| mattoliverau | clayg: o/ | 00:04 |
|---|---|---|
| mattoliverau | acoles_: o/ | 00:04 |
| kota_ | clayg, acoles: o/ | 00:05 |
| *** chsc has quit IRC | 00:11 | |
| *** lyrrad has quit IRC | 00:13 | |
| *** kota_ has quit IRC | 00:14 | |
| *** zhill has quit IRC | 00:20 | |
| *** garthb has quit IRC | 00:21 | |
| *** dmorita has joined #openstack-swift | 00:24 | |
| *** bill_az has quit IRC | 00:40 | |
| *** m_kazuhiro has joined #openstack-swift | 00:43 | |
| *** nexusz99 has joined #openstack-swift | 00:55 | |
| *** nexusz99 has quit IRC | 00:55 | |
| *** minwoob has quit IRC | 00:56 | |
| *** blmartin_ has quit IRC | 00:57 | |
| *** klrmn1 has quit IRC | 01:06 | |
| *** thumpba has joined #openstack-swift | 01:14 | |
| *** pberis has joined #openstack-swift | 01:30 | |
| *** dustins has joined #openstack-swift | 01:34 | |
| *** thumpba has quit IRC | 01:38 | |
| *** dencaval has quit IRC | 01:41 | |
| *** pberis has quit IRC | 01:41 | |
| *** kota_ has joined #openstack-swift | 01:42 | |
| *** ChanServ sets mode: +v kota_ | 01:42 | |
| kota_ | back again | 01:42 |
| *** thumpba has joined #openstack-swift | 01:51 | |
| *** jerrygb has joined #openstack-swift | 02:09 | |
| notmyname | zigo: just curious, in relation to your email about liberty in debian, what did you use for swift there? we haven't even done an RC for liberty yet, so what does "Liberty RC1 has been fully uploaded..." mean? | 02:11 |
| *** jrichli has joined #openstack-swift | 02:18 | |
| *** lcurtis has quit IRC | 02:20 | |
| openstackgerrit | Janie Richling proposed openstack/swift: Decrypting Container Listing https://review.openstack.org/214438 | 02:32 |
| openstackgerrit | Janie Richling proposed openstack/swift: Encrypting/Decrypting object metadata https://review.openstack.org/203310 | 02:32 |
| *** pberis has joined #openstack-swift | 02:33 | |
| hrou | jrichli, ^ rebase? | 02:33 |
| jrichli | hrou - yes, but i think I have something to fix from that already. give me a min | 02:35 |
| hrou | jrichli, oh great work though ! Yea lots changed from the base patch there | 02:35 |
| *** sanchitmalhotra has joined #openstack-swift | 02:44 | |
| *** klrmn1 has joined #openstack-swift | 02:45 | |
| *** pberis has quit IRC | 02:47 | |
| *** haomaiwang has joined #openstack-swift | 02:50 | |
| *** haomaiwang has quit IRC | 02:50 | |
| *** haomaiwa_ has joined #openstack-swift | 02:51 | |
| *** wbhuber_ has joined #openstack-swift | 02:56 | |
| *** m_kazuhiro has quit IRC | 02:59 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-swift | 03:01 | |
| *** m_kazuhiro has joined #openstack-swift | 03:03 | |
| *** kota_ has quit IRC | 03:06 | |
| *** dustins has quit IRC | 03:07 | |
| *** willsama has joined #openstack-swift | 03:07 | |
| *** silor has joined #openstack-swift | 03:09 | |
| openstackgerrit | Janie Richling proposed openstack/swift: Encrypting/Decrypting object metadata https://review.openstack.org/203310 | 03:09 |
| jrichli | hrou - thanks. Let me know if you see issues ^^ | 03:10 |
| *** lcurtis has joined #openstack-swift | 03:11 | |
| hrou | jrichli, sure thing ! I'll pull those down now. | 03:11 |
| jrichli | hrou: just do that last one right now. I still gotta rebase the container listing one again ;-) | 03:11 |
| *** wbhuber_ has quit IRC | 03:11 | |
| hrou | hehe sounds good : ) Playing around with the json fix suggested by timburke, so far so good. | 03:12 |
| jrichli | great! thanks | 03:12 |
| *** wbhuber_ has joined #openstack-swift | 03:12 | |
| *** wbhuber_ has quit IRC | 03:16 | |
| *** trifon has joined #openstack-swift | 03:24 | |
| openstackgerrit | Janie Richling proposed openstack/swift: Decrypting Container Listing https://review.openstack.org/214438 | 03:25 |
| *** sanchitmalhotra1 has joined #openstack-swift | 03:31 | |
| *** sanchitmalhotra has quit IRC | 03:33 | |
| *** trifon has quit IRC | 03:33 | |
| *** lcurtis has quit IRC | 03:41 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:51 | |
| *** sanchitmalhotra1 has quit IRC | 03:53 | |
| *** links has joined #openstack-swift | 03:55 | |
| *** jerrygb has quit IRC | 03:57 | |
| *** jerrygb has joined #openstack-swift | 03:58 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 03:59 | |
| *** hrou has quit IRC | 04:00 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-swift | 04:01 | |
| *** sanchitmalhotra has quit IRC | 04:02 | |
| *** klrmn1 has quit IRC | 04:18 | |
| *** jrichli has quit IRC | 04:19 | |
| *** willsama has quit IRC | 04:23 | |
| *** mahatic has joined #openstack-swift | 04:29 | |
| *** StevenK has quit IRC | 04:35 | |
| *** StevenK has joined #openstack-swift | 04:36 | |
| *** kota_ has joined #openstack-swift | 04:36 | |
| *** ChanServ sets mode: +v kota_ | 04:36 | |
| *** ppai has joined #openstack-swift | 04:44 | |
| *** ktsuyuzaki has joined #openstack-swift | 04:47 | |
| *** kota_ has quit IRC | 04:49 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-swift | 05:01 | |
| *** chsc has joined #openstack-swift | 05:01 | |
| *** trifon has joined #openstack-swift | 05:09 | |
| *** ktsuyuzaki has left #openstack-swift | 05:22 | |
| *** ktsuyuzaki has joined #openstack-swift | 05:22 | |
| *** ktsuyuzaki has left #openstack-swift | 05:24 | |
| *** kota_ has joined #openstack-swift | 05:25 | |
| *** ChanServ sets mode: +v kota_ | 05:25 | |
| *** chsc has quit IRC | 05:26 | |
| *** sanchitmalhotra has joined #openstack-swift | 05:33 | |
| *** sanchitmalhotra1 has quit IRC | 05:35 | |
| *** kota_ has quit IRC | 05:48 | |
| *** haomaiwang has quit IRC | 05:54 | |
| *** haomaiwang has joined #openstack-swift | 05:54 | |
| *** trifon has quit IRC | 06:00 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-swift | 06:01 | |
| *** trifon has joined #openstack-swift | 06:02 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 06:24 | |
| *** sanchitmalhotra has quit IRC | 06:26 | |
| *** ktsuyuzaki has joined #openstack-swift | 06:26 | |
| *** ktsuyuzaki has quit IRC | 06:28 | |
| *** sanchitmalhotra has joined #openstack-swift | 06:32 | |
| *** sanchitmalhotra1 has quit IRC | 06:34 | |
| *** ktsuyuzaki has joined #openstack-swift | 06:36 | |
| *** ktsuyuzaki has quit IRC | 06:36 | |
| *** SkyRocknRoll has joined #openstack-swift | 06:41 | |
| *** janonymous has joined #openstack-swift | 06:51 | |
| *** kota_ has joined #openstack-swift | 06:52 | |
| *** kota_ has quit IRC | 06:53 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 07:00 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** SkyRocknRoll has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:01 | |
| *** SkyRocknRoll_ has quit IRC | 07:01 | |
| *** kota_ has joined #openstack-swift | 07:01 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 07:02 | |
| kota_ | test, test | 07:02 |
| kota_ | I've setup znc in my cloud server | 07:02 |
| *** SkyRocknRoll_ has quit IRC | 07:04 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 07:04 | |
| *** kota_ has quit IRC | 07:05 | |
| *** rledisez has joined #openstack-swift | 07:05 | |
| *** kota_ has joined #openstack-swift | 07:06 | |
| kota_ | test, again | 07:06 |
| *** wolsen has quit IRC | 07:08 | |
| m_kazuhiro | kota_ :test | 07:08 |
| kota_ | ok, znc works well in my environment. | 07:10 |
| kota_ | I won't lose playbacks anymore, yey. Next, i wann get notifications when someone ping me... | 07:11 |
| kota_ | oh yeah, irssi which clayg said seems good. | 07:15 |
| *** thumpba has quit IRC | 07:21 | |
| *** thumpba has joined #openstack-swift | 07:25 | |
| mattoliverau | kota_: nice work! now you can always be pinged :) | 07:32 |
| mattoliverau | kota_: prepare to work more :P | 07:32 |
| *** eranrom has joined #openstack-swift | 07:32 | |
| *** thumpba has quit IRC | 07:33 | |
| kota_ | mattoliverau: might be no holidays? :\ | 07:37 |
| kota_ | kota_: notification test | 07:40 |
| mattoliverau | kota_: notification test 2 :) | 07:40 |
| kota_ | mattoliverau: just looking at so no notification at your ping :/ | 07:41 |
| kota_ | closed my phone, and please note me again. | 07:41 |
| mattoliverau | kota_: ping | 07:42 |
| kota_ | mattoliverau: I got it! | 07:42 |
| mattoliverau | \o/ | 07:42 |
| kota_ | mattoliverau: thanks for the test :) | 07:42 |
| mattoliverau | Nps, remember to out your phone on silent ;) | 07:43 |
| mattoliverau | At night I mean | 07:43 |
| *** openstackgerrit has quit IRC | 07:46 | |
| *** acoles_ is now known as acoles | 07:46 | |
| *** openstackgerrit has joined #openstack-swift | 07:46 | |
| acoles | kota_: good work on bouncer! | 07:48 |
| *** geaaru has joined #openstack-swift | 07:51 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** silor has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-swift | 08:01 | |
| *** haomaiwang has quit IRC | 08:03 | |
| zigo | notmyname: I have a bit given up on Swift package maintenance as someone else who uses it more took over. What he did is just upload of the latest Swift, so that is 2.4.0, and in Sid, not Experimental. | 08:05 |
| zigo | If you are releasing out of sync, then probably we will also do the packaging out of sync. | 08:05 |
| *** itlinux has joined #openstack-swift | 08:06 | |
| zigo | It's also not so much of a problem for Swift, because of nicer requirements/dependencies. | 08:07 |
| *** jordanP has joined #openstack-swift | 08:08 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 08:18 | |
| *** jistr has joined #openstack-swift | 08:18 | |
| *** sanchitmalhotra has quit IRC | 08:20 | |
| *** itlinux has quit IRC | 08:26 | |
| *** thumpba has joined #openstack-swift | 08:30 | |
| *** thumpba has quit IRC | 08:34 | |
| *** sanchitmalhotra has joined #openstack-swift | 08:36 | |
| *** haomaiwang has joined #openstack-swift | 08:37 | |
| *** ho has joined #openstack-swift | 08:38 | |
| *** sanchitmalhotra1 has quit IRC | 08:39 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 08:56 | |
| *** sanchitmalhotra has quit IRC | 08:58 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** 6A4AALXHF has joined #openstack-swift | 09:01 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Fix copy requests to service accounts in Keystone https://review.openstack.org/227133 | 09:02 |
| *** mac_ified has quit IRC | 09:02 | |
| acoles | ho: donagh : notmyname ^^ fixes done. thanks for reviews! | 09:02 |
| *** aix has joined #openstack-swift | 09:04 | |
| ho | acoles: thanks for it. looks perfect! | 09:05 |
| acoles | ho: great. if you are +1 then perhaps we can ask mattoliverau to take another look | 09:10 |
| ho | acoles: yeah, i'm going to test it now. please wait | 09:11 |
| acoles | ho: sure. thanks. | 09:11 |
| *** foexle has joined #openstack-swift | 09:14 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add further unit test for cross-prefix copy auth https://review.openstack.org/229787 | 09:17 |
| acoles | ho: donagh cschwede mattoliverau notmyname ^^ deliberately a follow-on patch so as to to not delay patch 227133 | 09:17 |
| patchbot | acoles: https://review.openstack.org/#/c/227133/ - Fix copy requests to service accounts in Keystone | 09:17 |
| *** proteusguy_ has quit IRC | 09:18 | |
| *** mahatic_ has joined #openstack-swift | 09:24 | |
| *** mahatic has quit IRC | 09:24 | |
| *** proteusguy_ has joined #openstack-swift | 09:31 | |
| *** marzif has joined #openstack-swift | 09:32 | |
| *** 6A4AALXHF has quit IRC | 09:52 | |
| *** haomaiwang has joined #openstack-swift | 09:52 | |
| acoles | ho: thanks for testing and +1 | 09:52 |
| ho | acoles: you are welcome :-) | 09:53 |
| acoles | ho: have a good evening | 09:53 |
| ho | acoles: thanks! good morning :-) | 09:53 |
| acoles | donagh: thanks too for +1 | 09:54 |
| acoles | hopefully we can land it this week | 09:54 |
| *** mahatic_ has quit IRC | 09:58 | |
| ho | donagh, acoles: we have similar tests for the problem so I would like to arrange them. acoles' unittest is overlapped with donagh's functional test and donagh's functional test is part of my functional test. I think acoles' unittest and my functional test are covered all cases but my patches are big so if we need to commit them ASAP it's better to proceed acoles' unittest and donagh's functional test first. | 10:00 |
| *** haomaiwang has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| acoles | ho: my unit test is not as good as a func test. last i looked donagh's test needed more work. i wish i had found time to review your func test patches more :( | 10:03 |
| *** silor has joined #openstack-swift | 10:06 | |
| *** silor has quit IRC | 10:06 | |
| *** sanchitmalhotra has joined #openstack-swift | 10:12 | |
| *** silor has joined #openstack-swift | 10:12 | |
| *** sanchitmalhotra1 has quit IRC | 10:14 | |
| hugokuo | had anyone experienced with container-sync SLO object ever ? (Seems not work) | 10:18 |
| *** silor has quit IRC | 10:20 | |
| *** silor has joined #openstack-swift | 10:21 | |
| onovy | acoles, cschwede, clayg, mattoliverau, notmyname: anyone pls? :) https://review.openstack.org/#/c/210736/ | 10:22 |
| *** thumpba has joined #openstack-swift | 10:32 | |
| *** thumpba has quit IRC | 10:37 | |
| *** dmorita has quit IRC | 10:47 | |
| *** acoles is now known as acoles_ | 10:55 | |
| ho | acoles: ok, now i reviewed your unittest but it's bit difficult to understand the necessity of COPY(L452-455). I think you might want to test internal behavior of copy. | 10:56 |
| ho | acoles: if so, my expectation is response code of the copy is same as response code of last request. | 10:57 |
| ho | acoles: s/my expectation is response/my expectation for response/ | 10:59 |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| ho | acoles: tomorrow I will continue to review it. have a nice day! | 11:01 |
| *** ho has quit IRC | 11:02 | |
| *** marzif has quit IRC | 11:04 | |
| *** marzif has joined #openstack-swift | 11:05 | |
| vinsh | Why am I swifting wrong? --> "ERROR: Invalid Storage Policy Configuration in /etc/swift/swift.conf (Invalid option 'policy_type' in storage-policy section" | 11:06 |
| vinsh | Isn't that a valid option? | 11:07 |
| *** SkyRocknRoll_ has quit IRC | 11:07 | |
| *** acoles_ is now known as acoles | 11:13 | |
| *** marzif has quit IRC | 11:14 | |
| *** mahatic has joined #openstack-swift | 11:14 | |
| *** marzif has joined #openstack-swift | 11:15 | |
| *** marzif has quit IRC | 11:16 | |
| *** marzif has joined #openstack-swift | 11:16 | |
| *** ppai has quit IRC | 11:17 | |
| vinsh | Why is policy_type an invalid option? | 11:17 |
| vinsh | policy_type = replication | 11:19 |
| acoles | vinsh: policy_type only got added as a valid option with erasure codes, so kilo release | 11:20 |
| vinsh | ah.. 2.2.2-0ubuntu1~cloud0 | 11:21 |
| vinsh | I had just assumed I was already there. d'oh | 11:21 |
| acoles | :) | 11:21 |
| vinsh | amateur hour here :) | 11:21 |
| *** marzif has quit IRC | 11:26 | |
| vinsh | acoles: Any idea why UCA skipped 2.3.0 ? | 11:26 |
| *** ccavanna_ has quit IRC | 11:26 | |
| *** marzif has joined #openstack-swift | 11:26 | |
| vinsh | I see 2.2.2 then 2.4.0 | 11:26 |
| *** aix has quit IRC | 11:27 | |
| acoles | vinsh: idk sorry | 11:28 |
| vinsh | Ok thanks. I gather 2.3.0 is kilo.. but don't see it in there. I'll dig. | 11:29 |
| acoles | yes kilo is 2.3.0 | 11:29 |
| vinsh | Might as well go right to liberty :) | 11:30 |
| vinsh | looks like 2.2.2 is part of the kilo series.. but is pre release. | 11:30 |
| vinsh | so no EC. | 11:30 |
| *** ppai has joined #openstack-swift | 11:31 | |
| *** thumpba has joined #openstack-swift | 11:33 | |
| *** marzif has quit IRC | 11:36 | |
| *** gustavo_ has quit IRC | 11:37 | |
| *** marzif has joined #openstack-swift | 11:37 | |
| *** thumpba has quit IRC | 11:37 | |
| *** kei_yama has quit IRC | 11:39 | |
| vinsh | Cononical seems to have just skipped kilo entirely. Going to ping them for feedback. | 11:46 |
| vinsh | also, how am I just now noticing this? :) | 11:46 |
| *** marzif has quit IRC | 11:46 | |
| *** marzif has joined #openstack-swift | 11:49 | |
| vinsh | acoles: there it is : https://javacruft.wordpress.com/ | 11:52 |
| vinsh | "NOTE: We’re shipping Swift 2.2.2 for release – due to the relatively late inclusion of new dependencies to support erasure coding in Swift, we’ve opted not to update to 2.3.0 this cycle in Ubuntu. | 11:52 |
| vinsh | Lame. | 11:52 |
| acoles | vinsh: interesting | 11:55 |
| vinsh | Yeah. Either build your own or be held to their whims I guess. | 11:56 |
| *** ppai has quit IRC | 12:00 | |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-swift | 12:01 | |
| *** MVenesio has joined #openstack-swift | 12:04 | |
| *** km has quit IRC | 12:05 | |
| *** marcusvrn_ has joined #openstack-swift | 12:06 | |
| *** marzif has quit IRC | 12:06 | |
| *** marzif has joined #openstack-swift | 12:07 | |
| *** ppai has joined #openstack-swift | 12:12 | |
| *** marzif has quit IRC | 12:17 | |
| *** jerrygb has quit IRC | 12:17 | |
| *** marzif has joined #openstack-swift | 12:17 | |
| *** haomaiwang has quit IRC | 12:18 | |
| *** haomaiwang has joined #openstack-swift | 12:19 | |
| *** alejandrito has joined #openstack-swift | 12:27 | |
| *** thumpba has joined #openstack-swift | 12:34 | |
| *** m_kazuhiro has quit IRC | 12:44 | |
| *** mahatic has quit IRC | 12:46 | |
| *** thumpba has quit IRC | 12:50 | |
| *** bill_az has joined #openstack-swift | 12:53 | |
| *** delattec has joined #openstack-swift | 12:56 | |
| *** cdelatte has quit IRC | 12:58 | |
| *** mac_ified has joined #openstack-swift | 12:58 | |
| *** mahatic has joined #openstack-swift | 12:59 | |
| *** haomaiwang has quit IRC | 13:01 | |
| *** haomaiwang has joined #openstack-swift | 13:01 | |
| *** pberis has joined #openstack-swift | 13:04 | |
| *** eranrom has quit IRC | 13:08 | |
| *** hrou has joined #openstack-swift | 13:11 | |
| *** ppai has quit IRC | 13:13 | |
| *** dustins has joined #openstack-swift | 13:18 | |
| *** marzif has quit IRC | 13:20 | |
| *** wbhuber_ has joined #openstack-swift | 13:20 | |
| *** marzif has joined #openstack-swift | 13:20 | |
| *** marzif has quit IRC | 13:25 | |
| *** DjamOlsky has joined #openstack-swift | 13:25 | |
| *** trifon has quit IRC | 13:28 | |
| *** jrichli has joined #openstack-swift | 13:33 | |
| *** wbhuber_ has quit IRC | 13:39 | |
| *** wbhuber_ has joined #openstack-swift | 13:40 | |
| *** wbhuber_ has quit IRC | 13:44 | |
| vinsh | Is the ec_object_segment_size buffered in proxy memory? | 13:47 |
| vinsh | I set that to 1G.. vs default.. but don't see any more memory consumed. | 13:48 |
| *** ccavanna has joined #openstack-swift | 13:48 | |
| vinsh | (after restarting services of course) | 13:48 |
| wbhuber | clayg: peluse: acoles: when you have a cluster running with replication in production, how do we migrate from a replication-only policy to multiple policies running on cluster without stopping any services? | 13:50 |
| wbhuber | I assume stop one or two of the proxy servers (at least three) to ensure that transacations are kept running for new/lookup objects | 13:50 |
| vinsh | Could just do a graceful reload. | 13:51 |
| wbhuber | vinsh: what are the steps to do a graceful reload? | 13:51 |
| vinsh | swift-init reload vs vestart | 13:51 |
| vinsh | *restart | 13:51 |
| vinsh | Take a look at the help output for swift-init | 13:51 |
| wbhuber | vinsh: did you try testing that yet? | 13:52 |
| vinsh | how so? | 13:52 |
| vinsh | Your going to have to restart services to load the new config.. thats a smoother way | 13:52 |
| wbhuber | but that'd cease some of the operations that are running in the meantime, correct? | 13:53 |
| vinsh | " reload: graceful shutdown then restart on supporting servers" | 13:53 |
| wbhuber | that's in production - we would like to have no downtime during migration | 13:53 |
| vinsh | meaning.. it lets current requests complete before reloading | 13:53 |
| vinsh | one proxy reloads the new config.. then the other. no down time. | 13:54 |
| wbhuber | ok, that sounds feasible | 13:54 |
| vinsh | are you using puppet or? | 13:54 |
| vinsh | what are you configuring swift with? | 13:54 |
| wbhuber | ansible if i understood your Q right | 13:55 |
| vinsh | Ok, that should work well with ansible. push config out. reload proxy one. then reload proxy two. | 13:55 |
| vinsh | Take a peek at the docs for swift-init... its got all the magic you need. | 13:56 |
| wbhuber | vinsh: thumbs up | 13:56 |
| *** links has quit IRC | 13:57 | |
| *** aix has joined #openstack-swift | 14:00 | |
| *** haomaiwang has quit IRC | 14:01 | |
| *** haomaiwang has joined #openstack-swift | 14:01 | |
| *** willsama has joined #openstack-swift | 14:07 | |
| *** lyrrad has joined #openstack-swift | 14:09 | |
| *** david-lyle has quit IRC | 14:11 | |
| *** david-lyle has joined #openstack-swift | 14:13 | |
| *** mahatic has quit IRC | 14:14 | |
| *** mahatic has joined #openstack-swift | 14:14 | |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift: Guess content type for PUT and probe tests https://review.openstack.org/220897 | 14:24 |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift: Modified tests to include real crypto https://review.openstack.org/211451 | 14:24 |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift: Encrypt flag for probe tests https://review.openstack.org/216974 | 14:24 |
| *** thumpba has joined #openstack-swift | 14:26 | |
| *** thumpba has quit IRC | 14:26 | |
| *** thumpba has joined #openstack-swift | 14:26 | |
| vinsh | I see my memory being consumed up to segment size now.. all is well. | 14:34 |
| *** jlhinson has joined #openstack-swift | 14:39 | |
| *** thurloat is now known as thurloat_isgone | 14:42 | |
| wbhuber | vinsh - an object up to 1 gb consuming your memory accordingly? | 14:42 |
| vinsh | I think | 14:43 |
| vinsh | yeah. failed.. but thats probably due to max chunk size in general. | 14:43 |
| wbhuber | 64k chunk size? | 14:43 |
| vinsh | I think 5G | 14:45 |
| wbhuber | what happens if you lower the chunk size but leave the seg size as is | 14:45 |
| vinsh | Not sure, have to work on other stuff first :) | 14:45 |
| wbhuber | :) | 14:45 |
| notmyname | good morning | 14:53 |
| vinsh | 'mornin | 14:54 |
| *** minwoob has joined #openstack-swift | 14:58 | |
| *** haomaiwang has quit IRC | 14:59 | |
| *** haomaiwang has joined #openstack-swift | 15:00 | |
| *** sayali has quit IRC | 15:00 | |
| *** haomaiwang has quit IRC | 15:01 | |
| *** haomaiwa_ has joined #openstack-swift | 15:01 | |
| notmyname | so canonical hasn't released anything for swift more recent than 2.2.2 because they don't think we're a responsible set of devs who know how to have a beta feature? | 15:02 |
| *** sayali has joined #openstack-swift | 15:02 | |
| *** haomaiwa_ has quit IRC | 15:04 | |
| *** wolsen has joined #openstack-swift | 15:05 | |
| *** nadeem has joined #openstack-swift | 15:08 | |
| *** nadeem has quit IRC | 15:11 | |
| vinsh | I guess... i'm grumpy at canonical for that. | 15:12 |
| vinsh | They have a 2.4.0 out there now. | 15:12 |
| vinsh | But skipped 2.3.0 entirely | 15:12 |
| notmyname | vinsh: I'm not surprised that they may skip versions | 15:13 |
| vinsh | yeah, and thats the cost I pay for consuming their packages vs building my own. | 15:13 |
| notmyname | swift has historically released between the openstack integrated releases, and also at the integrated release | 15:13 |
| vinsh | Saves time in general | 15:13 |
| notmyname | distros afaik just take the integrated one | 15:13 |
| notmyname | however, now in openstack there isn't an "integrated release". and many of the other projects are also doing mid-cycle releases | 15:14 |
| notmyname | what I'm annoyed at is that we spend extra effort to do the end-of-cycle release and distros seemingly skip it or don't pick it up and otherwise say "swift is different so deal with it" (which I reject) | 15:15 |
| vinsh | I'm with ya on that. :-/ | 15:16 |
| vinsh | I would have been more vocal with them had I realized it was the case back when. | 15:16 |
| vinsh | For now, I'll run 2.4.0 on my EC cluster. Move 2.4.0 without EC to production after a little soak time. | 15:18 |
| notmyname | cool | 15:18 |
| notmyname | many people are using 2.4.0. nobody is using ec in prod. that's ok | 15:18 |
| vinsh | this EC cluster is a POC to eval some beefy nodes we bought. | 15:18 |
| *** jerrygb has joined #openstack-swift | 15:19 | |
| notmyname | I think we'll start seeing people use ec in prod after this next release | 15:19 |
| vinsh | if this hardware pans out.. we will be buying about 10x more of it for a larger EC poc | 15:19 |
| *** jerrygb has quit IRC | 15:19 | |
| notmyname | (after the current priority patches land) | 15:19 |
| notmyname | very cool | 15:19 |
| vinsh | I'm excited :) | 15:19 |
| *** jerrygb has joined #openstack-swift | 15:19 | |
| *** MVenesio has quit IRC | 15:22 | |
| *** MVenesio has joined #openstack-swift | 15:23 | |
| *** chsc has joined #openstack-swift | 15:25 | |
| *** zhill has joined #openstack-swift | 15:31 | |
| *** chsc has quit IRC | 15:32 | |
| *** DjamOlsky has quit IRC | 15:33 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Enable object server to return non-durable data https://review.openstack.org/215276 | 15:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add POST capability to ssync for .meta files https://review.openstack.org/138498 | 15:47 |
| *** nadeem has joined #openstack-swift | 15:48 | |
| *** rohit_ has joined #openstack-swift | 15:48 | |
| jrichli | mahatic : I saw that you mentioned patch 220897 resolves the TestSerialization issue. Sorry I haven't reviewed that yet. I will do that very soon :-) | 15:52 |
| patchbot | jrichli: https://review.openstack.org/#/c/220897/ - Guess content type for PUT and probe tests | 15:52 |
| *** Fin1te has joined #openstack-swift | 15:56 | |
| *** chsc has joined #openstack-swift | 15:58 | |
| *** lyrrad has quit IRC | 15:58 | |
| *** MVenesio has quit IRC | 15:58 | |
| *** alejandrito has quit IRC | 15:58 | |
| *** chsc has quit IRC | 16:03 | |
| *** MVenesio has joined #openstack-swift | 16:05 | |
| *** alejandrito has joined #openstack-swift | 16:05 | |
| *** nadeem has quit IRC | 16:07 | |
| *** nadeem has joined #openstack-swift | 16:08 | |
| *** breitz has quit IRC | 16:14 | |
| *** breitz has joined #openstack-swift | 16:14 | |
| *** rledisez has quit IRC | 16:24 | |
| *** blmartin has joined #openstack-swift | 16:27 | |
| *** garthb has joined #openstack-swift | 16:28 | |
| *** jistr has quit IRC | 16:29 | |
| *** silor has quit IRC | 16:29 | |
| *** trifon has joined #openstack-swift | 16:34 | |
| *** lyrrad has joined #openstack-swift | 16:34 | |
| *** lyrrad has quit IRC | 16:35 | |
| *** lyrrad has joined #openstack-swift | 16:43 | |
| *** amit213 has joined #openstack-swift | 16:43 | |
| *** garthb has quit IRC | 16:46 | |
| *** mahatic has quit IRC | 16:47 | |
| *** chsc has joined #openstack-swift | 16:47 | |
| notmyname | acoles: I /think/ I'm understandign what's going on with the keystone patch. but your subsequent test leaves me with a quesion https://review.openstack.org/#/c/229787/ | 16:56 |
| acoles | notmyname: yeah i just saw your review | 16:56 |
| *** Fin1te has quit IRC | 16:57 | |
| acoles | there's no real proxy in the tests so the headers are irrelevant tbh | 16:57 |
| notmyname | mostly I'm trying to figure it out rather than say it's good or bad | 16:57 |
| acoles | its not a great test | 16:57 |
| *** chsc has quit IRC | 16:58 | |
| notmyname | ok | 16:58 |
| notmyname | so why the Destination headers to start with? | 16:58 |
| acoles | i said it simulates a COPY - basically it says 'if i use a request env that get's allowed, then re-use the SAME env with a different, disallowed path, it should get denied | 16:59 |
| acoles | why Destination header? ummm, to make it look authentic i guess :) on reflection maybe it just confuses - ho also tripped up on that patch | 17:00 |
| notmyname | he | 17:00 |
| notmyname | *heh | 17:00 |
| acoles | a func test would be MUCH better | 17:00 |
| notmyname | yeah | 17:00 |
| acoles | so I am just reviewing patch 202411 again cos that is the start of v comprehensive access control testing suite | 17:01 |
| patchbot | acoles: https://review.openstack.org/#/c/202411/ - Add functional test for access control (PUT/DELETE... | 17:01 |
| acoles | notmyname: unfortunately we need to do some devstack work to get these tests (and other existing tests) to run on jenkins | 17:02 |
| notmyname | ah | 17:02 |
| acoles | or we get charz to set it all up on community cluster :P | 17:02 |
| notmyname | :-) | 17:02 |
| notmyname | I'd prefer devstack to start with, depending on what that takes | 17:03 |
| *** chsc has joined #openstack-swift | 17:03 | |
| acoles | agree. I would do it once "everything else" is done. I think maybe ho would take it on. | 17:04 |
| *** jerrygb has quit IRC | 17:04 | |
| notmyname | despite the lack of functional tests, patch 227133 seems ok. the bug makes sense. the patch seems right. just the whole service vs operator roles and who sets what when seems to make my brain stutter | 17:04 |
| patchbot | notmyname: https://review.openstack.org/#/c/227133/ - Fix copy requests to service accounts in Keystone | 17:04 |
| acoles | notmyname: consider it a gentle warm up for 215276 | 17:05 |
| notmyname | lol | 17:05 |
| notmyname | yeah, I should move on now | 17:05 |
| charz | acoles: need add something or? | 17:05 |
| notmyname | charz: nah, not yet :-) | 17:05 |
| charz | notmyname: k | 17:06 |
| notmyname | charz: was there something going on with the QA cluster recently? did you fix something? | 17:06 |
| notmyname | seemed like it hadn't been reporting, but now it is | 17:06 |
| acoles | charz: no, we were just kidding. thanks. | 17:06 |
| charz | notmyname: idk, the VM just no response. I can't reach the VM, so I restart it. | 17:07 |
| notmyname | acoles: I think I need to take the 2 patches as one. look at just patch 215276 and make sure both together make sense | 17:07 |
| patchbot | notmyname: https://review.openstack.org/#/c/215276/ - Enable object server to return non-durable data | 17:07 |
| notmyname | charz: ah, ok. thanks | 17:07 |
| notmyname | acoles: does that make sense from your perspective? will I miss anything by doing that? | 17:09 |
| charz | acoles: ha, it's good to see my nickname in irc again. :-P | 17:09 |
| acoles | notmyname: makes sense. there are three main areas of change: obj.py - the get response handling 'brains', ssync - do POSTs, then diskfile.py for the enabling changes: make richer info about on disk files available, open diskfile without a durable etc. | 17:11 |
| acoles | notmyname: to start with, ignore the ssync stuff, concentrate on the proxy <-> obj server GET interaction | 17:12 |
| notmyname | ok | 17:12 |
| notmyname | yeah, from looking at it earlier (from a high level), the concept makes a lot of sense. I'm just worried about missing nuances and systemic effects | 17:13 |
| acoles | me too :) | 17:13 |
| notmyname | but that's why we don't stop writing code, I guess. we find a bug and fix it | 17:13 |
| *** SkyRocknRoll has joined #openstack-swift | 17:14 | |
| acoles | notmyname: once you get to wondering why the ssync change is needed, run nosetests test.probe.test_reconstructor_rebuild:TestReconstructorRebuild.test_main on master and take a look at your filesystem for .data files at different timestamps on different nodes :/ | 17:14 |
| notmyname | right. it's stuff like that that I'm worried about :-) | 17:15 |
| notmyname | I totally get how that's a thing. but would I have caught that? I'm not sure | 17:15 |
| acoles | notmyname: arguably even if 215276 doesn't make it we should really try to land patch 138498 - it scares me that real clusters will start to accumulate diverse timestamps for objects and then it'll be hard to put a fix into place when there are already confused timistamps. | 17:17 |
| patchbot | acoles: https://review.openstack.org/#/c/138498/ - Add POST capability to ssync for .meta files | 17:17 |
| acoles | notmyname: i'll be away for a couple of hours, school thing to attend, but i'll check back later in case anyone has questions or wants to flame me ;) | 17:18 |
| notmyname | :-) | 17:19 |
| *** SkyRocknRoll has quit IRC | 17:28 | |
| *** diogogmt has joined #openstack-swift | 17:32 | |
| diogogmt | notmyname: thanks for the links, initially i was trying to use the swift python client in debug mode to find the sequence of requests executed when uploading a file, however, i couldn’t find much info, this is what i got from the debug logs: http://pastebin.com/exBmNDbu | 17:35 |
| diogogmt | the actual file was uplodaded by i don’t see any indication of the request that uploaded the file in the debug logs | 17:35 |
| *** alejandrito has quit IRC | 17:36 | |
| notmyname | diogogmt: the equivalent curl commands are printed in that debug output | 17:36 |
| notmyname | diogogmt: eg line 29 | 17:37 |
| *** klrmn1 has joined #openstack-swift | 17:37 | |
| *** alejandrito has joined #openstack-swift | 17:38 | |
| * notmyname steps away for a bit | 17:38 | |
| *** alejandrito has quit IRC | 17:46 | |
| *** alejandrito has joined #openstack-swift | 17:47 | |
| *** garthb has joined #openstack-swift | 17:47 | |
| *** NM has joined #openstack-swift | 17:51 | |
| *** jordanP has quit IRC | 17:52 | |
| *** lcurtis has joined #openstack-swift | 17:54 | |
| diogogmt | notmyname: but where is the file content transfered in the request? is it set in the headers or in the body? | 17:56 |
| *** Fin1te has joined #openstack-swift | 17:58 | |
| *** cdelatte has joined #openstack-swift | 18:06 | |
| *** delattec has quit IRC | 18:08 | |
| *** geaaru has quit IRC | 18:21 | |
| *** delattec has joined #openstack-swift | 18:22 | |
| *** cdelatte has quit IRC | 18:24 | |
| *** aix has quit IRC | 18:26 | |
| *** marzif has joined #openstack-swift | 18:27 | |
| *** cdelatte has joined #openstack-swift | 18:30 | |
| *** openstackgerrit has quit IRC | 18:31 | |
| *** openstackgerrit has joined #openstack-swift | 18:31 | |
| *** delattec has quit IRC | 18:32 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Add search filter examples to swift-ring-builder dispersion help https://review.openstack.org/227058 | 18:34 |
| *** sanchitmalhotra1 has joined #openstack-swift | 18:38 | |
| *** sanchitmalhotra has quit IRC | 18:40 | |
| *** CR7 has joined #openstack-swift | 18:43 | |
| *** marzif has quit IRC | 18:45 | |
| *** delattec has joined #openstack-swift | 18:51 | |
| *** delatte has joined #openstack-swift | 18:51 | |
| *** eranrom has joined #openstack-swift | 18:52 | |
| *** cdelatte has quit IRC | 18:53 | |
| *** chao_li has joined #openstack-swift | 18:54 | |
| *** cdelatte has joined #openstack-swift | 18:55 | |
| chao_li | Hi, nice to meet you guys, just wonder have any of you guys integrated CADF with swift before | 18:55 |
| *** delattec has quit IRC | 18:56 | |
| *** delatte has quit IRC | 18:56 | |
| notmyname | diogogmt: the file content is transferred in the request body | 18:56 |
| notmyname | diogogmt: so eg to upload something with curl, use the -T flag and give it a local file | 18:57 |
| diogogmt | notmyname: so are the debugs logs complete? shouldn’t curl have more arguments? | 18:57 |
| diogogmt | do i need to set the transfer encoding to header to chunked as well as the encoding type to gzip? | 18:58 |
| notmyname | diogogmt: they are there for a guide, definitely. it's not actually using curl under the covers, and they are there as a hint or guideline | 18:58 |
| *** thumpba has quit IRC | 18:58 | |
| *** thumpba_ has joined #openstack-swift | 18:58 | |
| *** delattec has joined #openstack-swift | 18:59 | |
| diogogmt | notmyname: ok, that makes sense — so the api exects to receive a chunked multi part PUT/POST request? | 19:00 |
| *** lcurtis has quit IRC | 19:00 | |
| *** rohit_ has quit IRC | 19:00 | |
| *** cdelatte has quit IRC | 19:01 | |
| *** pberis has quit IRC | 19:01 | |
| *** pberis has joined #openstack-swift | 19:02 | |
| notmyname | diogogmt: swift supports both content-length and chunked transfer encoding | 19:03 |
| notmyname | diogogmt: eg here's somethign I had saved a few days ago for something I was working on. (I don't even remember what) https://gist.github.com/notmyname/11ed48cdb554085e8307 | 19:04 |
| diogogmt | notmyname: thanks, i’m trying to gather some info about how the swift REST api works to later on connect to a node-js application | 19:05 |
| notmyname | cool | 19:05 |
| notmyname | those links I gave you earlier are the best place to start. they show requests and responses | 19:05 |
| notmyname | diogogmt: http://www.amazon.com/OpenStack-Swift-Administering-Developing-Storage/dp/1491900822/ref=sr_1_1 | 19:08 |
| notmyname | diogogmt: that has several chapters on the swift api | 19:08 |
| diogogmt | notmyname: great! that’s really handy, thanks | 19:08 |
| notmyname | diogogmt: also, you could request it from https://swiftstack.com/books/ (disclosure: where I work) | 19:08 |
| diogogmt | notmyname: do you have any promo codes i could use there? | 19:09 |
| notmyname | diogogmt: if you request it from swiftstack, we'll mail you one | 19:09 |
| *** bill_az has quit IRC | 19:17 | |
| *** david-lyle has quit IRC | 19:18 | |
| *** ccavanna has quit IRC | 19:25 | |
| *** david-lyle has joined #openstack-swift | 19:27 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Make LengthWrappers resettable if their _readable is https://review.openstack.org/226897 | 19:28 |
| *** delatte has joined #openstack-swift | 19:30 | |
| *** cdelatte has joined #openstack-swift | 19:30 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Add py35 to default tox environments https://review.openstack.org/230100 | 19:32 |
| *** delattec has quit IRC | 19:33 | |
| *** tellesnobrega has quit IRC | 19:33 | |
| *** thumpba_ has quit IRC | 19:37 | |
| *** itlinux has joined #openstack-swift | 19:44 | |
| *** logan2 has quit IRC | 19:47 | |
| *** bill_az has joined #openstack-swift | 19:48 | |
| *** [1]eranrom has joined #openstack-swift | 19:52 | |
| *** eranrom has quit IRC | 19:52 | |
| *** [1]eranrom is now known as eranrom | 19:52 | |
| *** logan2 has joined #openstack-swift | 19:55 | |
| *** lcurtis has joined #openstack-swift | 19:56 | |
| clayg | py35!? what are we crazy!? | 20:01 |
| notmyname | that's like 32 versions from now | 20:02 |
| notmyname | http://d.not.mn/tshirt_hint.jpg | 20:03 |
| jrichli | what a tease! | 20:04 |
| clayg | notmyname: it's not loading for me :'( | 20:04 |
| notmyname | clayg: took a while for me too | 20:04 |
| notmyname | mlanner: did you break the office proxy? ;-) | 20:04 |
| jrichli | you've got me wondering where the bird factors in | 20:05 |
| hrou | Or the Japanese word for bird (but maybe that was just me assuming : - ) | 20:06 |
| jrichli | well, i guess those silver lines could be the bottom of a wing | 20:06 |
| jrichli | nah .. | 20:06 |
| jrichli | maybe its white | 20:07 |
| *** thumpba has joined #openstack-swift | 20:08 | |
| *** thumpba has quit IRC | 20:08 | |
| *** thumpba has joined #openstack-swift | 20:09 | |
| *** Fin1te has quit IRC | 20:10 | |
| *** thumpba has quit IRC | 20:11 | |
| mlanner | notmyname: nope. wfm. i was going to remove the caching, but haven't ... yet. | 20:14 |
| *** thumpba has joined #openstack-swift | 20:19 | |
| *** thumpba has quit IRC | 20:20 | |
| clayg | I wonder if there's a backwards compatible way for http clients & servers to negotiate a request pipeling for requests that return 4/5XX instead of 100-continue that wouldn't require the body... | 20:20 |
| clayg | like instead of Expect: 100-continue it's like Require: 100-continue - and if if you give me anything else I'm just going to drop the the next request on the wire - and not send you any body. | 20:21 |
| acoles | notmyname: is that part of a 'do not tumble dry' symbol? ;-) | 20:22 |
| clayg | I think maybe the symantic might also require that instead of timeing out waiting on the 100 line and sending the body anyway - you have to close the connection and retry with expect 100 continue. | 20:22 |
| clayg | like - either tell me I can proceed or I'm not sending it - and if you say don't I won't - promise | 20:23 |
| clayg | Require: 100-continue - should totally be a thing | 20:23 |
| clayg | (everytime I tcpdump ssync I have these thoughts) | 20:23 |
| clayg | we need to get mark nottingham on this | 20:23 |
| clayg | or write Sweventlet so we can add SPDY support - or make up our own thing with json (CC/troll redbo) | 20:24 |
| jrichli | acoles: lol. | 20:24 |
| acoles | jrichli: heh, well you know notmyname is playing with us | 20:26 |
| notmyname | :-) | 20:28 |
| acoles | clayg: ok i'm curious - ssync & Expect 100 ?? is that a thing? | 20:29 |
| *** thumpba has joined #openstack-swift | 20:29 | |
| acoles | clayg: thansk for that review btw | 20:29 |
| *** CR7 has quit IRC | 20:30 | |
| clayg | acoles: 'a161102fba1710ef912af194b8d4635a dm' - rly? like you're totally schizophrenic in your "invent my own encoding" vs. "just use json" | 20:30 |
| *** CR7 has joined #openstack-swift | 20:30 | |
| clayg | I can't wait to go read this packing and unpacking parser | 20:31 |
| clayg | acoles: NO! | 20:31 |
| clayg | acoles: RE: ssync & expect 100 | 20:31 |
| acoles | clayg: omg i was just waiting for that :) had same thought myself when i looked over it today | 20:31 |
| clayg | acoles: but like the hummingbird replicator does it way better | 20:31 |
| acoles | clayg: phew, thought i had missed something there | 20:31 |
| clayg | acoles: instead of all the back and forth "within" the SSYNC request it just does pipelined SYNC requests (which are basically PUT's to a timestamp" | 20:32 |
| clayg | er... PUT to a hash | 20:32 |
| acoles | clayg: right. makes sense. | 20:32 |
| clayg | but it's cool about 409's and shit | 20:32 |
| acoles | yup | 20:32 |
| *** thumpba has quit IRC | 20:32 | |
| clayg | but it doesn't work with HTTP because you can't do pipelined requests like that :'( | 20:32 |
| clayg | so it opens the socket and does the REPLICATE, then it just start's sending this '<length>\r\n<json-version-of-http-headers>\r\n[WAIT FOR JSON OK]<body>\r\n<next-request-length>\r\n><json-headers>\r\n[WAIT FOR JSON 409]\r\n...' | 20:34 |
| clayg | unless redbo re-wrote it again | 20:34 |
| clayg | anyway - totally better than what we've got with SSYNC | 20:34 |
| acoles | hrm. reminds me, i wondered if we should make the ssync_sender tolerate response content between the updates_start and updates_end so that one day we can start to send back subrequest success/fail info without raising exceptions in legacy senders. | 20:35 |
| clayg | acoles: maybe - I think we should just re-work the protocol tho :\ | 20:35 |
| clayg | acoles: or go ahead and build something compatible with the hummingbird replciator (wouldn't be so bad neway) | 20:36 |
| acoles | so they send *json* ;) | 20:36 |
| clayg | heh | 20:36 |
| clayg | a) why are you still working b) what do you call the ssync missing check line encoding? | 20:36 |
| acoles | i think i just use encode/decode for anything that transforms stuff to a string or w/e. i should get a thesaurus. | 20:37 |
| clayg | a161102fba1710ef912af194b8d4635a 1443730105.90698 m:18a915 and the response a161102fba1710ef912af194b8d4635a dm (vs. just a161102fba1710ef912af194b8d4635a m) | 20:38 |
| acoles | ? | 20:39 |
| acoles | re a) i came back to switch off my laptop! | 20:39 |
| clayg | lol | 20:39 |
| *** breitz has quit IRC | 20:41 | |
| acoles | seriously, the more i think about it the more i feel we should really try to land 138498 even if 215276 is a stretch, cos if we have real clusters spreading mixed timestamps for frags around then 215276 will never work on those clusters. | 20:42 |
| clayg | acoles: yeah it's all good - we'll totally get it in | 20:42 |
| clayg | hash is a builtin name btw - so you've got at least one push-over coming | 20:43 |
| acoles | clayg: hey thanks for looking at it and, you know, just tell me what can be better, i'm all ears | 20:43 |
| acoles | heh. ok there's one already | 20:43 |
| clayg | acoles: i'll just push over - don't worry about it | 20:43 |
| acoles | yeah, i used object recently, and dir. :/ fixed those though | 20:44 |
| clayg | I'ma wait till I make it through tho | 20:44 |
| clayg | lol | 20:44 |
| acoles | sure | 20:44 |
| clayg | acoles: get some better syntax highlighting bro | 20:44 |
| acoles | surprised my IDE didn't yell at me | 20:44 |
| clayg | acoles: what editor do you use? | 20:44 |
| acoles | or maybe my old eyes just don't see the colors no more | 20:44 |
| acoles | pycharm | 20:44 |
| clayg | rofl | 20:44 |
| acoles | its all shades of grey | 20:45 |
| acoles | no, wait, bad choice | 20:45 |
| clayg | jeez | 20:45 |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Improve SLO PUT error checking https://review.openstack.org/228099 | 20:46 |
| *** bill_az_ has joined #openstack-swift | 20:46 | |
| *** thumpba has joined #openstack-swift | 20:47 | |
| clayg | acoles: so how come old ssync_reciver's don't blow up when the line has the extra "metadata delta" bit? | 20:47 |
| acoles | clayg: ok i need some sleep, forgive my blunders, change it all, particularly header names. re-author the commit to yourself and I'll +2A. | 20:47 |
| * acoles spreads the blame | 20:47 | |
| clayg | acoles: RIGHT ON! | 20:47 |
| clayg | none of us are as dumb as all of us! | 20:47 |
| notmyname | :-) | 20:47 |
| acoles | clayg: oh, remember crazy time when we landed EC, well i sneaked in a change to ssync_receiver to tolerate extra parts in the missing check phrases. | 20:48 |
| acoles | clayg: that was the best thing i did on EC! | 20:48 |
| acoles | i saw this coming | 20:48 |
| clayg | acoles: yeah I thought it was something like that - nice work | 20:48 |
| acoles | and i think i try to do same here, like look for what i expect but don't blow up if there is other stuff | 20:49 |
| clayg | object_hash, timestamp = [urllib.unquote(v) for v in parts[:2]] | 20:49 |
| clayg | ^ boom! | 20:49 |
| acoles | well some pieces can't change ;) | 20:50 |
| *** delatte has quit IRC | 20:50 | |
| *** cdelatte has quit IRC | 20:50 | |
| clayg | no no, that's the bit on master, it's great | 20:50 |
| clayg | go to bed or something - stop listening to me | 20:50 |
| acoles | k. got it. go away coles | 20:51 |
| acoles | clayg: ok. have a good one. | 20:52 |
| *** [1]eranrom has joined #openstack-swift | 20:52 | |
| *** eranrom has quit IRC | 20:52 | |
| *** [1]eranrom is now known as eranrom | 20:52 | |
| acoles | huh, i remembered why i came here: | 20:52 |
| acoles | mattoliverau: if you have chance could you take another look at https://review.openstack.org/#/c/227133/ it has one +2. thanks. hope your phone is on silent :D | 20:53 |
| *** acoles is now known as acoles_ | 20:54 | |
| *** ekarlso has quit IRC | 20:55 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Improve SLO PUT error checking https://review.openstack.org/228099 | 20:56 |
| *** ekarlso has joined #openstack-swift | 20:58 | |
| *** itlinux has quit IRC | 21:01 | |
| *** itlinux has joined #openstack-swift | 21:01 | |
| *** bill_az has quit IRC | 21:08 | |
| *** Fin1te has joined #openstack-swift | 21:13 | |
| *** chao_li_ has joined #openstack-swift | 21:13 | |
| *** dustins has quit IRC | 21:15 | |
| *** pberis has quit IRC | 21:18 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Fix slorange on-disk format when including whole object https://review.openstack.org/230152 | 21:19 |
| timburke | torgomatic: ^^^ | 21:19 |
| torgomatic | timburke: you should really put some words in there about how the current behavior is only wasteful, not harmful | 21:20 |
| torgomatic | it sounds way scarier than it is :| | 21:20 |
| timburke | good point | 21:20 |
| openstackgerrit | Tim Burke proposed openstack/swift: Fix slorange on-disk format when including whole object https://review.openstack.org/230152 | 21:24 |
| *** wbhuber has quit IRC | 21:25 | |
| notmyname | clayg: you're working on a push for patch 138498? if so, I found a typo that needs fixed | 21:25 |
| patchbot | notmyname: https://review.openstack.org/#/c/138498/ - Add POST capability to ssync for .meta files | 21:25 |
| *** trifon has quit IRC | 21:26 | |
| *** zhill has quit IRC | 21:28 | |
| *** chao_li_ has quit IRC | 21:30 | |
| *** chao_li has quit IRC | 21:30 | |
| clayg | notmyname: yup - just stick it in a comment and I'll add it | 21:34 |
| clayg | or push over and I'll rebase before I push fixes | 21:35 |
| notmyname | will do. still going through it. simple thing, so if you were about to push something I wanted to make sure you got it | 21:35 |
| *** jrichli has quit IRC | 21:35 | |
| notmyname | clayg: in test_diskfile.py, in test_get_metafile_not_opened(), the parens are wrong | 21:36 |
| clayg | notmyname: can that be true? sounds like a bs test if it's wrong and not failing... | 21:41 |
| *** MVenesio has quit IRC | 21:41 | |
| notmyname | I'm surprised "self.assertRaises(Exception)" doesn't fail since it only has the one argument. but assuming it doesn't, then the line just ends up being a tuple | 21:42 |
| notmyname | which isn't wrong, syntactically | 21:42 |
| notmyname | somethingsomething type systems | 21:42 |
| *** zhill has joined #openstack-swift | 21:43 | |
| notmyname | of course there are 2 forms of assertRaises. why wouldn't there be? https://docs.python.org/2/library/unittest.html#unittest.TestCase.assertRaises | 21:43 |
| *** david-lyle has quit IRC | 21:43 | |
| *** eranrom has quit IRC | 21:49 | |
| *** ccavanna has joined #openstack-swift | 21:51 | |
| *** Fin1te has quit IRC | 21:52 | |
| clayg | well *yeah* | 21:54 |
| clayg | notmyname: so is the test wrong or not? | 21:54 |
| notmyname | other than the parens? | 21:54 |
| notmyname | I think it's fine, just needs the right parens to actually check something | 21:54 |
| clayg | notmyname: you might need to send me a diff - did you make a comment on the patch? | 21:54 |
| notmyname | self.assertRaises(DiskFileNotOpen), df.get_metafile_metadata # current line | 21:55 |
| notmyname | self.assertRaises(DiskFileNotOpen, df.get_metafile_metadata) # new line | 21:55 |
| clayg | notmyname: this is patch 138498 | 21:55 |
| patchbot | clayg: https://review.openstack.org/#/c/138498/ - Add POST capability to ssync for .meta files | 21:55 |
| notmyname | yup | 21:55 |
| clayg | test.unit.obj.test_diskfile you say? | 21:56 |
| notmyname | yup | 21:56 |
| clayg | this is in the diff - or preexisting? | 21:56 |
| notmyname | test_get_metafile_metadata_not_opened | 21:56 |
| notmyname | it's a test added in this patch | 21:56 |
| *** willsama has quit IRC | 21:56 | |
| *** david-lyle has joined #openstack-swift | 21:56 | |
| clayg | OMG | 21:57 |
| clayg | fuck this new gerrit view | 21:57 |
| clayg | ctrl-F doesn't work | 21:57 |
| notmyname | lol, you haven't turned it off yet? | 21:57 |
| clayg | well I'm going to NOW | 21:57 |
| clayg | what fucking stpuid - why - I don't even | 21:58 |
| *** foexle has quit IRC | 22:01 | |
| clayg | i just can't - i try to read through gerrit's issues but there are just so many of them - i ... i'm full of hate | 22:02 |
| *** CaioBrentano has quit IRC | 22:06 | |
| *** CaioBrentano has joined #openstack-swift | 22:06 | |
| *** CaioBrentano has quit IRC | 22:07 | |
| *** CaioBrentano has joined #openstack-swift | 22:08 | |
| *** bill_az has joined #openstack-swift | 22:10 | |
| *** jlhinson has quit IRC | 22:11 | |
| clayg | torgomatic: you remember that one time we had that bug environ caches and we were going to fix everything but then didn't because simplified diffs - but then never went back and made any of the other good changes? | 22:16 |
| clayg | ... cause that was great | 22:16 |
| torgomatic | yeah, that would be nice to clean up at some point | 22:16 |
| clayg | torgomatic: do we have a lp bug or *anything* | 22:17 |
| torgomatic | eh, I didn't make one | 22:17 |
| clayg | sweet - me either! | 22:17 |
| clayg | notmyname: do you remember the container-tempurls security patch with like 86 comments? | 22:17 |
| notmyname | yeah | 22:17 |
| clayg | notmyname: in some of those comments was some great code that we opt'd not to merge with that fix - but we should totally caputre somewhere to go back and add all that | 22:18 |
| *** NM has quit IRC | 22:18 | |
| clayg | just thought of that because I'm looking at another change with risk of subtle bugs because of leaking identity on dicts :'( | 22:18 |
| notmyname | https://bugs.launchpad.net/swift/+bug/1449212 | 22:19 |
| openstack | Launchpad bug 1449212 in OpenStack Object Storage (swift) kilo "Container level temp URLs can unintentionally leak data." [Undecided,Fix committed] | 22:19 |
| *** lcurtis has quit IRC | 22:25 | |
| clayg | torgomatic: notmyname: lp bug #1501939 | 22:25 |
| openstack | Launchpad bug 1501939 in OpenStack Object Storage (swift) "get_info environ caching is leaky and brittle" [Undecided,New] https://launchpad.net/bugs/1501939 | 22:25 |
| notmyname | clayg: thanks | 22:26 |
| *** thumpba has quit IRC | 22:38 | |
| *** thumpba has joined #openstack-swift | 22:41 | |
| *** thumpba has quit IRC | 22:42 | |
| *** MVenesio has joined #openstack-swift | 22:49 | |
| *** chsc has quit IRC | 22:49 | |
| *** MVenesio has quit IRC | 22:53 | |
| *** zhill has quit IRC | 22:54 | |
| *** hrou has quit IRC | 22:55 | |
| *** alejandrito has quit IRC | 23:03 | |
| *** chlong has quit IRC | 23:07 | |
| *** km has joined #openstack-swift | 23:08 | |
| *** itlinux has quit IRC | 23:13 | |
| *** CR7 has quit IRC | 23:19 | |
| *** blmartin has quit IRC | 23:20 | |
| *** fbo has quit IRC | 23:22 | |
| *** kei_yama has joined #openstack-swift | 23:30 | |
| *** CR7 has joined #openstack-swift | 23:36 | |
| *** gyee has joined #openstack-swift | 23:39 | |
| *** minwoob has quit IRC | 23:48 | |
| *** wbhuber has joined #openstack-swift | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
