 6JTABTZX7 has joined #openstack-swift | 02:02 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 02:09 | |
| *** peluse1 has quit IRC | 02:11 | |
| *** sanchitmalhotra has quit IRC | 02:11 | |
| *** wbhuber has quit IRC | 02:14 | |
| *** logan2 has quit IRC | 02:18 | |
| *** sanchitmalhotra has joined #openstack-swift | 02:25 | |
| *** sanchitmalhotra1 has quit IRC | 02:27 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 02:36 | |
| *** sanchitmalhotra has quit IRC | 02:38 | |
| notmyname | ho: good mornign! did you make it back home ok? | 02:39 |
|---|---|---|
| *** haigang has joined #openstack-swift | 02:39 | |
| *** hezhiqiang has joined #openstack-swift | 02:45 | |
| *** haigang has quit IRC | 02:46 | |
| *** hezhiqiang has quit IRC | 02:47 | |
| jrichli | notmyname: good to see you made it back ok. Now I am worried about ho :-\ ! | 02:50 |
| notmyname | :-) | 02:50 |
| ho | hotmyname: yeah, i don't have any jet lag. thanks! :-) | 02:50 |
| notmyname | turns out he was right behind me the whole time! | 02:50 |
| jrichli | ho: nice to hear! | 02:50 |
| notmyname | ho: great! | 02:50 |
| jrichli | you guys are at the ops thing? | 02:51 |
| notmyname | that starts tuesday. (and is just a short drive away) | 02:51 |
| jrichli | ah. ok | 02:51 |
| jrichli | so ... I guess I dont like all mezcal :-) | 02:52 |
| jrichli | my intuition failed me on that one | 02:52 |
| notmyname | lol | 02:52 |
| notmyname | sorry about that | 02:52 |
| jrichli | thats ok - I got the casa noble | 02:53 |
| notmyname | what did the new one taste like? | 02:53 |
| jrichli | rubber! | 02:53 |
| notmyname | lol | 02:53 |
| notmyname | ok, I can help with that actually | 02:53 |
| notmyname | depending on what else you have | 02:53 |
| jrichli | ooo .. well, its gone. i actually am suspecting it is not meant for human consumtion. | 02:54 |
| notmyname | lol | 02:54 |
| notmyname | ok, then | 02:54 |
| jrichli | its ok. i just need to find that one i liked the first night. what was that exactly? | 02:55 |
| jrichli | the illegal | 02:55 |
| jrichli | i remember | 02:55 |
| *** logan2 has joined #openstack-swift | 02:57 | |
| notmyname | ya http://www.ilegalmezcal.com | 02:57 |
| jrichli | thx | 03:00 |
| *** 6JTABTZX7 has quit IRC | 03:01 | |
| notmyname | one of the first mezcals I had smelled like burning tires. and tasted similar. but I learned more, tasted more, and figured out what was good. | 03:02 |
| *** haomaiwang has joined #openstack-swift | 03:02 | |
| jrichli | I dont think I will just pick one without a recommendation next time! | 03:04 |
| notmyname | happy to help, whenever | 03:05 |
| *** dbarman has quit IRC | 03:05 | |
| *** peluse1 has joined #openstack-swift | 03:25 | |
| *** jrichli has quit IRC | 03:42 | |
| *** SkyRocknRoll has joined #openstack-swift | 03:43 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:45 | |
| *** sanchitmalhotra1 has quit IRC | 03:47 | |
| *** mahatic has joined #openstack-swift | 03:53 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-swift | 04:01 | |
| *** hrou has quit IRC | 04:03 | |
| *** changbl has quit IRC | 04:04 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 04:10 | |
| *** sanchitmalhotra has quit IRC | 04:12 | |
| *** sanchitmalhotra has joined #openstack-swift | 04:15 | |
| *** openstack has joined #openstack-swift | 04:19 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (GET) with Keystone https://review.openstack.org/202994 | 04:25 |
| *** peluse1 has quit IRC | 04:36 | |
| *** baojg has joined #openstack-swift | 04:48 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (HEAD) with Keystone https://review.openstack.org/202997 | 04:49 |
| *** haomaiwa_ has joined #openstack-swift | 04:49 | |
| *** haomaiwang has quit IRC | 04:49 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (POST) with Keystone https://review.openstack.org/202999 | 05:00 |
| *** haomaiwa_ has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-swift | 05:02 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (OPTIONS) with Keystone https://review.openstack.org/203012 | 05:08 |
| *** SkyRocknRoll_ has joined #openstack-swift | 05:13 | |
| *** silor has joined #openstack-swift | 05:24 | |
| *** silor has quit IRC | 05:24 | |
| *** silor has joined #openstack-swift | 05:28 | |
| *** janonymous has joined #openstack-swift | 05:31 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwa_ has joined #openstack-swift | 06:02 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (PUT ACL) with Keystone https://review.openstack.org/213608 | 06:24 |
| *** mahatic has quit IRC | 06:29 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (PUT ACL) with Keystone https://review.openstack.org/213608 | 06:30 |
| *** baojg has quit IRC | 06:39 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (PUT ACL) with Keystone https://review.openstack.org/213608 | 06:47 |
| *** geaaru has joined #openstack-swift | 06:54 | |
| *** sileht has quit IRC | 06:55 | |
| ho | gerrit is quite slow... | 06:55 |
| *** sileht has joined #openstack-swift | 06:56 | |
| *** haomaiwa_ has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:02 | |
| *** rledisez has joined #openstack-swift | 07:02 | |
| *** baojg has joined #openstack-swift | 07:03 | |
| -openstackstatus- NOTICE: Gerrit is currently under very high load and may be unresponsive. infra are looking into the issue. | 07:07 | |
| *** mahatic has joined #openstack-swift | 07:22 | |
| *** haomaiwa_ has joined #openstack-swift | 07:40 | |
| *** haomaiwang has quit IRC | 07:40 | |
| *** haomaiwa_ has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-swift | 08:02 | |
| *** mahatic has quit IRC | 08:18 | |
| *** akle has joined #openstack-swift | 08:21 | |
| *** jistr has joined #openstack-swift | 08:21 | |
| *** baojg has quit IRC | 08:23 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (PUT ACL) with Keystone https://review.openstack.org/213608 | 08:27 |
| *** fbo has joined #openstack-swift | 08:29 | |
| *** mahatic has joined #openstack-swift | 08:30 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (DELETE ACL) with Keystone https://review.openstack.org/213629 | 08:36 |
| *** baojg has joined #openstack-swift | 08:39 | |
| *** haomaiwang has quit IRC | 08:43 | |
| *** haomaiwang has joined #openstack-swift | 08:44 | |
| *** ho has quit IRC | 08:44 | |
| *** kei_yama has quit IRC | 08:52 | |
| *** jordanP has joined #openstack-swift | 08:57 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-swift | 09:02 | |
| *** SkyRocknRoll_ has quit IRC | 09:09 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient: Convert http response(byte string) to utf8 string. https://review.openstack.org/213640 | 09:11 |
| *** DaveBCN has joined #openstack-swift | 09:12 | |
| *** akle has quit IRC | 09:17 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient: Convert http response(byte string) to utf8 string in python3. https://review.openstack.org/213640 | 09:19 |
| *** akle|2 has joined #openstack-swift | 09:26 | |
| *** SkyRocknRoll has quit IRC | 09:29 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 09:31 | |
| *** SkyRocknRoll has joined #openstack-swift | 09:43 | |
| *** silor has quit IRC | 09:43 | |
| *** Guest47951 is now known as d0ugal | 09:49 | |
| *** d0ugal has quit IRC | 09:49 | |
| *** d0ugal has joined #openstack-swift | 09:49 | |
| *** DaveBCN has quit IRC | 09:55 | |
| *** acoles_ is now known as acoles | 09:59 | |
| *** haomaiwang has quit IRC | 10:01 | |
| acoles | good morning | 10:01 |
| *** haomaiwang has joined #openstack-swift | 10:02 | |
| *** silor has joined #openstack-swift | 10:08 | |
| mahatic | good morning, acoles | 10:13 |
| -openstackstatus- NOTICE: review.openstack.org (aka gerrit) is going down for an emergency restart | 10:23 | |
| *** ChanServ changes topic to "review.openstack.org (aka gerrit) is going down for an emergency restart" | 10:23 | |
| *** baojg has quit IRC | 10:24 | |
| *** baojg has joined #openstack-swift | 10:25 | |
| *** baojg has quit IRC | 10:48 | |
| *** ChanServ changes topic to "Review Dashboard: https://goo.gl/eqeGwE | Summary Dashboard: https://goo.gl/jL0byl | Hackathon topics: https://etherpad.openstack.org/p/swift-midcycle-aug-2015 | https://etherpad.openstack.org/p/swift_encryption_issues | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/" | 10:49 | |
| -openstackstatus- NOTICE: Gerrit restart has resolved the issue and systems are back up and functioning | 10:49 | |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:02 | |
| *** leopoldj has joined #openstack-swift | 11:04 | |
| *** marcusvrn has joined #openstack-swift | 11:05 | |
| *** cdelatte has joined #openstack-swift | 11:15 | |
| *** delattec has joined #openstack-swift | 11:15 | |
| *** sakaYK has joined #openstack-swift | 11:23 | |
| *** sakaYK has quit IRC | 11:23 | |
| *** delattec has quit IRC | 11:26 | |
| *** cdelatte has quit IRC | 11:26 | |
| *** sanchitmalhotra has quit IRC | 11:36 | |
| *** haomaiwang has quit IRC | 11:36 | |
| *** leopoldj has quit IRC | 11:44 | |
| *** janonymous has quit IRC | 11:53 | |
| *** chlong has quit IRC | 11:59 | |
| *** robefran has quit IRC | 12:19 | |
| *** hrou has joined #openstack-swift | 12:33 | |
| onovy | PLS https://review.openstack.org/#/c/192562/ https://review.openstack.org/#/c/192198/ someone for review? waiting ~2mons | 12:34 |
| *** ekarlso has quit IRC | 12:44 | |
| *** ekarlso has joined #openstack-swift | 12:44 | |
| *** robefran has joined #openstack-swift | 12:49 | |
| *** jrichli has joined #openstack-swift | 12:50 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Fix disparsion-reports error message https://review.openstack.org/213697 | 12:52 |
| *** robefran has quit IRC | 13:03 | |
| *** lifeless has quit IRC | 13:05 | |
| tdasilva | good morning | 13:11 |
| *** haomaiwa_ has joined #openstack-swift | 13:13 | |
| *** dustins has joined #openstack-swift | 13:13 | |
| pgbridge | good morning | 13:16 |
| *** dustins has quit IRC | 13:21 | |
| *** dustins has joined #openstack-swift | 13:26 | |
| *** lifeless has joined #openstack-swift | 13:28 | |
| *** cdelatte has joined #openstack-swift | 13:29 | |
| *** delattec has joined #openstack-swift | 13:29 | |
| *** NM has joined #openstack-swift | 13:30 | |
| *** lpabon has joined #openstack-swift | 13:39 | |
| *** changbl has joined #openstack-swift | 13:43 | |
| *** ccavanna has joined #openstack-swift | 13:45 | |
| *** SkyRocknRoll_ has quit IRC | 13:46 | |
| *** peluse1 has joined #openstack-swift | 13:50 | |
| *** petertr7_away is now known as petertr7 | 13:56 | |
| *** zul has joined #openstack-swift | 13:58 | |
| *** sakaYK has joined #openstack-swift | 13:59 | |
| *** breitz has joined #openstack-swift | 14:00 | |
| *** minwoob has joined #openstack-swift | 14:00 | |
| *** haomaiwa_ has quit IRC | 14:01 | |
| *** tongli has joined #openstack-swift | 14:01 | |
| *** haomaiwang has joined #openstack-swift | 14:02 | |
| *** lpabon has quit IRC | 14:03 | |
| *** chlong has joined #openstack-swift | 14:03 | |
| *** dustins has quit IRC | 14:06 | |
| *** bill_az has joined #openstack-swift | 14:06 | |
| *** dustins has joined #openstack-swift | 14:08 | |
| *** jkugel has joined #openstack-swift | 14:12 | |
| *** lpabon has joined #openstack-swift | 14:15 | |
| *** janonymous has joined #openstack-swift | 14:19 | |
| *** lpabon has quit IRC | 14:28 | |
| *** lpabon has joined #openstack-swift | 14:30 | |
| *** baojg has joined #openstack-swift | 14:43 | |
| *** NM has quit IRC | 14:43 | |
| jrichli | acoles: I guess you made it back in one piece :-) | 14:43 |
| jrichli | acoles: for the 'encrypt-me' flag for probe tests, what do you think of setting the crypto override in the environment upfront instead of using a new flag? | 14:45 |
| *** jlhinson has joined #openstack-swift | 14:48 | |
| *** annegentle has joined #openstack-swift | 14:48 | |
| acoles | jrichli: yep not too bad a journey back | 14:48 |
| acoles | jrichli: the probe tests would (i imagine) send requests to the server so would not have access to the request environ to set a flag there. But an encrypt-me=False header could be mapped to crypto.override, by the keymaster i guess | 14:50 |
| jrichli | acoles: ok. I didn't know what types of things the probetests could do - since they dont seem as restricted to testing the api as functests | 14:51 |
| acoles | jrichli: in general they still call the API but can also go dig around the backend servers, run/kill processes etc | 14:52 |
| *** lpabon has quit IRC | 14:53 | |
| jrichli | acoles: ok. could we make the flag a crypto override, so its in the request only to say - don't encrypt? this would map more directly to the current override concept. | 14:54 |
| *** changbl has quit IRC | 14:55 | |
| *** wbhuber has joined #openstack-swift | 14:55 | |
| acoles | jrichli: didn't we discuss having a probe test use BYOK (so enable rather than disable crypto) so that the test could then verify on disk cipher text? | 14:55 |
| *** dustins has quit IRC | 14:57 | |
| jrichli | acoles: yes, but I thought we also talked about being able to control the key used from a simulated key server. we can do the BYOK for now, tho | 14:57 |
| acoles | jrichli: oic, so BYOK to force they key, and an override flag to force no generated key. | 14:59 |
| acoles | s/they/the/ | 14:59 |
| jrichli | acoles: yes, I think that is what I was thinking - good summary! | 14:59 |
| *** changbl has joined #openstack-swift | 15:00 | |
| *** lpabon has joined #openstack-swift | 15:00 | |
| *** haomaiwang has quit IRC | 15:01 | |
| *** peluse1 has quit IRC | 15:01 | |
| *** dustins has joined #openstack-swift | 15:02 | |
| *** haomaiwang has joined #openstack-swift | 15:02 | |
| *** morgan_503 is now known as morgan_2549 | 15:06 | |
| *** esker has joined #openstack-swift | 15:08 | |
| *** baojg has quit IRC | 15:09 | |
| *** dustins has quit IRC | 15:10 | |
| *** haomaiwang has quit IRC | 15:12 | |
| *** ajiang_ has joined #openstack-swift | 15:13 | |
| jrichli | acoles mahatic: I just summarized this in the trello card https://trello.com/c/8Y3yOJWZ/49-add-a-header-encrypt-flag-for-trivial-keymaster-to-use-in-probetests | 15:14 |
| notmyname | good morning | 15:16 |
| *** zul has quit IRC | 15:18 | |
| *** chlong has quit IRC | 15:18 | |
| *** tsubic has quit IRC | 15:20 | |
| janonymous | Morning | 15:22 |
| hurricanerix | Thanks again jrichli and IBM for hosting the hackathon. I had a great time (especially the panic house thing). | 15:23 |
| jrichli | hurricanerix: you and all swifters are certainly welcome! I am glad you enjoyed it. | 15:24 |
| hrou | jrichli, acoles, glad you both made it back ! Nice summary jrichli, the one other point I recall is when you supply your own key we're going to want the probe test to find out the IV (but I recall acoles mentioned this would be possible), this is if you want to know the cipher text | 15:24 |
| *** dustins has joined #openstack-swift | 15:26 | |
| acoles | hrou: yes. so the test would PUT an obj with BYOK, then go find the obj on disk, read metadata to get IV (which is in plaintext sysmeta) then calculate expected on disk ciphertext (given BYOK value and IV) and compare. | 15:26 |
| acoles | s/IV/nonce/ :) | 15:26 |
| jrichli | yes! thanks acoles and hrou | 15:27 |
| hrou | acoles, I do like nonce better ; ) Yep perfect ! | 15:27 |
| hrou | jrichli, did you want me to update the trello ? Though I realized I don't have permission to do that : ) | 15:27 |
| jrichli | hrou - thanks for the offer. since you dont have perms yet, i can do that. notmyname: could you add hrou (Hamdi Roumani) to the encryption trello? | 15:29 |
| notmyname | hrou: what's your trello id? | 15:29 |
| mattoliverau | morning | 15:30 |
| hrou | notmyname, thanks ! hamdiroumani | 15:30 |
| notmyname | done | 15:31 |
| notmyname | also added mahatic | 15:31 |
| jrichli | mattoliverau! you aren't usually on now. | 15:31 |
| notmyname | he stayed in texas! | 15:31 |
| *** chlong has joined #openstack-swift | 15:31 | |
| hrou | notmyname, thanks, and thanks jrichli for the update. | 15:31 |
| jrichli | notmyname: thanks! i will have one more for you to add soon - but i dont think jon has an account yet | 15:32 |
| hrou | mattoliverau, how was the drive down ? (or up I guess : - ). | 15:32 |
| mahatic | hello! | 15:32 |
| mahatic | jrichli: cool cookies! :) | 15:32 |
| jrichli | mahatic: hello! | 15:32 |
| notmyname | jrichli: yeah, when is he getting his paperwork to contribute done? | 15:32 |
| jrichli | mahatic: thanks :-) | 15:32 |
| mahatic | nice pictures on twitter notmyname | 15:32 |
| * mahatic reading scrollback | 15:33 | |
| jrichli | notmyname: he got is official stamp during this week - now just has to make all the accounts | 15:33 |
| notmyname | great | 15:33 |
| jrichli | notmyname mattoliverau: oh, right ... with rackers. Having fun? | 15:33 |
| mattoliverau | jrichli: yeah still in Texas, in SAT. | 15:33 |
| *** chsc has joined #openstack-swift | 15:34 | |
| mattoliverau | hrou: it was fine, much shorter then the inital flights the travel agent booked. | 15:34 |
| *** dustins has quit IRC | 15:35 | |
| *** ccavanna has quit IRC | 15:35 | |
| notmyname | emails and IRC buffers read, kids dropped off for first day of school. time to commute | 15:37 |
| mattoliverau | jrichli: yup, hurricanerix took mrda and I out shooting like real Texans yesterday :) | 15:37 |
| * notmyname back online later | 15:38 | |
| hrou | mattoliverau, ha! yea can't believe they were suggesting that at some point : ) | 15:38 |
| *** chlong has quit IRC | 15:38 | |
| jrichli | mattoliverau: was that out in the wild, or at a facility? | 15:39 |
| mattoliverau | jrichli: shooting range, middle of the day, really hot.. but we had fun :) | 15:40 |
| jrichli | nice! Now you just need boots and a hat | 15:40 |
| janonymous | Sounds cool :) | 15:40 |
| *** chlong has joined #openstack-swift | 15:40 | |
| mattoliverau | jrichli: lol, yup, then I'd be a real Texan ;) | 15:40 |
| janonymous | Texas Poker :D | 15:41 |
| mattoliverau | Well I guess I'd better write an update to the sharding spec reflecting hackathon while it''s still fresh. | 15:42 |
| mattoliverau | janonymous: lol, hat first :P | 15:42 |
| mrda | ...and fwiw, mattoliverau is a gun shark. He's a pretty good shot :) | 15:42 |
| jrichli | [10:26] <notmyname> overall, I think it went well. it got off to a rough start, IMO, as we were figuring out the "schedule" and how to work together. but it ended (last 2 days) really well | 15:42 |
| jrichli | [10:27] <notmyname> adn you and IBM did great as a host. it was great | 15:42 |
| jrichli | janonymous: you mean Texas Hold'em? | 15:42 |
| jrichli | oops | 15:42 |
| janonymous | Yup :P | 15:42 |
| jrichli | paste didnt work out for me so well today | 15:42 |
| mattoliverau | mrda: lol, beginners luck. | 15:44 |
| janonymous | Sounds crazy holidays guys.. seems like hangover part 4 ;) | 15:44 |
| *** dustins has joined #openstack-swift | 15:50 | |
| *** robefran has joined #openstack-swift | 15:51 | |
| *** nadeem has joined #openstack-swift | 15:51 | |
| acoles | mattoliverau: heh, so topgolf and topgun in same week | 15:53 |
| mattoliverau | acoles: lol, yup. And we drove past a topgolf here in SAT.. maybe I should have joined as a life member ;) | 15:56 |
| *** lcurtis has joined #openstack-swift | 15:56 | |
| *** gyee has joined #openstack-swift | 15:56 | |
| mattoliverau | acoles: how was the flight home? | 15:56 |
| *** SkyRocknRoll has quit IRC | 15:56 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:57 | |
| *** SkyRocknRoll has quit IRC | 15:57 | |
| acoles | mattoliverau: not too bad thanks. got lucky with an empty seat next to me. | 15:57 |
| acoles | mattoliverau: turns out there is topgolf in uk. but nowhere near me. | 15:58 |
| *** SkyRocknRoll has joined #openstack-swift | 15:58 | |
| mattoliverau | acoles: roadtrip!! | 16:00 |
| acoles | lol | 16:01 |
| jrichli | acoles: globalization. you travel across the ocean to see the same stuff :-) | 16:01 |
| acoles | jrichli: mattoliverau i bet the uk version has wooden clubs and you have to go pick up your own balls | 16:02 |
| jrichli | acoles: lol | 16:03 |
| wbhuber | acoles: also the uk version has higher wind speed, thanks to st.andrews! | 16:03 |
| *** dosaboy_ has joined #openstack-swift | 16:04 | |
| *** dosaboy_ has quit IRC | 16:04 | |
| acoles | wbhuber: that too ;) | 16:04 |
| *** dosaboy has quit IRC | 16:04 | |
| mattoliverau | lol | 16:04 |
| pgbridge | hard mode | 16:07 |
| lcurtis | hello all...so trying to verify something... so I was under general impression that for a failed drive being swapped with a day that one would not have to set weight to zero--just simply replace drive | 16:07 |
| lcurtis | though from Joe Arnold's book, recommending to remove the failed drive from the ring | 16:08 |
| *** rledisez has quit IRC | 16:08 | |
| lcurtis | above should read a drive being swapped within a day | 16:08 |
| *** dosaboy has joined #openstack-swift | 16:09 | |
| lcurtis | should i indeed be setting weight to zero for a failed drive | 16:09 |
| lcurtis | ? | 16:10 |
| lcurtis | then replacing? | 16:10 |
| *** pbourke has quit IRC | 16:10 | |
| *** pbourke has joined #openstack-swift | 16:11 | |
| ctennis | you only need to set it to weight 0 if it will be an extended period of time before you replace it | 16:11 |
| ctennis | if it's going to be within the day, I woudn't worry about it | 16:11 |
| hrou | jrichli, mattoliverau I wonder if you could combine top golf and a shooting range, that way if someone is getting ahead if you score wise, you just start shooting their golf balls in flight | 16:12 |
| *** jordanP has quit IRC | 16:12 | |
| jrichli | hrou: lol | 16:13 |
| *** david-ly_ is now known as david-lyle | 16:15 | |
| *** jistr has quit IRC | 16:19 | |
| lcurtis | ctennis: thanks | 16:19 |
| ctennis | lcurtis: the only downside to leaving the disk in the ring is that operations that want to read/write to it will have some added latency as they try, fail, then move to the next disk. | 16:19 |
| lcurtis | okay...awesome | 16:21 |
| lcurtis | makes sense | 16:21 |
| *** peluse1 has joined #openstack-swift | 16:24 | |
| peluse1 | acoles, yo, how was the trip home? | 16:24 |
| acoles | peluse1: ok thx. bit groggy today but not too bad. | 16:25 |
| peluse1 | that's cool, takes me forever to recover from that kind of time change anymore... | 16:26 |
| acoles | peluse1: got your review on my todo but had cam back to some firefighting | 16:26 |
| peluse1 | hey, I need to lobby hard for reviews on patch 212187 patch 213147 this week... next week a few SS guys (maybe clayg included) are coming to Chandler to help debug remaining issues on running EC under heavy load | 16:26 |
| patchbot | peluse1: https://review.openstack.org/#/c/212187/ | 16:26 |
| peluse1 | two of them now :) | 16:26 |
| peluse1 | patch 21347 | 16:27 |
| patchbot | peluse1: https://review.openstack.org/#/c/21347/ | 16:27 |
| peluse1 | hmm, patchbot didn't like listing two patches in one thingy.... | 16:27 |
| acoles | peluse1: k. your one down from top of my list! | 16:27 |
| peluse1 | acoles, sweet. let me know if there's something I can look at for you in the meantime - be happy to | 16:27 |
| acoles | nah patchbot is like, 'one thing at a time man' | 16:28 |
| peluse1 | acoles, also, what did you and clayg end up wrt plan for opt. GET? you on diskfile and me on proxy? | 16:28 |
| acoles | peluse1: oh were we supposed to make a plan ? :P | 16:29 |
| acoles | peluse1: that split would work. at least i would start with diskfile changes then work out from there. | 16:29 |
| peluse1 | heh, wasn't sure :) | 16:29 |
| wbhuber | peluse1: what kind of remaining issues running EC under heavy load that you saw? | 16:29 |
| acoles | so if you get round to proxy changes then let me know when you make a start | 16:30 |
| peluse1 | acoles, OK I will start on proxy probably today and just use the 2 headers we talked about as our i/f and dummy something up in my patch for testing... | 16:30 |
| wbhuber | are those issues the ones we discussed last week at hackathon (503 overwrite, etc.)? | 16:30 |
| acoles | peluse1: cool | 16:30 |
| peluse1 | wbhuber, unclear because the cluster wasn't under my control. I saw some exceptions but by the time I got to the cluster the benchmark team had already done like 6 things to it and I couldn't seem to get them trained to operate in debug mode... | 16:30 |
| peluse1 | so I'm taking over the cluster next week :) | 16:31 |
| wbhuber | ah, need a real person controller to see what went wrong. | 16:31 |
| wbhuber | do keep minwoo and i in the loop with your findings. | 16:31 |
| peluse1 | wbhuber, the last exception I was able to debug turned out to be a bogus log statement that clayg addressed and might already be merged | 16:31 |
| wbhuber | peluse1: were you and bill able to saturate the cluster over the network while running workloads on EC? | 16:32 |
| *** lyrrad has joined #openstack-swift | 16:33 | |
| peluse1 | wbhuber, no. we couold no problem with 3x so we are confident in the HW and SW setup (finally) however they kept running into stupid little things that blocked us from really going ful throttle and since they are not staffed for doing SW debug they pretty much punted | 16:33 |
| wbhuber | peluse1: at least they know how to punt. minwoo and i are doing both SW debug and HW cluster setup. | 16:35 |
| peluse1 | yeah, its been frustrating from both sides but at least they're letting me 'borrow' the full rack to get things in order before turning it back over to them for perf testing | 16:36 |
| *** obiwandev has joined #openstack-swift | 16:38 | |
| peluse1 | wbhuber, if you have a chance and can take a look at those 2 patches I just mentioned that'd be great too | 16:48 |
| *** lpabon has quit IRC | 16:50 | |
| *** mfalatic has joined #openstack-swift | 16:51 | |
| *** haypo has joined #openstack-swift | 16:51 | |
| *** mfalatic_ has joined #openstack-swift | 16:52 | |
| haypo | hello! is someone available to review my patch https://review.openstack.org/#/c/199034/ "Fix tox -e py34" -- it provides a basic python3 support to start working on porting swift to python3 | 16:52 |
| haypo | it allows to make the py34 gate voting to ensure that we don't go backward in python3 support | 16:52 |
| haypo | PyEClib 1.0.8 was tagged in Git two weeks ago, but there is no release on PyPI!? https://bitbucket.org/kmgreen2/pyeclib/issues/69/need-a-release-108-with-python-3-fixes | 16:53 |
| haypo | (PyEClib 1.0.8 includes my python3 changes, so my sed hack in tox.ini is no more needed) | 16:54 |
| *** NM has joined #openstack-swift | 16:54 | |
| notmyname | acoles: peluse1: patches welcome for patchbot ;-) | 16:55 |
| *** mfalatic has quit IRC | 16:56 | |
| haypo | notmyname: hi. i'm back from 2 weeks of holiday. i don't remind if you like my change or not :) | 16:56 |
| peluse1 | haypo, I'l ping tsg wrt the pyeclib release | 16:56 |
| haypo | i have to read again the latest comments | 16:56 |
| notmyname | haypo: I don't either ;-) | 16:56 |
| haypo | notmyname: haha, no problem | 16:57 |
| haypo | peluse1: oh, cool | 16:57 |
| haypo | peluse1: i don't understand why a version was tagged without tarball released on PyPI | 16:57 |
| *** zaitcev has joined #openstack-swift | 16:57 | |
| *** ChanServ sets mode: +v zaitcev | 16:57 | |
| haypo | zaitcev: ah! you already voted +2 on the patch set 12 of https://review.openstack.org/#/c/199034/ you should revote ;) (i had to rebase my patch) | 17:04 |
| peluse1 | haypo, texting with tsg - both of us are laptop-less today :( He said they can't push latest to pypi because of a jerasure change | 17:04 |
| peluse1 | eta getting close though | 17:05 |
| haypo | peluse1: ah? which change? | 17:05 |
| haypo | peluse1: a bug? | 17:05 |
| zaitcev | I wish someone made Kevin to cut releases and not just "push latest to pypi" and consider it done. | 17:05 |
| peluse1 | haypo, I do't beleive so. I don't want to mis-speak though. The last push should be noon tpday | 17:06 |
| notmyname | yeah, we're working on improving the situation around pyeclib/liberasurecode, with regard to releases | 17:06 |
| notmyname | it was something I brought up with quite a few people last week at the hackathon. I should have a resolution for what to do Real Soon Now | 17:07 |
| wbhuber | notmyname: such as incorporating pyeclib into swift? since it is not openstack material i don't think we reached a consensus on what to do with the package yet... right? | 17:10 |
| haypo | notmyname: pyeclib 1.0.8 installs liberasurecode 1.0.8, pyeclib has an embedded copy of liberasurecode-1.0.8.tar.gz | 17:11 |
| haypo | for me, it's the best option, but i heard that it's compromise | 17:11 |
| zaitcev | wbhuber: pyeclib did not shed all of C, it still builds an adapter for Python. | 17:11 |
| notmyname | wbhuber: no, it won't be in the swift source tree. it's more of getting more people with commit rights and where the official repo lives. with more people, especially those who are using it, we'll have better releases. the releases are the only real problem I'm aware of right now (although some of those issues are related to some internal coupling) | 17:14 |
| *** obiwandev has quit IRC | 17:14 | |
| mattoliverau | hrou: lol (sorry for the late response, I get pulling into meeting people in the office) :) | 17:18 |
| hrou | mattoliverau, jrichli oh no worries at all ! Yea seems like a good franchise opportunity, best of both worlds ; - ) | 17:19 |
| peluse1 | notmyname, so my laptop is dead today so I'm on a different machine and had to start from scratch and haven't done so since that hacking update not too long ago. I think it ight be breaking me as I can run unitests w/o issue but once I isntall tox things go to hell and I get like 28 tests fail. I can't find that patch though that you pointed me to before where clayg commented "do this if things pass locally and fail in the gate"? | 17:23 |
| hrou | peluse1, 205977 ? | 17:25 |
| hrou | patch 205977 | 17:26 |
| patchbot | hrou: https://review.openstack.org/#/c/205977/ | 17:26 |
| peluse1 | yes, thanks :) | 17:26 |
| hrou | And you'll probably need tox -r as well btw | 17:26 |
| peluse1 | trying it now... | 17:27 |
| *** geaaru has quit IRC | 17:37 | |
| notmyname | oh good :-) | 17:38 |
| *** SkyRocknRoll has quit IRC | 17:39 | |
| peluse1 | hrou, so that didn't fix me but I resoted a snapshot from before I tried to install tox using pip and this time installed using apt-get and it works fine. older version I supose - I dunno | 17:39 |
| hrou | peluse1, hmm odd, but does the older version predate that patch above that changed the test requirements ? Let me know if you run into it again though! | 17:41 |
| peluse1 | hrou, beats me :) I did upgrade the test-reqs before trying with both the pip tox and the apt-get tox (1.6.0 BTW) and only the apt-get works for me at least | 17:42 |
| hrou | peluse1, : ) Interesting ! Let me know if you have issues again, the magic for me was -r which rebuilds the tox env (and as such the package versions) | 17:43 |
| hrou | peluse1, glad its working for you now though! | 17:43 |
| peluse1 | hrou, will do. since I have a working setup saved as snapshot I'll try upgrading tox and see if it works - if not I'll give up for now and just use the old one | 17:43 |
| hrou | peluse1, it'd be funny to see how many snapshots folks have while trying to setup something, I know when I was trying to get storlets, keystone, swift all working (which is an effort) I had over 20 snapshots at one point ; ) | 17:44 |
| peluse1 | hrou, nice! yeah what did we do before VMs?? | 17:48 |
| *** lpabon has joined #openstack-swift | 17:50 | |
| *** dbarman has joined #openstack-swift | 17:50 | |
| dbarman | Can anyone tell me how the replication works? How often does it send replication packets? my swift-recon -r shows the last replication was 57 minutes ago | 17:51 |
| dbarman | is there a way to speed up the replication process? | 17:51 |
| dbarman | Also, is there was to tell when the replication is 100% complete? | 17:52 |
| hrou | peluse1, cry a lot, or break things, depending on your personality, Kids these days just don't appreciate how easy it is : - ) | 17:52 |
| peluse1 | lol! | 17:52 |
| mahatic | jrichli: about the patch, you mean, remove from "encrypter.py"? | 17:54 |
| *** lpabon has quit IRC | 17:55 | |
| mahatic | jrichli: I thought you were talking about the test file! got it | 17:56 |
| jrichli | mahatic: yes. and when you modify decrypter tests to stop using that method, then you can remove that method from the decrypter | 17:56 |
| *** zul has joined #openstack-swift | 17:56 | |
| mahatic | jrichli: right, thanks | 17:56 |
| *** acoles is now known as acoles_ | 17:58 | |
| dbarman | Can anyone tell me how the replication works? How often does it send replication packets? my swift-recon -r shows the last replication was 57 minutes ago | 17:59 |
| *** dustins has quit IRC | 17:59 | |
| notmyname | dbarman: you can find a lot of technical explanations of how different parts of swift work at http://docs.openstack.org/developer/swift/ and https://swiftstack.com/openstack-swift/ | 18:00 |
| *** dustins has joined #openstack-swift | 18:01 | |
| wbhuber | peluse: acoles: re: increasing durable quorum to parity + 1 to tell the world that an objec tis rebuildable with enough fragments, do we do it in the 2nd phase when we send a commit confirmation to participating object servers? i'm looking inside _get_put_responses with need_quorum enabled. | 18:01 |
| *** zhill has joined #openstack-swift | 18:02 | |
| *** lpabon has joined #openstack-swift | 18:02 | |
| haypo | peluse1: maybe it's better to wait for the PyEClib release on PyPI | 18:04 |
| haypo | peluse1: instead of pushing my sed hack | 18:04 |
| haypo | peluse1: since we are now really close to an effictive release ;) | 18:04 |
| peluse1 | haypo, yeah probably | 18:05 |
| peluse1 | wbhuber, yes | 18:06 |
| peluse1 | wbhuber, today its the sending of the commit messge that only requires 2 | 18:06 |
| haypo | for my "pep8" patch serie, i'm fine with keeping checks skipped | 18:07 |
| haypo | we just have to agree on which tests are revelant or not | 18:07 |
| haypo | checks* | 18:07 |
| *** zhill_ has joined #openstack-swift | 18:07 | |
| peluse1 | wbhuber, look in _store_object() where min_conns is set to 2 | 18:07 |
| *** zhill has quit IRC | 18:07 | |
| wbhuber | yes | 18:07 |
| haypo | i may open a thread on openstack-dev to discuss pep8 in swift | 18:08 |
| dbarman | dustins: is it common for replication to wait over an hour? | 18:20 |
| dbarman | there other issue I am not sure about is. I have two storage nodes. Once is almost 100% full, the other one is just 7% full. Doesn't it spread the data over all the nodes somewhat evenly? | 18:21 |
| dustins | dbarman: Hmm? Sorry I don't quite know the context | 18:21 |
| dbarman | dustins: Are you referring to my last inquiry or the question about the over 1 hour for replication? | 18:22 |
| dustins | the one you mentioned me on :) | 18:22 |
| *** kutija has joined #openstack-swift | 18:23 | |
| ctennis | in general yes it will dbarman | 18:23 |
| dbarman | dustins: ok. The context is that I recently added a second node. So swift started to replicate to the new node. However, when I checked the replication status this afternoon, it indicated that the last replication was about 1 hour ago. I am wondering why it is waiting to continue the replication or how I can tell if the replication is 100% complete | 18:24 |
| dbarman | ctennis: nice to hear from you again. | 18:24 |
| dbarman | ctennis: So does it just take a long time to attempt to move the data to the new node or do you think someting is wrong? | 18:25 |
| ctennis | dbarman: you need to post the layout of your ring files so we can verify you have them configured right. Also please check for any additional errors. | 18:25 |
| dustins | dbarman: I'll be honest, I don't really know the first thing about Swift (yet), I was just curious as to why you pinged me :P | 18:25 |
| dustins | that was the context I wanted, haha | 18:25 |
| ctennis | dbarman: I think something is wrong. It would help if we could see log files as well. | 18:25 |
| *** lpabon has quit IRC | 18:25 | |
| dbarman | ctennis: ok. what is the best way to show my ring file layout? | 18:26 |
| dbarman | ctennis: and the log files | 18:26 |
| notmyname | haypo: what are you wanting to discuss on the mailing list about pep8 in swift that isn't discussed in here? | 18:26 |
| ctennis | swift-ring-builder object.builder and post the output somewhere list gist.github.com you can share | 18:26 |
| ctennis | for log files, find a place you can upload and share and give a link | 18:26 |
| dbarman | ctennis: ok. you mean the syslog file, right? | 18:28 |
| *** lpabon has joined #openstack-swift | 18:28 | |
| dbarman | ctennis: That is the only log I am aware of | 18:28 |
| haypo | notmyname: the question is to decide which checks should stay skipped | 18:28 |
| ctennis | dbarman whatever you can find to share | 18:28 |
| haypo | notmyname: i started to send patches to reduce the number of skipped checks | 18:29 |
| dbarman | ctennis: lol. ok | 18:29 |
| haypo | notmyname: i'm not really motivated to fix all pep8 checks, it's just a follow up of my patch to upgrade hacking from 0.8 to 0.10 | 18:30 |
| notmyname | haypo: for the time being, if there are style changes that are required for py3 (ie py27 formatting is simply not allowed/not supported in py3), then fix it. if it's something else, then don't | 18:30 |
| notmyname | the list of stuff we currently don't check on is not a todo list of style fixes | 18:30 |
| haypo | notmyname: there is a single check related to python3, and janonymous already wrote a patch for it | 18:30 |
| haypo | notmyname: samuel likes 3 of my patches | 18:31 |
| haypo | https://review.openstack.org/#/c/213491/ pep8: replace deprecated calls to assert_() | 18:31 |
| notmyname | great | 18:31 |
| haypo | https://review.openstack.org/#/c/213494/ pep8: Fix usage of the l10n _('...') function | 18:31 |
| haypo | https://review.openstack.org/#/c/213495/ pep8: Don't override '_' symbol | 18:31 |
| haypo | notmyname: my current patches, https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:master+topic:pep8,n,z | 18:31 |
| *** annegentle has quit IRC | 18:32 | |
| notmyname | style changes are one of the things that are a "doesn't matter, default to yes" sort of change. it's not something that is prioritized, and it shouldn't have a ton of time spent on it. there's a *lot* of other things going on that need the mental effort | 18:33 |
| notmyname | gotta run for a bot | 18:34 |
| haypo | notmyname: tox -e pep8 has different kind of checks. some of them finds real bugs | 18:35 |
| haypo | notmyname: like H702 "Formatting operation should be outside of localization method call" - https://review.openstack.org/#/c/213494/ | 18:36 |
| *** silor has quit IRC | 18:38 | |
| dbarman | ctennis: will have logs posted shortly | 18:42 |
| openstackgerrit | Bill Huber proposed openstack/swift: Quorum on durable response is too low https://review.openstack.org/213822 | 19:01 |
| *** mahatic has quit IRC | 19:01 | |
| *** sakaYK has quit IRC | 19:02 | |
| wbhuber | peluse: for the patch 213822, let me know if you'd like more testing on putters that have adequate successes (i.e. statuses > min_of_responses). | 19:02 |
| patchbot | wbhuber: https://review.openstack.org/#/c/213822/ | 19:02 |
| *** annegentle has joined #openstack-swift | 19:05 | |
| openstackgerrit | Janie Richling proposed openstack/swift: Encrypter and Decrypter middleware - object body only https://review.openstack.org/194191 | 19:06 |
| *** lpabon has quit IRC | 19:13 | |
| *** lpabon has joined #openstack-swift | 19:19 | |
| peluse1 | wbhuber, looks like clayg just commented on exactly that! | 19:22 |
| wbhuber | peluse1: yes, lol. GMTA. | 19:22 |
| dbarman | ctennis: can't upload files to github. is there somewhere else I can upload the logs files to? | 19:23 |
| ctennis | how big are they? | 19:23 |
| *** lpabon has quit IRC | 19:23 | |
| *** petertr7 is now known as petertr7_away | 19:24 | |
| dbarman | ctennis: one's about 100MB the other about 30MB | 19:27 |
| ctennis | I dunno, maybe something like dropbox? | 19:28 |
| ctennis | or google drive | 19:28 |
| ctennis | or amazon s3 | 19:28 |
| dbarman | ctennis: oh boy.. ok | 19:28 |
| *** petertr7_away is now known as petertr7 | 19:29 | |
| *** silor has joined #openstack-swift | 19:31 | |
| dbarman | ctennis: try this: https://drive.google.com/folderview?id=0B1e8GRJ1mpZTfmZHY1BhRmNmbmY2cC1jb2xxeTNwNHEzSU9mUXpXU0ZBZHZTbm5BY1VfbUU&usp=sharing | 19:33 |
| dbarman | ctennis: let me know if you can view the files or not | 19:36 |
| openstackgerrit | Merged openstack/swift: Fix some minor typos https://review.openstack.org/213378 | 19:36 |
| ctennis | dbarman: all of the quarantine information is a bad sign | 19:37 |
| ctennis | dbarman: on the drives that are full, how much data is in the objects/ directory vs. how much is in the quarantine directory? | 19:37 |
| dbarman | ctennis: where is the quarantine directory? | 19:38 |
| ctennis | on the disk | 19:38 |
| ctennis | your object replicator is far from being done running | 19:40 |
| ctennis | Aug 17 14:19:57 OSSN1 object-replicator: 167/2047 (8.16%) partitions replicated in 93002.61s (0.00/sec, 290h remaining) | 19:40 |
| ctennis | Aug 17 14:24:57 OSSN1 object-replicator: 167/2047 (8.16%) partitions replicated in 93302.62s (0.00/sec, 291h remaining) | 19:40 |
| ctennis | Aug 17 14:29:57 OSSN1 object-replicator: 167/2047 (8.16%) partitions replicated in 93602.62s (0.00/sec, 292h remaining) | 19:40 |
| *** annegentle has quit IRC | 19:42 | |
| *** annegentle has joined #openstack-swift | 19:43 | |
| dbarman | ctennis: that is what I thought. So that is why I was wondering that the swift-recon -r showed the last replication as over an hour ago. what do you think? | 19:45 |
| ctennis | not sure | 19:45 |
| ctennis | was your cluster just 2 disks before? | 19:45 |
| dbarman | ctennis: I just did it again. now it says it has been 2 hours since most recent completion | 19:46 |
| dbarman | yes, it was two disks on the first node and now the second node has 2 more disks | 19:46 |
| ctennis | perhaps that's just on the new node where the data is being replicated to and not from | 19:47 |
| ctennis | yes that's right, your new node is only taking about 2 hours to run through replication | 19:47 |
| dbarman | this is what it does: [2015-08-17 15:45:36] Checking on replication [replication_time] low: 80, high: 154, avg: 117.2, total: 234, Failed: 0.0%, no_result: 0, reported: 2 Oldest completion was 2015-08-14 17:13:19 (3 days ago) by 10.0.0.51:6000. Most recent completion was 2015-08-17 16:51:54 (2 hours ago) by 10.0.0.52:6000. =============================================================================== | 19:47 |
| ctennis | right, so your old node has been 3 days since it completed | 19:48 |
| ctennis | because it's still in the middle of trying to complete | 19:48 |
| dbarman | friday is when I put the second node in play | 19:48 |
| dbarman | ie. 3 days ago | 19:48 |
| ctennis | keep in mind, going from 2 disks to 4 disks isn't going to buy you much..swift wanted to keep 3 copies of data anyway and you only had two copies to begin with, so you didn't double your capacity | 19:49 |
| dbarman | I was afraid of that | 19:49 |
| dbarman | but I should get 50% increase??? | 19:49 |
| ctennis | yeah | 19:50 |
| minwoob | zaitcev: Can you expound on what you mean in https://review.openstack.org/#/c/198909/ ? | 19:50 |
| dbarman | it just seems strange that it's been 2 hours. in the earlier stages it was updating every 15 minutes or even faster | 19:51 |
| dbarman | I am wonder if it stopped for some reason. Did you see any error in the logs that might be an issue? | 19:52 |
| dbarman | What should I do about quarantine (not sure where it is located). Should they be deleted? Does swift remove them automatically? | 19:52 |
| ctennis | it didn't stop. All of the new data coming onto that node has to be scanned by the replicator which is adding to the time | 19:52 |
| ctennis | once things settle down it will go back to ~15 minutes or wahtever you saw before | 19:52 |
| ctennis | I don't know what you tell you about the quarantine, swift just moves the data into the quarantine directory on the same disk as it existed before and lets you decide what to do with it. The amount of quarnatine data you have indicates to me something is gravely wrong. | 19:53 |
| onovy | notmyname, hi, what about Tokyo summit? we didn't finish discussion :) | 19:54 |
| dbarman | Yeah, my hash pre/suffix values were incorrect on the second node. once I corrected it the quarantine logs stopped. pretty sure that was the source of the problem. | 19:54 |
| dbarman | Do you know where the quarantine is located, I think I should just delete them | 19:54 |
| ctennis | it's on the same disks as your swift data, in a directory called "quarantine" | 19:55 |
| onovy | mattoliverau, hi, did you had time for 192562 please? | 19:55 |
| dbarman | I meant before pre/suffix hash values were wrong on the second node | 19:55 |
| notmyname | onovy: oh yeah. you were asking if it's "worth it", right? | 19:55 |
| onovy | notmyname, yep | 19:55 |
| onovy | [09:07:05] <onovy> notmyname, get "hot news" and meet another developers / admins of openstack swift. | 19:55 |
| onovy | my last message :) | 19:55 |
| notmyname | onovy: right | 19:55 |
| notmyname | onovy: I *always* think in-person and face-to-face conversation is good :-) | 19:56 |
| dbarman | is it safe to delete the quarantine folder? | 19:56 |
| dbarman | or it's contents | 19:56 |
| notmyname | onovy: so in that sense, Tokyo (and every openstack summit) is "worth it" | 19:56 |
| ctennis | it's safe, as long as you're not concerned about losing all of the data associated with it | 19:56 |
| onovy | notmyname, ok. so it's possible to meet some of dev/core-devs here and it's time to talk | 19:57 |
| notmyname | onovy: but it's like other tech conferences. you'll get out of it what you look for. if you're looking for a lot of conversations with people and wanting to meet people and you make that happen, it will be great! | 19:57 |
| notmyname | onovy: and if you want to learn more and go listen to a lot of talks, you can do that too | 19:57 |
| dbarman | well, I assume the data is bogus since my hashing was wrong. What do you think? | 19:57 |
| notmyname | onovy: yes. most core devs and many people who are active in here will be there | 19:57 |
| ctennis | dbarman: the data doesn't change, just where it gets stored. The risk is you may be deleting data you cannot recover/retrieve. If you're okay with that, then delete. | 19:58 |
| onovy | perfect, i will try to come | 19:58 |
| *** mfalatic_ has quit IRC | 19:58 | |
| *** mfalatic has joined #openstack-swift | 19:58 | |
| zaitcev | minwoob: I do not like the "except Exception" construct, because I had an experience with them consuming unexpected exceptions. It is a condition that is difficult to debug. I allow for them to get caught in one place in a program. That's just me, however. | 19:59 |
| onovy | notmyname, so thanks. is there something what can i do for review of my patch please? waiting ~2mons and feature freeze is near. | 20:00 |
| dbarman | other than deleted what is typical for quarantine? can it be recovered? if so, in what method/way? | 20:02 |
| ctennis | would take some work. beyond my pay grade. | 20:02 |
| notmyname | onovy: I'm not sure where you are based, but know that Japan has some tougher visa requirements. please look into it today. http://lists.openstack.org/pipermail/openstack/2015-August/013749.html | 20:03 |
| dbarman | lol | 20:03 |
| onovy | eu/cz | 20:03 |
| dbarman | looks like the quarantine folder is large. I am trying to figure out it's size. would du -hs tell me anything? | 20:04 |
| dbarman | I am trying it but it is just sitting there..hmmm.. | 20:04 |
| ctennis | there's a lot of disk activity, give it a few minutes | 20:04 |
| dbarman | good point | 20:05 |
| *** silor1 has joined #openstack-swift | 20:05 | |
| dbarman | I will see if it ever responds | 20:05 |
| minwoob | zaitcev: It's better to have them in those places, so we know where it is caught, no? | 20:05 |
| zaitcev | minwoob: You're assuming that logging is reliable, but it's not. | 20:06 |
| *** silor has quit IRC | 20:07 | |
| *** silor1 is now known as silor | 20:07 | |
| onovy | notmyname, visa is not needed for eu/cz, so it's ok. | 20:09 |
| notmyname | great. I'm glad you checked | 20:09 |
| onovy | perfect and what about review? i'm really unmotivated about this long process | 20:10 |
| clayg | onovy: what patch? | 20:13 |
| notmyname | onovy: which patch. I'm not currently mapping IRC to gerrit in my head ;-) | 20:13 |
| onovy | https://review.openstack.org/#/c/192198/ | 20:14 |
| onovy | https://review.openstack.org/#/c/192562/ | 20:14 |
| onovy | this two | 20:14 |
| dbarman | ctennis: wow, du -hs command is still running, no response yet! | 20:15 |
| clayg | onovy: looks like you've had reviews and made changes based on feedback - it just feel off the radar - pinging people in channel (particuarlly the people who have already reviewed them) is exactly what you're supposed to do | 20:16 |
| clayg | onovy: I haven't looked at the recon patch - so I'll let mattoliverau torgomatic and notmyname review that (hit them up tomorrow if they don't) | 20:16 |
| onovy | thanks | 20:16 |
| clayg | I've already got the rsync one starred - it's a big change - so don't expect to hear anything from me until tomorrow - but you're welcome to ask me again if I have any first impressions | 20:17 |
| notmyname | and it looks like I owe you a review on the recon time sync patch | 20:17 |
| notmyname | onovy: at a meta level, though, I'm sorry those have been open for such a long time. it's something we're working on chaning, and it's something we specifically talked about last week at the swift hackathon. | 20:18 |
| notmyname | I hope we'll be getting better in the future (but I realize that's small comfort to you now about these patches) | 20:19 |
| clayg | notmyname: although in fairness I don't think we came up with anything I expect to make the situation drastically different other than... | 20:19 |
| clayg | a) ask for reviews (it's not rude) b) help with reviews (obviously we have too many) | 20:19 |
| notmyname | there were a couple of things, I think. yeah, those were some of them | 20:20 |
| notmyname | I've got it in my notes. haven't had time to collect it all yet :-) | 20:20 |
| onovy | exactly i'm thinking about employ in one person for part/full time for Swift coding in our company, but it's hard when review process takes so long | 20:20 |
| clayg | onovy: if you want to employ them full time you should have them do reviews full time | 20:20 |
| *** jlhinson has quit IRC | 20:20 | |
| onovy | thats good point | 20:21 |
| *** jlhinson has joined #openstack-swift | 20:21 | |
| onovy | i have already done few reviews and i'm not sure if it is usefull | 20:21 |
| clayg | onovy: -1's are the most useful | 20:21 |
| notmyname | it is! | 20:21 |
| onovy | already gave some -1 :) | 20:22 |
| clayg | +1's are only useful if they were "I checked this out and ran it, all the tests (including probe & func) pass on my machine, code is all readable, commit clearly explains the change and references launchpad/bugs/DocImpact, it has new unittests, unittests fail if you remove the change, and here's some additional notes for reviews to help them get started testing this change functionally" | 20:23 |
| *** dbarman1 has joined #openstack-swift | 20:26 | |
| dbarman1 | ctennis: you still around? | 20:26 |
| ctennis | yes | 20:26 |
| onovy | ok | 20:26 |
| dbarman1 | still no go on the du command. Seem reasonable? | 20:27 |
| ctennis | not sure | 20:27 |
| dbarman1 | lol | 20:27 |
| dbarman1 | ok. | 20:27 |
| clayg | onovy: do you have the review dashboard -> https://wiki.openstack.org/wiki/Swift/PriorityReviews | 20:27 |
| onovy | yep, it's in topic here | 20:27 |
| clayg | going after the old changes is a good way to help keep the fodder down - or if you want to get more involved (social capital) you can grab some of the big high priority reviews and help out | 20:28 |
| clayg | onovy: great! you could also go dig up a change that's pending from one of the people that have done reviews for you and stick a -1 on that | 20:28 |
| clayg | ... you know - unless by some off chance it's acctually ready to merge - then feel free to +1, or ping a core maintainer and let them know you think you found a change that's good to merge! | 20:29 |
| clayg | onovy: good luck! | 20:30 |
| onovy | thanks a lot :) | 20:30 |
| clayg | c | 20:44 |
| *** silor has quit IRC | 20:45 | |
| tdasilva | notmyname: hi, can you share a link to the community graphs you talked about last week? | 20:46 |
| openstackgerrit | Minwoo Bae proposed openstack/swift: Error handling of DiskFileNoSpace() https://review.openstack.org/198909 | 20:46 |
| notmyname | tdasilva: https://github.com/notmyname/git-stats | 20:48 |
| *** robefran has quit IRC | 20:49 | |
| notmyname | tdasilva: gotta run ./build_swift_gerrit_history.sh and then ./swift_contrib_stats.py | 20:49 |
| notmyname | tdasilva: and probably update it to not use my gerrit id and local paths ;-) | 20:50 |
| tdasilva | notmyname: cool, thanks! | 20:50 |
| onovy | zigo, hi, are you here? | 20:51 |
| *** dustins has quit IRC | 20:52 | |
| *** dbarman has quit IRC | 20:55 | |
| *** dbarman1 has quit IRC | 20:55 | |
| openstackgerrit | Bill Huber proposed openstack/swift: Quorum on durable response is too low https://review.openstack.org/213822 | 21:06 |
| clayg | wbhuber: I may be naming the "phases" of the ec PUT business in correctly | 21:12 |
| clayg | wbhuber: replication has getexpect and getresponse | 21:12 |
| clayg | wbhuber: most of the time in unittests when it's needed the fake's getexpect will return 100 | 21:13 |
| clayg | wbhuber: in EC land there's *another* 100 continue and the fake grew some arms to deal with introducing failures there - whatever *that* pahse is called - the second 100-continue - that's a different failure than when you just give the fake connect a 503 | 21:14 |
| onovy | mattoliverau, thanks! :) | 21:14 |
| *** zul has quit IRC | 21:14 | |
| wbhuber | clayg: ah, gotcha. | 21:14 |
| clayg | and since that "second" phase (or whatever it's called) has to do with durables - seems like an interesting place to play with errors | 21:14 |
| mattoliverau | onovy: sorry I didn't answer before, I'm not at my computer too much today :) | 21:14 |
| clayg | mattoliverau: how's that san antonio heat!? | 21:14 |
| mattoliverau | clayg: its cooler then Austin (but only be a few degrees) :( | 21:15 |
| mattoliverau | clayg: luckly the castle is nice and cool :) | 21:16 |
| wbhuber | mattoliverau: that's rather odd. SAT is actually 60 miles south of Austin - you should be feeling more heat than Austin | 21:16 |
| clayg | mattoliverau: yay AC! | 21:16 |
| wbhuber | Austin got much cooler after the hackathon. | 21:16 |
| wbhuber | Guess it got too heated when you guys were here. | 21:16 |
| mattoliverau | wbhuber: well it feels cooler, maybe I'm now just aclimatised ;) | 21:16 |
| clayg | wbhuber: lol! | 21:16 |
| mattoliverau | wbhuber: lol, it was great to meet you in person :) | 21:17 |
| wbhuber | mattoliverau: likewise and i hope you remember the two signs "good morning and good night." | 21:17 |
| mattoliverau | wbhuber: I'll make sure I practice for next time! | 21:18 |
| wbhuber | mattoliverau: i think it is a combination of you getting acclimatized & weather cooling down. that's the icing of your trip. | 21:18 |
| onovy | what do you think about setting ionice/nice of daemons directly in swift configuration files? we are using cron job for this, but it's not perfect. | 21:19 |
| mattoliverau | wbhuber: ready for me to travel back home to winter | 21:19 |
| wbhuber | mattoliverau: it's really that brisk back home enough to wear a trapper hat? | 21:20 |
| mattoliverau | wbhuber: nah, it's Australia.. never really gets that cold ;) | 21:21 |
| wbhuber | ;-) | 21:21 |
| *** robefran has joined #openstack-swift | 21:29 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Make swift-ring-builder filename usage more consistent https://review.openstack.org/209787 | 21:31 |
| mattoliverau | cschwede: ^^ that patchset is just replacing some tabs with spaces that you placed in the test. | 21:32 |
| mattoliverau | this is weird, I'm used to acoles_ and cschwede to start coming online now.. but they wont cause I'm not in my timezone. So I guess cschwede will see that later. | 21:34 |
| notmyname | mattoliverau: now you know how the rest of us feel about our european friends | 21:35 |
| mattoliverau | yeah :( why can't they be like clayg and never sleep. | 21:35 |
| onovy | nobody needs ionice? | 21:43 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:45 |
| notmyname | onovy: and there you go. the timing recon patch has been marked as approved | 21:46 |
| clayg | notmyname: oh sure - take the ducks | 21:46 |
| onovy | notmyname, noway! :) | 21:47 |
| notmyname | take the ducks? mattoliverau and torgomatic reviewed it | 21:47 |
| onovy | cool | 21:47 |
| notmyname | what is "take the ducks"? | 21:47 |
| clayg | lol | 21:47 |
| mattoliverau | clayg: are you calling me a duck :P | 21:48 |
| clayg | apparently this is not a widely used euphemism | 21:49 |
| clayg | it's a billards term for balls that are close to the pocket | 21:50 |
| clayg | i.e. "the easy ones" | 21:50 |
| clayg | https://en.wiktionary.org/wiki/Appendix:Glossary_of_pool,_billiards_and_snooker | 21:50 |
| notmyname | clayg: google says it's this: http://www.wikihow.com/Take-Care-of-Ducks | 21:50 |
| onovy | notmyname, lol | 21:51 |
| clayg | notmyname: yeah that's it | 21:51 |
| hrou | wow I love the fact there is a part 3 "activities for the ducks", apparently they like to swim : ) | 21:52 |
| *** wbhuber has quit IRC | 22:00 | |
| *** petertr7 is now known as petertr7_away | 22:01 | |
| *** jkugel has quit IRC | 22:01 | |
| *** hrou has quit IRC | 22:05 | |
| mattoliverau | clayg: a duck in cricket (yes that game you don't understand) is a batter getting out with 0 runs. | 22:06 |
| *** NM has quit IRC | 22:06 | |
| notmyname | I understand it more now, thanks to mattoliverau and acoles_ | 22:06 |
| jrichli | lol, the things you can learn on swift IRC! | 22:07 |
| notmyname | (at swift hackathons) | 22:07 |
| mattoliverau | notmyname: then we've done our jobs well :) | 22:07 |
| jrichli | yes, we all learned how to select good tequila. :-) | 22:09 |
| * jrichli is outta here | 22:11 | |
| *** jrichli has quit IRC | 22:11 | |
| notmyname | mattoliverau: ok, here's what I learned about cricket: | 22:12 |
| notmyname | Cricket is a game where Batman" has to protect Christian "hay" Bales from the bowler. | 22:12 |
| notmyname | The bowler is called such because he wears a bowler hat. After Batman hits the ball away from the hay bales, he switches places with someone at another wicket (a stick in the ground). | 22:12 |
| notmyname | The more times he runs to the wicket, the more points he scores. The trick is that everyone moves around the field during the game, including the bowler. | 22:12 |
| notmyname | Games last for days and days and they take tea-time breaks. The winner gets to burn down the other team's hay bales, thus the name given to the big tournament: The Ashes. | 22:12 |
| mattoliverau | notmyname: lol! close enough :P | 22:13 |
| notmyname | I left the part out where there are 60 ways to get someone out while they are runnign to the wicket. sometimes even without them getting to a wicket at all | 22:14 |
| *** nadeem has quit IRC | 22:14 | |
| mattoliverau | and now you know about ducks | 22:15 |
| notmyname | ah! that's it! the visiting team brings ducks and has to hold them during the game. the winner takes the ducks home. thus, "taking the ducks" | 22:16 |
| notmyname | also, I had to be careful to not s/u/i/ there ;-) | 22:16 |
| *** jlhinson has quit IRC | 22:16 | |
| mattoliverau | notmyname: lol, I'd love you see you start up cricket in SFO with those rules :P | 22:18 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 22:20 |
| mattoliverau | notmyname: so mrda sent me this.. and it'll clear up cricket for you :P | 22:20 |
| mattoliverau | You have two sides, one out in the field and one in. Each man that's in the side that's in goes out, and when he's out he comes in and the next man goes in until he's out. When they are all out, the side that's out comes in and the side that's been in goes out and tries to get those coming in, out. | 22:20 |
| mattoliverau | Sometimes you get men still in and not out. When a man goes out to go in, the men who are out try to get him out, and when he is out he goes in and the next man in goes out and goes in. | 22:20 |
| mattoliverau | There are two men called umpires who stay all out all the time and they decide when the men who are in are out. When both sides have been in and all the men have been out, and both sides have been out twice after all the men have been in, including those who are not out, that is the end of the game! | 22:20 |
| mattoliverau | </sarcasm> | 22:21 |
| notmyname | that actually makes sense to me now! ;-) | 22:21 |
| mattoliverau | lol | 22:22 |
| *** marcusvrn has quit IRC | 22:23 | |
| mattoliverau | k, I'm getting a lift back to my hotel now | 22:24 |
| torgomatic | what the duck just happened in here | 22:27 |
| peluse1 | i didn't know tere were fellow pool players in the community... do they have pool tables in Tokyo? :) | 22:28 |
| *** annegentle has quit IRC | 22:30 | |
| * peluse1 off to Apple store to try and get laptop fixed... | 22:30 | |
| *** zul has joined #openstack-swift | 22:33 | |
| *** esker has quit IRC | 22:33 | |
| *** chlong has quit IRC | 22:34 | |
| *** minwoob has quit IRC | 22:38 | |
| *** lpabon has joined #openstack-swift | 22:44 | |
| *** km has joined #openstack-swift | 22:57 | |
| *** dbarman1 has joined #openstack-swift | 23:01 | |
| *** ho has joined #openstack-swift | 23:03 | |
| *** zul has quit IRC | 23:04 | |
| *** hrou has joined #openstack-swift | 23:07 | |
| *** nadeem has joined #openstack-swift | 23:10 | |
| ho | good morning! | 23:11 |
| *** lcurtis has quit IRC | 23:17 | |
| *** kei_yama has joined #openstack-swift | 23:17 | |
| *** sakaYK has joined #openstack-swift | 23:20 | |
| *** lpabon has quit IRC | 23:22 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (PUT ACL) with Keystone https://review.openstack.org/213608 | 23:37 |
| *** nadeem has quit IRC | 23:39 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (DELETE ACL) with Keystone https://review.openstack.org/213629 | 23:39 |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add functional test for access control (GET ACL) with Keystone https://review.openstack.org/213934 | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!