| *** bjkeller has joined #openstack-swift | 00:00 | |
| *** haomaiwa_ has joined #openstack-swift | 00:01 | |
| *** ho_away has quit IRC | 00:02 | |
| *** chlong_ has quit IRC | 00:03 | |
| *** RichardRaseley has joined #openstack-swift | 00:06 | |
| RichardRaseley | Hello all, I have a question regarding TempAuth. I have a small cluster I am using for a test and I intend to use TempAuth. Do those accounts *only* have to be defined on my one proxy node, or do the storage nodes also have to be made aware of the accounts? | 00:07 |
|---|---|---|
| *** dmorita has joined #openstack-swift | 00:08 | |
| *** haomaiwa_ has quit IRC | 00:09 | |
| clayg | mrmoje - sure why not - priority wishlist | 00:10 |
| clayg | RichardRaseley: just in the proxy server conf | 00:10 |
| RichardRaseley | clayg: Thank you. I just get the error message that the account doesn't exist. Could I validate the syntax with you? | 00:11 |
| clayg | RichardRaseley: where the passwords are in plain text | 00:11 |
| RichardRaseley | Yeah, that's fine. | 00:11 |
| clayg | RichardRaseley: you could do a paste - but the format/syntax is in the example configs - are you getting some error or something? | 00:11 |
| RichardRaseley | user_test_user = password .admin is what I have for a user, but I see 'Account not found' when using the client. | 00:12 |
| clayg | RichardRaseley: other thing with tempauth that's sort of a bummer is that there is no caonicial source to validate tokens with except memcache - so if you have some memcache issues it causes availability problems instead of just a lot of requests to auth your auth endpoints to validate incoming tokens | 00:12 |
| RichardRaseley | I seem to remember something else about tempauth and account creation, but I can't put my finger on it. | 00:12 |
| RichardRaseley | swift -A http://swift-test.domain.local:8080/auth/v1.0 -U test:user -K password stat | 00:13 |
| clayg | the allow account manage and account autocreate options effect all auth systems equally - tempauth included | 00:13 |
| clayg | yeah seems reasonable | 00:13 |
| RichardRaseley | OK, let me check those options. | 00:15 |
| *** mwheckmann has quit IRC | 00:15 | |
| RichardRaseley | OK, those both were false. Bouncing proxy. | 00:16 |
| RichardRaseley | clayg: That did it, thank you. | 00:16 |
| *** Jeffrey4l has joined #openstack-swift | 00:32 | |
| *** daemontool has quit IRC | 00:47 | |
| *** RichardRaseley has quit IRC | 00:57 | |
| openstackgerrit | Ben Keller proposed openstack/swift: Rename test_object_delete_at_aysnc https://review.openstack.org/284998 | 00:59 |
| kota_ | good morning | 01:00 |
| *** ho_away has joined #openstack-swift | 01:02 | |
| *** haomaiwa_ has joined #openstack-swift | 01:06 | |
| *** ho_away has quit IRC | 01:07 | |
| mattoliverau | kota_: morning | 01:07 |
| kota_ | mattoliverau: \o/ | 01:08 |
| kota_ | The last working day before Bristol Hackathon! | 01:09 |
| *** haomaiwa_ has quit IRC | 01:13 | |
| mattoliverau | kota_: rub it in why dont ya :P | 01:13 |
| kota_ | mattoliverau: that's because you get (got?) more happiness :P | 01:15 |
| mattoliverau | kota_: 'get' baby still isn't born yet. Thanks man, yeah that's true. | 01:16 |
| kota_ | mattoliverau: hope all safe :) | 01:18 |
| mattoliverau | kota_: so far so good :) | 01:19 |
| *** haomaiwang has joined #openstack-swift | 01:25 | |
| *** garthb has quit IRC | 01:28 | |
| *** garthb_ has quit IRC | 01:29 | |
| *** haomaiwang has quit IRC | 01:33 | |
| *** bjkeller has quit IRC | 01:34 | |
| *** chlong_ has joined #openstack-swift | 01:49 | |
| *** lpabon has quit IRC | 01:50 | |
| *** hezhiqiang has joined #openstack-swift | 01:54 | |
| *** gyee has quit IRC | 01:55 | |
| *** ho_away has joined #openstack-swift | 02:03 | |
| *** ho_away has quit IRC | 02:07 | |
| *** bjkeller has joined #openstack-swift | 02:09 | |
| *** diazjf has joined #openstack-swift | 02:10 | |
| *** diazjf has quit IRC | 02:13 | |
| *** mingdang1 has joined #openstack-swift | 02:16 | |
| *** haomaiwang has joined #openstack-swift | 02:21 | |
| *** bjkeller has quit IRC | 02:21 | |
| *** lpabon has joined #openstack-swift | 02:31 | |
| *** mwheckmann has joined #openstack-swift | 02:43 | |
| *** delatte has quit IRC | 02:56 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-swift | 03:01 | |
| *** ho_away has joined #openstack-swift | 03:04 | |
| *** lpabon has quit IRC | 03:09 | |
| *** ho_away has quit IRC | 03:09 | |
| *** links has joined #openstack-swift | 03:26 | |
| *** mingyu has quit IRC | 03:27 | |
| *** mingyu has joined #openstack-swift | 03:27 | |
| *** dmorita has quit IRC | 03:43 | |
| *** rickyrem has quit IRC | 03:43 | |
| *** mingdang1 has quit IRC | 03:50 | |
| *** mingyu has quit IRC | 03:55 | |
| *** mingyu has joined #openstack-swift | 03:55 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-swift | 04:01 | |
| *** mwheckmann has quit IRC | 04:05 | |
| *** ho_ has joined #openstack-swift | 04:05 | |
| *** mingyu has quit IRC | 04:10 | |
| *** mingyu has joined #openstack-swift | 04:11 | |
| *** hezhiqiang has quit IRC | 04:12 | |
| *** hezhiqiang has joined #openstack-swift | 04:13 | |
| *** dmorita has joined #openstack-swift | 04:17 | |
| *** dmorita has quit IRC | 04:22 | |
| *** hezhiqiang has quit IRC | 04:57 | |
| *** hezhiqiang has joined #openstack-swift | 04:58 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-swift | 05:01 | |
| *** ho_away has joined #openstack-swift | 05:01 | |
| mahatic | good morning | 05:02 |
| *** loth1 has joined #openstack-swift | 05:08 | |
| *** loth has quit IRC | 05:09 | |
| *** loth has joined #openstack-swift | 05:11 | |
| *** loth1 has quit IRC | 05:12 | |
| *** dmorita has joined #openstack-swift | 05:25 | |
| *** ppai has joined #openstack-swift | 05:30 | |
| *** klrmn has quit IRC | 05:30 | |
| *** garthb_ has joined #openstack-swift | 05:31 | |
| *** garthb has joined #openstack-swift | 05:31 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:32 | |
| *** mingyu has quit IRC | 05:32 | |
| *** mingyu has joined #openstack-swift | 05:33 | |
| *** ho_away has quit IRC | 05:37 | |
| *** asettle has joined #openstack-swift | 05:37 | |
| *** mingdang1 has joined #openstack-swift | 05:38 | |
| *** dmorita has quit IRC | 05:41 | |
| *** asettle has quit IRC | 05:42 | |
| *** mingdang1 has quit IRC | 05:46 | |
| *** diazjf has joined #openstack-swift | 05:58 | |
| *** ho_away has joined #openstack-swift | 06:00 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-swift | 06:01 | |
| *** mingdang1 has joined #openstack-swift | 06:01 | |
| *** loth1 has joined #openstack-swift | 06:09 | |
| *** ho_away has quit IRC | 06:10 | |
| openstackgerrit | Merged openstack/python-swiftclient: adding .manpages script to swift client https://review.openstack.org/284928 | 06:11 |
| *** loth has quit IRC | 06:11 | |
| *** loth has joined #openstack-swift | 06:13 | |
| *** loth1 has quit IRC | 06:14 | |
| *** loth1 has joined #openstack-swift | 06:17 | |
| *** loth has quit IRC | 06:18 | |
| *** loth has joined #openstack-swift | 06:20 | |
| openstackgerrit | Merged openstack/swift: Rename test_object_delete_at_aysnc https://review.openstack.org/284998 | 06:21 |
| *** loth1 has quit IRC | 06:21 | |
| *** links has quit IRC | 06:26 | |
| *** loth1 has joined #openstack-swift | 06:28 | |
| *** loth has quit IRC | 06:30 | |
| *** ChubYann has quit IRC | 06:34 | |
| *** loth has joined #openstack-swift | 06:34 | |
| *** loth1 has quit IRC | 06:35 | |
| *** loth1 has joined #openstack-swift | 06:48 | |
| *** loth has quit IRC | 06:49 | |
| *** haomaiwang has quit IRC | 06:52 | |
| *** _JZ_ has quit IRC | 06:53 | |
| *** haomaiwang has joined #openstack-swift | 06:53 | |
| *** loth has joined #openstack-swift | 06:55 | |
| *** loth1 has quit IRC | 06:57 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:01 | |
| *** loth1 has joined #openstack-swift | 07:01 | |
| *** loth has quit IRC | 07:02 | |
| *** rcernin has joined #openstack-swift | 07:05 | |
| *** zaitcev has quit IRC | 07:07 | |
| *** loth has joined #openstack-swift | 07:07 | |
| *** loth1 has quit IRC | 07:09 | |
| *** vinsh_ has joined #openstack-swift | 07:09 | |
| *** vinsh has quit IRC | 07:12 | |
| *** loth1 has joined #openstack-swift | 07:13 | |
| *** asettle has joined #openstack-swift | 07:14 | |
| *** loth has quit IRC | 07:14 | |
| *** asettle has quit IRC | 07:19 | |
| *** mrmoje has joined #openstack-swift | 07:19 | |
| *** garthb_ has quit IRC | 07:19 | |
| *** garthb has quit IRC | 07:19 | |
| *** loth has joined #openstack-swift | 07:21 | |
| *** loth1 has quit IRC | 07:23 | |
| *** loth1 has joined #openstack-swift | 07:24 | |
| *** loth has quit IRC | 07:25 | |
| *** chlong_ has quit IRC | 07:26 | |
| *** mrmoje has quit IRC | 07:28 | |
| *** loth has joined #openstack-swift | 07:30 | |
| mingdang1 | @acoles_ excuse me,I have a question,can you help me? | 07:30 |
| mingdang1 | In my swift cluster, it has 3 replication.If I modify one replica,how the swift-replicator fix it to correct? | 07:30 |
| *** daemontool has joined #openstack-swift | 07:31 | |
| *** loth1 has quit IRC | 07:31 | |
| *** _JZ_ has joined #openstack-swift | 07:32 | |
| *** AJaeger has joined #openstack-swift | 07:33 | |
| *** AJaeger has left #openstack-swift | 07:33 | |
| *** AJaeger has joined #openstack-swift | 07:33 | |
| AJaeger | Hi, is Samuel Merrit around? | 07:33 |
| *** diazjf has quit IRC | 07:33 | |
| AJaeger | his change https://review.openstack.org/#/c/212824/ has a python27 test that runs endlessly - gets killed by jenkins and then restarts on another node | 07:34 |
| patchbot | AJaeger: patch 212824 - swift - Let developers/operators add watchers to object audit | 07:34 |
| openstackgerrit | Andreas Jaeger proposed openstack/swift: Let developers/operators add watchers to object audit https://review.openstack.org/212824 | 07:37 |
| * AJaeger disabled the failing test now to give other jobs the chance to get nodes aswell. | 07:37 | |
| AJaeger | Running 11 hours of python27 - each 3 mins starting a new node - is enough | 07:37 |
| *** loth1 has joined #openstack-swift | 07:38 | |
| *** loth has quit IRC | 07:39 | |
| *** dmorita has joined #openstack-swift | 07:41 | |
| *** loth has joined #openstack-swift | 07:45 | |
| mingdang1 | @kota_ excuse me :) | 07:45 |
| *** dmorita has quit IRC | 07:46 | |
| *** loth1 has quit IRC | 07:47 | |
| *** trifon has joined #openstack-swift | 07:50 | |
| kota_ | mingdang1: what's up? | 07:52 |
| *** loth1 has joined #openstack-swift | 07:53 | |
| *** loth has quit IRC | 07:56 | |
| mingdang1 | @kota_ In my swift cluster, it has 3 replication.If I modify one replica,how the swift-replicator fix it to correct? | 07:58 |
| *** haomaiwang has quit IRC | 08:01 | |
| mingdang1 | I am confused at swift-replicator work | 08:01 |
| *** haomaiwang has joined #openstack-swift | 08:01 | |
| kota_ | basically if the object corrupted, object-auditor will find the difference between etag and actuall object and move it into qualantine. | 08:01 |
| kota_ | quarantine | 08:02 |
| kota_ | quarantine directory | 08:02 |
| *** loth has joined #openstack-swift | 08:02 | |
| mingdang1 | yeah ,yeah, I understand auditor | 08:02 |
| *** loth1 has quit IRC | 08:02 | |
| kota_ | and then, replicator will find the missing object on *remote* and then push the local replica into remote site. | 08:02 |
| mingdang1 | If local object is corrupted, how swift judge which is correct? | 08:04 |
| kota_ | each object has "etag" as xattr metadata calculated from md5sum of the whole content body. | 08:05 |
| mingdang1 | currently, I read the swift-object source code,it said swift get hashes.pkl from remote node,and compare them,then invoke rsync command push different object to remote. | 08:06 |
| mingdang1 | but I am confused,how it know who's hashes.pkl is correct? | 08:07 |
| kota_ | auditor will calculate the md5sum and if the calculateion result is different from the etag in xattr, it moves into quarantine | 08:07 |
| kota_ | oh, | 08:08 |
| kota_ | you know. | 08:08 |
| mingdang1 | yeah, auditor is in charge of move it to quarantine, | 08:08 |
| *** loth1 has joined #openstack-swift | 08:08 | |
| kota_ | they all don't know correct hashes.pkl | 08:08 |
| kota_ | just know we have difference between local and remote. | 08:08 |
| *** loth has quit IRC | 08:08 | |
| mingdang1 | In case of local replicator push currupted object to remote correct object | 08:10 |
| mingdang1 | whether it could happened | 08:10 |
| kota_ | mingdang1: I dont't think so. | 08:10 |
| kota_ | rsync protocol based on the file path, right? | 08:11 |
| mingdang1 | yes | 08:11 |
| kota_ | replicator will not push the corrupted object if remote has correct object. | 08:12 |
| kota_ | because, local and remote should have same objects (except corrupted body) and hash.pkl should be same. | 08:12 |
| mingdang1 | why? replicator judge whether it will push only depend on hashes.pkl | 08:14 |
| kota_ | if hash.pkl is missing at remote but replicator ask to remote object-server for current hashes. | 08:14 |
| mingdang1 | but now object is moved to quarantine directory? hashes.pkl will be calculated wrong? | 08:16 |
| kota_ | so clarify the states. | 08:17 |
| kota_ | for example | 08:17 |
| mingdang1 | oh | 08:17 |
| kota_ | node1 has correct object and node2 has corrupted | 08:18 |
| kota_ | not yet audititing | 08:18 |
| *** loth has joined #openstack-swift | 08:18 | |
| kota_ | node1 and node2 has same hash.pkl, right? | 08:18 |
| mingdang1 | maybe replicator will recalculate the hash.pkl | 08:19 |
| kota_ | that is because both node1 and node2 has objects in the same /dev/object/part/suffix/hash. | 08:19 |
| *** loth1 has quit IRC | 08:19 | |
| kota_ | not yet recaculate before auditing | 08:20 |
| *** loth1 has joined #openstack-swift | 08:21 | |
| mingdang1 | If before auditding the replicator deal with the currupted object,it don't recalculate? | 08:22 |
| kota_ | no? | 08:22 |
| mingdang1 | oh,oh ,it will not recalculate | 08:22 |
| *** loth has quit IRC | 08:22 | |
| mingdang1 | you are right | 08:23 |
| mingdang1 | I understand this step | 08:23 |
| kota_ | alright. | 08:23 |
| mingdang1 | go on please :) | 08:24 |
| kota_ | so thinking the state change by auditor | 08:25 |
| kota_ | the auditor in node2 moves the object into quarantine dir | 08:25 |
| kota_ | and invalidate_hash to hash.pkl | 08:26 |
| mingdang1 | yes | 08:26 |
| *** acoles_ is now known as acoles | 08:27 | |
| kota_ | in this state, node1 has correct object but node2 doesn't anything in the path and hash.pkl is invalidated. | 08:28 |
| *** loth has joined #openstack-swift | 08:28 | |
| acoles | good morning | 08:29 |
| kota_ | node1 replicator will make REPLICATE request to node2 to recalc the hash and node2 does from the path (actually nothing) | 08:29 |
| kota_ | acoles: morning | 08:29 |
| mingdang1 | @acoles morning | 08:29 |
| *** rledisez has joined #openstack-swift | 08:29 | |
| kota_ | so node1 can find node2 doesn't have the object and then do rysnc. | 08:29 |
| *** mmuffinman has quit IRC | 08:30 | |
| mingdang1 | yeah,it is right | 08:30 |
| *** loth1 has quit IRC | 08:31 | |
| mingdang1 | And I have another confused | 08:31 |
| kota_ | go on? | 08:32 |
| mingdang1 | If replica in node2 find remote hash.pkl is different with local, it will push local object to remote? | 08:33 |
| *** loth1 has joined #openstack-swift | 08:33 | |
| mingdang1 | swift can ensure object in node1 will go first? | 08:33 |
| kota_ | node2 can find the difference but nothing to push into node1 because the corrupted one is already in quarantine dir. | 08:34 |
| mingdang1 | bengo | 08:34 |
| mingdang1 | nice | 08:34 |
| mingdang1 | I understand | 08:34 |
| kota_ | great | 08:34 |
| mingdang1 | @kota_ thanks very much | 08:35 |
| *** daemontool has quit IRC | 08:35 | |
| *** loth has quit IRC | 08:35 | |
| kota_ | mingdang1: np :) | 08:35 |
| kota_ | acoles: I'm now thinking when I can get Bristol in Sunday evening. | 08:36 |
| *** daemontool has joined #openstack-swift | 08:36 | |
| kota_ | acoles: Did you have some plans for Sunday evening? | 08:37 |
| acoles | kota_: do you mean a plan to take you somewhere for dinner? | 08:38 |
| acoles | :) | 08:38 |
| kota_ | acoles: sure :) | 08:38 |
| acoles | what time do you think you will arrive at hotel? | 08:38 |
| kota_ | not sure, my flight will land LHW at 15:10 in the schedule. | 08:39 |
| *** loth has joined #openstack-swift | 08:39 | |
| acoles | ok I guess around 7pm | 08:39 |
| kota_ | I cannot expect how much time i have to expect to spend the time to pass the gate | 08:39 |
| kota_ | yup | 08:39 |
| kota_ | depends on which time I can get the coach | 08:40 |
| *** loth1 has quit IRC | 08:40 | |
| kota_ | 16:10 vs 17:10 | 08:40 |
| kota_ | maybe | 08:40 |
| kota_ | if I can get the first one, I can reach to Bristol around 18:00 | 08:41 |
| kota_ | but the second one will be after 1 hour from the first one :/ | 08:43 |
| *** loth1 has joined #openstack-swift | 08:44 | |
| *** m_kazuhiro has joined #openstack-swift | 08:45 | |
| mingdang1 | @kota_ excuse me again :),I hava another question | 08:45 |
| kota_ | mingdang1: yup | 08:45 |
| mingdang1 | when i get a object from swift cluster,how it decide which replica it will get | 08:46 |
| *** loth has quit IRC | 08:46 | |
| kota_ | no ways by client. It will be decided by proxy. | 08:48 |
| kota_ | basically, random among primary replicas. | 08:48 |
| *** loth has joined #openstack-swift | 08:48 | |
| *** loth1 has quit IRC | 08:48 | |
| mingdang1 | how it ensure one is correct | 08:49 |
| mingdang1 | R(1) + W(2) = N(3) | 08:49 |
| *** jordanP has joined #openstack-swift | 08:49 | |
| mingdang1 | it's weak consistency? | 08:50 |
| kota_ | sorry what does it mean 'R', 'W', 'N'? | 08:50 |
| mingdang1 | cap theory | 08:51 |
| mingdang1 | maybe I understand wrong | 08:51 |
| kota_ | basically, swift employ "write quorum, read one" for cap. | 08:51 |
| mingdang1 | 3 replica, quorum is 2 | 08:52 |
| mingdang1 | read one? | 08:52 |
| kota_ | (though, trying write all but assume quorum success is ok) | 08:52 |
| kota_ | yup | 08:52 |
| kota_ | read one. | 08:52 |
| mingdang1 | oh,it is consistency last | 08:53 |
| kota_ | read one of any replicas. | 08:53 |
| kota_ | doesn't resolve the consistency at the read except "X-Newest" | 08:54 |
| mingdang1 | yeah,the replicator will ensure the 3 replicator is same correct | 08:54 |
| kota_ | exactly | 08:54 |
| *** loth1 has joined #openstack-swift | 08:56 | |
| kota_ | mingdang1: does it make you sense? | 08:56 |
| mingdang1 | yeah,yeah, I understand, thanks | 08:56 |
| kota_ | :) | 08:57 |
| acoles | mingdang1: its "eventual consistency" - a read is not guaranteed to return the most recently written version of an object, (unless X-Newest is used, which has a cost) | 08:57 |
| *** loth has quit IRC | 08:58 | |
| mingdang1 | yeah. I really want to invite both you for dinner。You help me for my confused long time | 08:59 |
| *** asettle has joined #openstack-swift | 08:59 | |
| *** ho_ has quit IRC | 09:00 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwa_ has joined #openstack-swift | 09:01 | |
| *** takashi_ has quit IRC | 09:01 | |
| *** loth has joined #openstack-swift | 09:02 | |
| *** loth1 has quit IRC | 09:02 | |
| *** asettle has quit IRC | 09:04 | |
| *** loth1 has joined #openstack-swift | 09:06 | |
| *** loth has quit IRC | 09:07 | |
| *** loth has joined #openstack-swift | 09:12 | |
| *** loth1 has quit IRC | 09:13 | |
| *** hezhiqiang has quit IRC | 09:13 | |
| *** hezhiqiang has joined #openstack-swift | 09:13 | |
| *** rcernin has quit IRC | 09:14 | |
| *** loth1 has joined #openstack-swift | 09:17 | |
| *** loth has quit IRC | 09:18 | |
| *** loth has joined #openstack-swift | 09:24 | |
| *** loth1 has quit IRC | 09:25 | |
| *** loth1 has joined #openstack-swift | 09:27 | |
| *** jistr has joined #openstack-swift | 09:27 | |
| *** loth has quit IRC | 09:28 | |
| *** rcernin has joined #openstack-swift | 09:28 | |
| *** _JZ__ has joined #openstack-swift | 09:32 | |
| *** jordanP has quit IRC | 09:32 | |
| *** loth has joined #openstack-swift | 09:33 | |
| *** then3rd_1 has joined #openstack-swift | 09:34 | |
| *** then3rd has quit IRC | 09:35 | |
| *** _JZ_ has quit IRC | 09:35 | |
| *** loth1 has quit IRC | 09:35 | |
| *** loth1 has joined #openstack-swift | 09:37 | |
| *** loth has quit IRC | 09:38 | |
| *** asettle has joined #openstack-swift | 09:40 | |
| *** joeljwright has joined #openstack-swift | 09:41 | |
| *** ChanServ sets mode: +v joeljwright | 09:41 | |
| *** loth has joined #openstack-swift | 09:41 | |
| *** loth1 has quit IRC | 09:42 | |
| acoles | joeljwright: you about? | 09:42 |
| joeljwright | acoles: I am now | 09:43 |
| acoles | joeljwright: looking at https://review.openstack.org/#/c/284822/2/tests/unit/test_swiftclient.py line 2208 | 09:43 |
| patchbot | acoles: patch 284822 - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 09:43 |
| *** mmcardle has joined #openstack-swift | 09:43 | |
| acoles | joeljwright: you don't encode the unicode token there? | 09:43 |
| joeljwright | damn it! | 09:44 |
| *** loth1 has joined #openstack-swift | 09:44 | |
| acoles | oh, so that wasn't deliberate :D | 09:44 |
| joeljwright | oh, hang on | 09:44 |
| joeljwright | I think that's okay | 09:44 |
| joeljwright | no, I'm wrong | 09:45 |
| joeljwright | that's the bit I should have encoded | 09:45 |
| acoles | if you're confused then I have no hope | 09:45 |
| joeljwright | acoles: thanks for pointing it out | 09:45 |
| acoles | i just thought it odd that the etag is bytes and the token not | 09:46 |
| joeljwright | acoles: just a bit confused because I was fixing another bug last night until 1am, so my brain's still a bit fuzzy | 09:46 |
| *** loth has quit IRC | 09:46 | |
| *** jordanP has joined #openstack-swift | 09:46 | |
| acoles | i bet! np | 09:47 |
| *** Jeffrey4l_ has joined #openstack-swift | 09:47 | |
| joeljwright | acoles: I think most of the things in those dicts should be encoded | 09:48 |
| joeljwright | they should all the byte strings at that point | 09:48 |
| acoles | makes sense | 09:48 |
| *** Jeffrey4l has quit IRC | 09:49 | |
| *** loth has joined #openstack-swift | 09:50 | |
| *** davideagle_ has joined #openstack-swift | 09:50 | |
| *** loth1 has quit IRC | 09:52 | |
| *** asettle has quit IRC | 09:52 | |
| *** asettle has joined #openstack-swift | 09:53 | |
| *** loth1 has joined #openstack-swift | 09:55 | |
| *** loth has quit IRC | 09:56 | |
| *** loth has joined #openstack-swift | 09:59 | |
| *** asettle has quit IRC | 10:00 | |
| *** kei_yama has quit IRC | 10:00 | |
| *** asettle has joined #openstack-swift | 10:00 | |
| *** loth1 has quit IRC | 10:00 | |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| *** loth1 has joined #openstack-swift | 10:02 | |
| *** stantonnet has quit IRC | 10:03 | |
| *** loth has quit IRC | 10:03 | |
| *** acoles is now known as acoles_ | 10:05 | |
| *** stantonnet has joined #openstack-swift | 10:06 | |
| *** admin0_ has joined #openstack-swift | 10:07 | |
| *** loth has joined #openstack-swift | 10:08 | |
| admin0_ | like amazon says in their object store 1 in 1 billion object can disappear, is this the same in openstack swift ? | 10:08 |
| *** loth1 has quit IRC | 10:09 | |
| *** Raymii has joined #openstack-swift | 10:10 | |
| *** loth1 has joined #openstack-swift | 10:13 | |
| *** loth has quit IRC | 10:14 | |
| *** daemontool has quit IRC | 10:16 | |
| *** mrmoje has joined #openstack-swift | 10:16 | |
| *** hezhiqiang has quit IRC | 10:19 | |
| *** loth has joined #openstack-swift | 10:21 | |
| *** loth1 has quit IRC | 10:22 | |
| *** haomaiwang has quit IRC | 10:23 | |
| *** m_kazuhiro has quit IRC | 10:28 | |
| *** haomaiwang has joined #openstack-swift | 10:29 | |
| ahale | it depends on how you run it, its certainly possible. things like letting disks sit at 100%, hardware failure, other unknown bad things.. they can cause situations where data could be lost I guess | 10:32 |
| *** davideagle_ has quit IRC | 10:34 | |
| *** admin0_ has quit IRC | 10:40 | |
| *** OutBackDingo has quit IRC | 10:43 | |
| *** loth1 has joined #openstack-swift | 10:46 | |
| *** davideagle_ has joined #openstack-swift | 10:47 | |
| *** loth has quit IRC | 10:47 | |
| *** mrmoje has quit IRC | 10:50 | |
| *** felixcantournet has joined #openstack-swift | 10:54 | |
| openstackgerrit | Merged openstack/python-swiftclient: Follow-up to patch 282363 https://review.openstack.org/284490 | 10:56 |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| *** mingdang1 has quit IRC | 11:03 | |
| *** haomaiwang has quit IRC | 11:14 | |
| *** admin6 has joined #openstack-swift | 11:19 | |
| *** SkyRocknRoll has quit IRC | 11:22 | |
| *** mingyu has quit IRC | 11:23 | |
| *** mingyu has joined #openstack-swift | 11:24 | |
| *** mingyu has quit IRC | 11:27 | |
| *** mingyu has joined #openstack-swift | 11:28 | |
| *** mingyu has quit IRC | 11:32 | |
| *** mingyu has joined #openstack-swift | 11:32 | |
| *** mingyu has quit IRC | 11:36 | |
| *** mingyu has joined #openstack-swift | 11:37 | |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Fix test for redacting sensitive data in client.http_log() https://review.openstack.org/284822 | 11:38 |
| *** SkyRocknRoll has joined #openstack-swift | 11:38 | |
| *** davideagle_ has quit IRC | 11:48 | |
| *** davideagle_ has joined #openstack-swift | 11:51 | |
| *** asettle has quit IRC | 12:02 | |
| *** hezhiqiang has joined #openstack-swift | 12:06 | |
| *** haomaiwang has joined #openstack-swift | 12:14 | |
| *** daemontool has joined #openstack-swift | 12:14 | |
| *** davideagle_ has quit IRC | 12:18 | |
| *** cdelatte has joined #openstack-swift | 12:18 | |
| *** haomaiwang has quit IRC | 12:19 | |
| *** mingdang1 has joined #openstack-swift | 12:19 | |
| *** asettle has joined #openstack-swift | 12:24 | |
| *** mingyu has quit IRC | 12:24 | |
| *** mingyu has joined #openstack-swift | 12:24 | |
| *** mingyu has quit IRC | 12:29 | |
| *** mingyu has joined #openstack-swift | 12:29 | |
| *** davideagle_ has joined #openstack-swift | 12:34 | |
| *** jordanP has quit IRC | 12:35 | |
| *** jordanP has joined #openstack-swift | 12:36 | |
| *** hezhiqiang has quit IRC | 12:56 | |
| *** hezhiqiang has joined #openstack-swift | 12:57 | |
| *** mingyu has quit IRC | 12:58 | |
| *** mingyu has joined #openstack-swift | 12:58 | |
| *** SkyRocknRoll has quit IRC | 13:04 | |
| *** km has quit IRC | 13:08 | |
| *** mmcardle has quit IRC | 13:10 | |
| *** mingyu has quit IRC | 13:14 | |
| *** mingyu has joined #openstack-swift | 13:14 | |
| *** daemontool has quit IRC | 13:15 | |
| *** daemontool has joined #openstack-swift | 13:16 | |
| *** daemontool has quit IRC | 13:17 | |
| *** daemontool has joined #openstack-swift | 13:18 | |
| *** mingyu has quit IRC | 13:18 | |
| *** mingyu has joined #openstack-swift | 13:18 | |
| *** daemontool has quit IRC | 13:24 | |
| *** admin6 has left #openstack-swift | 13:26 | |
| *** mmcardle has joined #openstack-swift | 13:27 | |
| *** acoles_ is now known as acoles | 13:37 | |
| *** haomaiwang has joined #openstack-swift | 13:38 | |
| *** AndreiaKumpera has joined #openstack-swift | 13:48 | |
| *** ppai has quit IRC | 13:54 | |
| *** daemontool has joined #openstack-swift | 13:54 | |
| *** haomaiwang has quit IRC | 14:01 | |
| *** haomaiwang has joined #openstack-swift | 14:01 | |
| *** vinsh_ has quit IRC | 14:08 | |
| *** petertr7_away is now known as petertr7 | 14:15 | |
| *** daemontool has quit IRC | 14:16 | |
| *** daemontool has joined #openstack-swift | 14:16 | |
| *** vinsh has joined #openstack-swift | 14:17 | |
| *** daemontool_ has joined #openstack-swift | 14:18 | |
| tdasilva | good morning | 14:21 |
| *** daemontool has quit IRC | 14:22 | |
| *** chsc has joined #openstack-swift | 14:34 | |
| *** mingyu has quit IRC | 14:35 | |
| *** mingyu has joined #openstack-swift | 14:35 | |
| *** davideagle_ has quit IRC | 14:36 | |
| *** mingyu has quit IRC | 14:36 | |
| *** mingyu has joined #openstack-swift | 14:36 | |
| *** davideagle_ has joined #openstack-swift | 14:44 | |
| *** mrmoje has joined #openstack-swift | 14:44 | |
| *** mingyu has quit IRC | 14:50 | |
| *** mingyu has joined #openstack-swift | 14:50 | |
| *** mingdang1 has quit IRC | 14:50 | |
| *** mrmoje has quit IRC | 14:51 | |
| acoles | aargh gerrit! | 14:51 |
| *** mrmoje has joined #openstack-swift | 14:55 | |
| acoles | notmyname: see my comment on patch 284645 - I'm thinking it should maybe have a different change id. What's the deal on reviews sharing same change id? | 14:56 |
| patchbot | acoles: https://review.openstack.org/#/c/284645/ - python-swiftclient (stable/liberty) - Do not reveal auth token in swiftclient log messag... | 14:56 |
| tdasilva | lol | 14:56 |
| acoles | tdasilva: what's the *longest* comment you have typed into gerrit only for it to decide you are no longer signed in and give you no opportunity to copy your comment before signing in? | 14:57 |
| acoles | joeljwright: thanks for sorting out the header redaction backport | 14:58 |
| *** haomaiwang has quit IRC | 15:01 | |
| *** daemontool_ has quit IRC | 15:01 | |
| *** haomaiwang has joined #openstack-swift | 15:01 | |
| tdasilva | acoles: hehe..gerrit can be maddening. I've noticed a bunch of little changes they've made recently that haven't really made our jobs easier | 15:01 |
| *** mrmoje has quit IRC | 15:04 | |
| *** delattec has joined #openstack-swift | 15:04 | |
| *** daemontool has joined #openstack-swift | 15:05 | |
| *** cdelatte has quit IRC | 15:06 | |
| *** vinsh has quit IRC | 15:08 | |
| *** mrmoje has joined #openstack-swift | 15:09 | |
| *** trifon has quit IRC | 15:09 | |
| *** arch-nemesis has joined #openstack-swift | 15:09 | |
| *** vinsh has joined #openstack-swift | 15:14 | |
| *** asettle has quit IRC | 15:16 | |
| *** mrmoje has quit IRC | 15:16 | |
| acoles | oh it did it again, so I clicked "Go Anonymous" which gave me chance to cut'n'paste my comment before signing in. :) | 15:18 |
| *** delatte has joined #openstack-swift | 15:28 | |
| *** delattec has quit IRC | 15:30 | |
| *** davideagle_ has quit IRC | 15:31 | |
| *** davideagle_ has joined #openstack-swift | 15:33 | |
| *** bjkeller has joined #openstack-swift | 15:34 | |
| *** asettle has joined #openstack-swift | 15:34 | |
| *** lcurtis has joined #openstack-swift | 15:39 | |
| *** _JZ__ has quit IRC | 15:41 | |
| *** _JZ_ has joined #openstack-swift | 15:42 | |
| *** diazjf has joined #openstack-swift | 15:49 | |
| *** jmccarthy has quit IRC | 15:51 | |
| *** jmccarthy has joined #openstack-swift | 15:52 | |
| *** diazjf has left #openstack-swift | 15:54 | |
| *** asettle has quit IRC | 15:56 | |
| *** mwheckmann has joined #openstack-swift | 15:57 | |
| *** arch-nemesis has quit IRC | 16:00 | |
| *** asettle has joined #openstack-swift | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-swift | 16:01 | |
| jaakkos | is teh Swift COPY operation O(1) in terms of object size? | 16:03 |
| jaakkos | ie. it just adds a new name and maybe increases reference count? | 16:04 |
| notmyname | good morning | 16:04 |
| notmyname | torgomatic: see the ping from AJaeger earlier about your patch | 16:05 |
| notmyname | acoles: I don't think the same change-id matters? | 16:05 |
| *** davideagle_ has quit IRC | 16:06 | |
| AJaeger | morning, notmyname and thanks for the ping to torgomatic ! | 16:06 |
| tdasilva | jaakkos: copy will actually copy the contents of the object to a new object, most likely to new storage nodes as the name of the object changes | 16:07 |
| notmyname | torgomatic: on patch 212824 | 16:07 |
| patchbot | notmyname: https://review.openstack.org/#/c/212824/ - swift - Let developers/operators add watchers to object audit | 16:07 |
| *** daemontool has quit IRC | 16:07 | |
| AJaeger | torgomatic: I'm still around if you have a question. Please revert my change but let's figure out why Jenkins was not happy with it. | 16:08 |
| jaakkos | tdasilva: alright. so, i suppose it's not possible to rename an object without copying data | 16:08 |
| tdasilva | jaakkos: correct, but of course this copying happens on the server side. user doesn't have to download data and upload again | 16:09 |
| tdasilva | jaakkos: just in case that wasn't clear | 16:09 |
| jaakkos | thanks for the clarification | 16:11 |
| tdasilva | jaakkos: and a "move" would be COPY+DELETE | 16:12 |
| *** logan- has quit IRC | 16:14 | |
| *** logan- has joined #openstack-swift | 16:14 | |
| notmyname | AJaeger: FWIW, torgomatic's am schedule usually doesn't start for another 60-90 minutes | 16:20 |
| *** _JZ_ has quit IRC | 16:21 | |
| AJaeger | notmyname: thanks. I don't need to discuss with him, just wanted him to be aware of the problem | 16:21 |
| AJaeger | but if torgomatic has questions, I'm happy to discuss;) | 16:22 |
| notmyname | thanks for bringing it up. looks like a really weird issue | 16:22 |
| AJaeger | yep ;( | 16:22 |
| *** klrmn has joined #openstack-swift | 16:25 | |
| *** garthb has joined #openstack-swift | 16:28 | |
| *** garthb_ has joined #openstack-swift | 16:28 | |
| *** rcernin has quit IRC | 16:31 | |
| *** garthb has quit IRC | 16:36 | |
| *** garthb_ has quit IRC | 16:36 | |
| *** daemontool has joined #openstack-swift | 16:40 | |
| *** nadeem has joined #openstack-swift | 16:43 | |
| openstackgerrit | John Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release https://review.openstack.org/281974 | 16:43 |
| notmyname | acoles: joeljwright: ^ best changelog yet | 16:43 |
| joeljwright | :) | 16:43 |
| *** nadeem has quit IRC | 16:43 | |
| *** nadeem has joined #openstack-swift | 16:44 | |
| *** garthb has joined #openstack-swift | 16:44 | |
| notmyname | if you could take a look before you stop for the day, I'll bug timburke about it later, and I think we'll be able to put in the request for a release today | 16:44 |
| *** gyee has joined #openstack-swift | 16:44 | |
| joeljwright | sure, will look now | 16:44 |
| acoles | notmyname: k. are you planning to wait for patch 284822? | 16:46 |
| patchbot | acoles: https://review.openstack.org/#/c/284822/ - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 16:46 |
| *** trifon has joined #openstack-swift | 16:47 | |
| notmyname | acoles: no, I wouldn't block a release on a test-only update. but that one should still land :-) | 16:48 |
| *** davideagle_ has joined #openstack-swift | 16:51 | |
| *** davideagle_ has quit IRC | 16:55 | |
| notmyname | acoles: joeljwright: I get that encoding to utf8 seems like the right thing to do, but it's a little weird that tests pass with and without the change | 16:55 |
| notmyname | for patch 284822 | 16:56 |
| patchbot | notmyname: https://review.openstack.org/#/c/284822/ - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 16:56 |
| acoles | notmyname: think its cos parse_header_string does a good job of handling any type of value and morphing it to unicode | 16:56 |
| joeljwright | notmyname: I think the tests pass without the change because the parse_headers() function is so defensive | 16:57 |
| acoles | joeljwright and i think the same :) | 16:57 |
| notmyname | yeah, so it's a test for if parse_headers() changes and misses some unicode thing | 16:57 |
| joeljwright | yes | 16:57 |
| acoles | notmyname: the wrangling was really to make what is passed in by the test realistic | 16:58 |
| acoles | notmyname: but joeljwright and I are also debating timburke's suggestion to include some non-utf-8 binary in the test value | 16:59 |
| joeljwright | notmyname: it's been one of those fun unicode days again :) | 16:59 |
| acoles | funicode day :) | 16:59 |
| notmyname | heh | 17:00 |
| acoles | notmyname: joel found this in api docs, "The metadata values in Object Storage do not follow HTTP/1.1 rules for character encodings. You must use a UTF-8 encoding to get a byte array for any string that contains characters that are not in the 7-bit ASCII 0-127 range. Otherwise, Object Storage returns the 404 response code for ISO-8859-1 characters in the 128-255 range, which is a direct violation of the HTTP/1.1 http://ww | 17:00 |
| acoles | w.w3.org/Protocols/rfc2616/rfc2616-sec2.html#sec2. | 17:00 |
| notmyname | so where are you now on that non utf8 unicode question? | 17:00 |
| notmyname | yeah | 17:00 |
| *** haomaiwang has quit IRC | 17:01 | |
| acoles | but I am trying to see where that is enforced ie where swift checks for utf8 in header values, so far i see a check on content-type | 17:01 |
| *** haomaiwang has joined #openstack-swift | 17:01 | |
| timburke | yeah, the 404 bit seems...weird. that seems much more likely to be talking about the actual *path*, not metadata | 17:02 |
| acoles | yeah the path is definitely checked | 17:03 |
| timburke | fwiw, i just got a 503 when trying to post non-utf8 data...digging into why | 17:03 |
| openstackgerrit | Bing Hu proposed openstack/python-swiftclient: make download --prefix working correct https://review.openstack.org/284851 | 17:03 |
| *** jordanP has quit IRC | 17:06 | |
| *** asettle has quit IRC | 17:08 | |
| timburke | here: https://github.com/openstack/swift/blob/2.6.0/swift/common/db.py#L793 | 17:08 |
| timburke | json.dumps will raise a UnicodeDecodeError | 17:09 |
| notmyname | tdasilva: wow. your first commit to swiftclient just landed | 17:09 |
| tdasilva | yay! :P | 17:09 |
| openstackgerrit | John Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release https://review.openstack.org/281974 | 17:10 |
| tdasilva | and it was just a file copy | 17:10 |
| tdasilva | :D | 17:10 |
| *** mwheckma1n has joined #openstack-swift | 17:10 | |
| *** rledisez has quit IRC | 17:11 | |
| *** wbhuber_ has joined #openstack-swift | 17:12 | |
| *** hezhiqia_ has joined #openstack-swift | 17:12 | |
| *** taneee has joined #openstack-swift | 17:12 | |
| *** swifterdarrell_ has joined #openstack-swift | 17:13 | |
| *** mingyu_ has joined #openstack-swift | 17:13 | |
| tdasilva | so, who is running this project: http://openqa.sed.hu/dashboard/index?did=1&id=OpenStack%3Aswift | 17:13 |
| acoles | timburke: ok so metadata to account/container db needs to be utf8. i just tried writing metadata to diskfile and it seemed happy to take \xff | 17:14 |
| joeljwright | looks like we need a 'you can only post utf8' middleware :) | 17:15 |
| *** gmmaharaj has joined #openstack-swift | 17:15 | |
| *** j_king_ has joined #openstack-swift | 17:15 | |
| acoles | there could be some place between proxy and diskfile that barfs on non utf8 | 17:15 |
| *** tonyb_ has joined #openstack-swift | 17:16 | |
| *** jeblair_ has joined #openstack-swift | 17:17 | |
| *** schizo has joined #openstack-swift | 17:17 | |
| *** schizo has joined #openstack-swift | 17:17 | |
| *** csmart_ has joined #openstack-swift | 17:17 | |
| *** coreycb` has joined #openstack-swift | 17:18 | |
| acoles | the bonus of doing a backport is you get to hit all the intermittent test failures that we fixed | 17:18 |
| * acoles types recheck on patch 284645 | 17:19 | |
| patchbot | acoles: https://review.openstack.org/#/c/284645/ - python-swiftclient (stable/liberty) - Do not reveal auth token in swiftclient log messag... | 17:19 |
| *** klrmn has quit IRC | 17:19 | |
| *** mlanner_ has joined #openstack-swift | 17:20 | |
| *** mwheckmann has quit IRC | 17:20 | |
| *** mingyu has quit IRC | 17:20 | |
| *** hezhiqiang has quit IRC | 17:20 | |
| *** then3rd_1 has quit IRC | 17:20 | |
| *** proteusguy__ has quit IRC | 17:20 | |
| *** zul has quit IRC | 17:20 | |
| *** ekarlso- has quit IRC | 17:20 | |
| *** gmmaha has quit IRC | 17:20 | |
| *** McMurlock has quit IRC | 17:20 | |
| *** notmyname has quit IRC | 17:20 | |
| *** _fortis has quit IRC | 17:20 | |
| *** onovy has quit IRC | 17:20 | |
| *** kota_ has quit IRC | 17:20 | |
| *** j_king has quit IRC | 17:20 | |
| *** csmart has quit IRC | 17:20 | |
| *** tonyb has quit IRC | 17:20 | |
| *** hrou has quit IRC | 17:20 | |
| *** thurloat has quit IRC | 17:20 | |
| *** BAKfr has quit IRC | 17:20 | |
| *** pdardeau has quit IRC | 17:20 | |
| *** fbo_ has quit IRC | 17:20 | |
| *** coreycb has quit IRC | 17:20 | |
| *** tanee has quit IRC | 17:20 | |
| *** jeblair has quit IRC | 17:20 | |
| *** swifterdarrell has quit IRC | 17:20 | |
| *** timburke has quit IRC | 17:20 | |
| *** mlanner has quit IRC | 17:20 | |
| *** wbhuber has quit IRC | 17:20 | |
| *** bapalm has quit IRC | 17:20 | |
| *** bwall has quit IRC | 17:20 | |
| *** gmmaharaj is now known as gmmaha | 17:20 | |
| *** swifterdarrell_ is now known as swifterdarrell | 17:20 | |
| *** mlanner_ is now known as mlanner | 17:20 | |
| *** onovy_ has joined #openstack-swift | 17:20 | |
| *** bapalm has joined #openstack-swift | 17:21 | |
| *** zhill has joined #openstack-swift | 17:21 | |
| *** onovy_ is now known as onovy | 17:21 | |
| *** notmyname has joined #openstack-swift | 17:21 | |
| *** ChanServ sets mode: +v notmyname | 17:21 | |
| *** thurloat has joined #openstack-swift | 17:21 | |
| *** BAKfr has joined #openstack-swift | 17:21 | |
| *** fbo has joined #openstack-swift | 17:21 | |
| joeljwright | there are still a few intermittent race conditions on the swiftclient tests | 17:21 |
| *** hrou has joined #openstack-swift | 17:22 | |
| joeljwright | I've been thinking about how to solve them, but not come up with anything concrete yet | 17:22 |
| joeljwright | and they are quite rare to hit | 17:22 |
| Zyric_ | Good Morning | 17:23 |
| *** timburke has joined #openstack-swift | 17:23 | |
| *** ChanServ sets mode: +v timburke | 17:23 | |
| *** mmuffinman has joined #openstack-swift | 17:24 | |
| acoles | notmyname: did you change your mind about co-authors in AUTHORS? | 17:25 |
| notmyname | I did | 17:25 |
| notmyname | I was telling joeljwright that I keep going back and forth | 17:25 |
| acoles | notmyname: you want a few mins before I +A? :) | 17:25 |
| notmyname | nope. land it :-) | 17:26 |
| acoles | timburke: you want a final look at patch 281974 or shall I land it? | 17:27 |
| patchbot | acoles: https://review.openstack.org/#/c/281974/ - python-swiftclient - authors/changelog updates for 2.8 release | 17:27 |
| *** proteusguy__ has joined #openstack-swift | 17:27 | |
| *** ekarlso- has joined #openstack-swift | 17:27 | |
| *** zul has joined #openstack-swift | 17:27 | |
| notmyname | (1) I don't think my own CLA enforcement argument is very strong. (2) we've got a known person in the community vouching for it. (3) better to credit someone and maybe necourage them than to get into a debate about when someone is truly considered and author | 17:27 |
| timburke | go for it. i was about to +A it myself but also wanted to ask notmyname about AUTHORS | 17:27 |
| notmyname | well now authors has tdasilva so it's all good :-) | 17:27 |
| *** then3rd_1 has joined #openstack-swift | 17:28 | |
| *** McMurlock has joined #openstack-swift | 17:28 | |
| notmyname | if https://review.openstack.org/#/c/284822/ lands shortly, then we can include it in the release | 17:29 |
| patchbot | notmyname: patch 284822 - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 17:29 |
| *** wasmum has joined #openstack-swift | 17:30 | |
| joeljwright | c'mon 284822, you can make it! | 17:30 |
| notmyname | timburke: oh, a 503 response on a unicode error seems like a serious bug in swift. we shouldn't ever return 503 | 17:30 |
| timburke | oh yeah, acoles -- that makes sense about the diskfile metadata. i had this recollection that i could POST non-utf8 data but couldn't remember the exact circumstances | 17:31 |
| acoles | notmyname: it needs a second +2 first | 17:31 |
| *** _fortis_ has joined #openstack-swift | 17:31 | |
| notmyname | acoles: yeah, I was hoping timburke would swoop in and save the day | 17:31 |
| timburke | notmyname: agreed. we need to clean that up a bit | 17:31 |
| acoles | timburke: can you repeat your 503 generating POST but to an object? | 17:32 |
| timburke | acoles: confirmed. POST -> 202; subsequent HEAD -> 200 with my funky data | 17:34 |
| timburke | ...which seems to be an argument in favor of including some non-utf8 data to redact in patch 284822 | 17:35 |
| patchbot | timburke: https://review.openstack.org/#/c/284822/ - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 17:35 |
| *** _fortis_ is now known as _fortis | 17:35 | |
| timburke | although, maybe that'd be better as yet-another-follow-up. the docstring fix *would* be nice to get in ahead of release | 17:36 |
| joeljwright | yeah, the docstring fix would be nice | 17:37 |
| *** schizo is now known as bwall | 17:38 | |
| notmyname | so push in the non-utf8 data. who's pushign that? joeljwright, timburke, me? | 17:38 |
| joeljwright | I can add the test changes, but the current method for collecting the output won;t be happy | 17:39 |
| joeljwright | u''.join() | 17:39 |
| joeljwright | :( | 17:39 |
| timburke | nah, it'll be fine. i trust that parse_header_string will *always* produce unicode | 17:40 |
| timburke | it should turn b'\xff' into u'%ff' (or maybe u'%FF') | 17:40 |
| acoles | timburke: thanks. nice to have it confirmed. | 17:41 |
| joeljwright | yeah, I think you're right :) | 17:42 |
| acoles | so is someone pushing a new version of 284822? | 17:45 |
| *** zhill has quit IRC | 17:46 | |
| *** silor has joined #openstack-swift | 17:46 | |
| joeljwright | 284822 is in the process of being merged | 17:47 |
| notmyname | patch 284822 | 17:48 |
| patchbot | notmyname: https://review.openstack.org/#/c/284822/ - python-swiftclient - Fix test for redacting sensitive data in client.ht... | 17:48 |
| joeljwright | will be a new patch for the binary test data | 17:48 |
| notmyname | ok | 17:48 |
| acoles | oic. cool | 17:48 |
| *** jistr has quit IRC | 17:49 | |
| *** davideagle_ has joined #openstack-swift | 17:50 | |
| *** mmcardle has quit IRC | 17:51 | |
| *** lpabon has joined #openstack-swift | 17:59 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** lpabon has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-swift | 18:01 | |
| *** ChubYann has joined #openstack-swift | 18:05 | |
| *** permalac has joined #openstack-swift | 18:06 | |
| permalac | hello guys. swift newbie here. | 18:06 |
| jaakkos | I suppose it's not possible to remove a certain version of an object - only the latest? | 18:06 |
| permalac | I installed a two nodes swift with kilo , only one controller as swift proxy-server , and then I installed an HA 2 controller nodes new installation with other machines | 18:07 |
| *** klrmn has joined #openstack-swift | 18:08 | |
| permalac | I would like to move the rings .gz to the new openstack installation ( the HA one) | 18:08 |
| permalac | can I just move the .gz files ? | 18:08 |
| permalac | can I create again the rings on the new installation and they will cohexist with the old one? | 18:09 |
| *** petertr7 is now known as petertr7_away | 18:11 | |
| *** RichardRaseley has joined #openstack-swift | 18:11 | |
| RichardRaseley | Hello all. Can anyone comment on their experience mounting Swift via FUSE (or something similar)? | 18:12 |
| notmyname | RichardRaseley: FUSE adds a lot of overhead to swift. swift (ie http protocol) adds a lot of overhead to fuse. it can all work, just slow | 18:13 |
| RichardRaseley | notmyname: Thanks. Is there a better way than FUSE to map a Swift cluster to a traditional file system? Or something that you (SwiftStack) tend to recommend in this situation? I am just trying to support a legacy workflow. | 18:14 |
| *** kota_ has joined #openstack-swift | 18:15 | |
| *** ChanServ sets mode: +v kota_ | 18:15 | |
| timburke | jaakkos: if you know when the old version was overwritten (or know some other characteristics of it, like file size, content type, etc.) you can delete it directly from the X-Versions-Location | 18:15 |
| notmyname | well, swiftstack has that as part of the product. so naturally that's what I'd recommend :-) | 18:15 |
| *** Zyric_ has quit IRC | 18:16 | |
| RichardRaseley | notmyname: I am looking at that now on your website, FS GW, correct? And that would provide superior performance to something like FUSE? | 18:17 |
| *** zhill has joined #openstack-swift | 18:17 | |
| notmyname | RichardRaseley: for the inside details, the current FS GW does use FUSE. you could almost think about it as a write-back cache for swift that has a filesystem interface. however, we've also got a different implementation under dev right now | 18:18 |
| jaakkos | does the SwiftStack FS GW have custom index and object format like S3QL, or can you point it to any existing Swift storage? | 18:19 |
| *** Zyric has joined #openstack-swift | 18:19 | |
| *** hezhiqia_ has quit IRC | 18:20 | |
| RichardRaseley | notmyname: Thanks so much, that helps me think about the problem. | 18:20 |
| jaakkos | sounds like you can use any existing (which means rename is going to be slow) | 18:20 |
| *** awelleck has joined #openstack-swift | 18:21 | |
| jaakkos | timburke: alright - so is there a consistent way to get a reference to the versioned object, when the object is modified? | 18:22 |
| jaakkos | timburke: maybe in HTTP response headers or something | 18:22 |
| joeljwright | to everyone travelling to the hackathon - have a good journey and see you in Bristol! | 18:24 |
| *** bjkeller has quit IRC | 18:25 | |
| notmyname | thanks | 18:25 |
| * notmyname has a lot to do before getting on the plane | 18:26 | |
| tdasilva | joeljwright: thanks | 18:26 |
| tdasilva | notmyname: are you leaving today? | 18:26 |
| notmyname | tdasilva: yeah. tonight. will be in bristol saturday evening local time | 18:26 |
| tdasilva | notmyname: cool, have a safe flight | 18:27 |
| notmyname | you too | 18:27 |
| jaakkos | <versions_container>/<length><object_name>/<timestamp> is nice and all but i would really need to know what Swift thought was the timestamp, if I'm to keep a reference to the version (because I'm about to delete it soonish) | 18:27 |
| *** dmorita has joined #openstack-swift | 18:28 | |
| timburke | as i recall, the response doesn't include any indication that versioning was even enabled, much less what the timestamp was used for the old version. immediately after the overwrite, though, you can do a prefixed container listing on the versions container and it should be the last one | 18:30 |
| jaakkos | alright, thanks | 18:31 |
| timburke | in newer versions of swift, include a reverse=on query param and you won't have to worry about the possibility of receiving multiple pages worth of results | 18:31 |
| *** awelleck has quit IRC | 18:37 | |
| *** fthiagogv has quit IRC | 18:46 | |
| *** haomaiwang has quit IRC | 18:46 | |
| *** haomaiwang has joined #openstack-swift | 18:48 | |
| *** arch-nemesis has joined #openstack-swift | 18:50 | |
| *** joeljwright has quit IRC | 18:52 | |
| clayg | timburke: got me a broken probetest on master as best I can tell it might be related to scrub headers? you can sanity check me -> https://gist.github.com/clayg/e7fd810ed043319cbcae | 18:57 |
| timburke | seems likely...still looking... | 19:00 |
| *** haomaiwang has quit IRC | 19:01 | |
| timburke | wait. int? what? why isn't that a '2'? | 19:01 |
| *** 92AAAEL3R has joined #openstack-swift | 19:01 | |
| clayg | because httplib doesn't care - i mean i always send in header values as ints - but only because I can i suppose | 19:02 |
| *** petertr7_away is now known as petertr7 | 19:02 | |
| timburke | ah, right, request header, not response header...client really needs to be robust to ints in request headers | 19:03 |
| clayg | torgomatic: what's the patch # to merge for the plane? something about tests suck when you have no internet? | 19:04 |
| timburke | notmyname: looks like we've got another fix we want before release.... | 19:04 |
| timburke | thanks clayg! | 19:04 |
| torgomatic | clayg: patch 283828 | 19:04 |
| patchbot | torgomatic: https://review.openstack.org/#/c/283828/ - swift - Fix StatsD tests to not use real DNS | 19:04 |
| *** permalac has quit IRC | 19:05 | |
| notmyname | timburke: no problem | 19:05 |
| *** permalac has joined #openstack-swift | 19:05 | |
| clayg | timburke: well... i mean... is that what happened? we used to not care and now scrub headers cares... and but like it needs to because on some headers it does some truncation? I feel like - well idk | 19:05 |
| *** permalac has quit IRC | 19:05 | |
| *** permalac has joined #openstack-swift | 19:05 | |
| *** RichardRaseley has quit IRC | 19:05 | |
| *** delattec has joined #openstack-swift | 19:06 | |
| timburke | clayg: scrub_headers needs to look at str(value). that's all | 19:06 |
| *** awelleck has joined #openstack-swift | 19:07 | |
| *** delatte has quit IRC | 19:08 | |
| *** delatte has joined #openstack-swift | 19:09 | |
| clayg | wow, swiftclient acctually has good number of tests! | 19:09 |
| clayg | timburke: so like this works too -> https://gist.github.com/clayg/8be5b7309f631ba59a32 | 19:10 |
| clayg | timburke: and I like that's it not another conditional type cast in that already messy function | 19:10 |
| clayg | timburke: really it removes a whole nother call for a bunch of headers' values | 19:10 |
| clayg | idk, you're better at all the touchy feely designy stuff - but I'm pretty sure my probetests are broken with out at least one of those | 19:11 |
| clayg | gah - speaking of which I should try to figure out what's going on with the vm-saio-probe :\ | 19:11 |
| *** delattec has quit IRC | 19:11 | |
| timburke | yup, lots of tests. at some point i want to refactor them a bit to be less wordy, though. there's a lot of common or similar setup that could be abstracted; i'm thinking that something similar to the TestEnvs in swift might be nice | 19:12 |
| *** wasmum has quit IRC | 19:15 | |
| *** petertr7 is now known as petertr7_away | 19:17 | |
| *** bjkeller has joined #openstack-swift | 19:18 | |
| clayg | torgomatic: yeah patch 283828 looks fully legit! I almost +A'd it | 19:19 |
| patchbot | clayg: https://review.openstack.org/#/c/283828/ - swift - Fix StatsD tests to not use real DNS | 19:19 |
| torgomatic | notmyname: you heard the man ^^ | 19:19 |
| notmyname | clayg: you just missed it? | 19:19 |
| torgomatic | ;) | 19:19 |
| notmyname | yeah, I've got it open. will look in a bit | 19:20 |
| clayg | timburke: oh no! I always get lost with TestEnvs | 19:20 |
| clayg | timburke: do you want me to push up one or the other of those swiftclient changes or are you looking at it? I'm pretty sure the issue could be demonstrated in a fairly stright forward test... | 19:21 |
| timburke | clayg: working on it. yeah, writing the test was trivial | 19:21 |
| tdasilva | torgomatic: i was just looking at it too | 19:22 |
| clayg | timburke: awesome, I'm going get back on patch 267788 with working probetests - yay! | 19:23 |
| patchbot | clayg: https://review.openstack.org/#/c/267788/ - swift - Fix inconsistent suffix hashes after ssync of tomb... | 19:23 |
| acoles | timburke: I'll add my -1 to clayg's for TestEnvs | 19:23 |
| clayg | timburke: ping me when you cook something up | 19:23 |
| clayg | acoles: !!! | 19:23 |
| acoles | clayg:!!!! | 19:24 |
| timburke | acoles: clayg: duly noted. but it really seems like *something* ought to be refactored there. maybe it's just going to be pulling out some helper methods | 19:24 |
| *** wasmum has joined #openstack-swift | 19:25 | |
| notmyname | timburke: I'll +1 your test envs ;-) | 19:25 |
| acoles | timburke: oh sure, +1 for refactor | 19:27 |
| acoles | timburke: at one point I tried to make it so that a bunch of tests could run using different auth options through inheritance. I can't remember how successful I was. | 19:28 |
| clayg | notmyname: you like swift functests Env mixins? | 19:28 |
| * clayg stares at that tab you know you opened for a reason, but can't remember why | 19:29 | |
| clayg | ha! vm-saio-probe! | 19:30 |
| acoles | patch 26778 | 19:30 |
| patchbot | acoles: https://review.openstack.org/#/c/26778/ - tripleo-image-elements - Make keystone config consistent with other elements. (MERGED) | 19:30 |
| acoles | oh :( | 19:30 |
| acoles | safe travels everyone. good night. | 19:30 |
| *** acoles is now known as acoles_ | 19:31 | |
| *** Jeffrey4l__ has joined #openstack-swift | 19:32 | |
| *** Jeffrey4l_ has quit IRC | 19:35 | |
| *** silor1 has joined #openstack-swift | 19:35 | |
| *** petertr7_away is now known as petertr7 | 19:36 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Force header keys/values to bytes/unicode before coercing to unicode https://review.openstack.org/285514 | 19:36 |
| timburke | clayg: ^^ | 19:37 |
| *** trifon has quit IRC | 19:37 | |
| *** porunov has joined #openstack-swift | 19:38 | |
| *** silor has quit IRC | 19:39 | |
| *** silor1 is now known as silor | 19:39 | |
| timburke | and notmyname: the real question is, would you +2? | 19:39 |
| notmyname | clayg: I kinda like the pattern of setting up some tests and then running them for a set of different envs. like "do this with ascii, then do it all again for unicode" etc | 20:00 |
| *** 92AAAEL3R has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-swift | 20:01 | |
| clayg | notmyname: i like the idea, the implementation in probetests warps my brain | 20:04 |
| clayg | notmyname: maybe if i thought about it harder | 20:04 |
| notmyname | mostly I'm thinking about the unittests. I haven't looked at the probetest code in a while. I think it would warp my brain regardless of envs ;-) | 20:05 |
| clayg | oh it looks like charz already fixed vm-saio-probe - mostly | 20:07 |
| clayg | i probably broke it by updating swift-client :\ | 20:07 |
| openstackgerrit | Merged openstack/python-swiftclient: Fix test for redacting sensitive data in client.http_log() https://review.openstack.org/284822 | 20:07 |
| clayg | notmyname: did I say probe? oh crap - probe on the brain - i ment test envs in *func*tests | 20:08 |
| clayg | i don't think probetests have anything like that - unittests *sorta* have some of that kinda stuff - timburke was talking about doing to refactor swiftclient unittests - i'm sure it will be great | 20:09 |
| notmyname | tdasilva: thanks for patch 283828 | 20:10 |
| patchbot | notmyname: https://review.openstack.org/#/c/283828/ - swift - Fix StatsD tests to not use real DNS | 20:10 |
| timburke | clayg: yeah, might be a bit tho -- still have to fix all the other bugs/deficiencies :P | 20:11 |
| notmyname | heh. step 1: fix all the bugs | 20:11 |
| notmyname | `rm -rf .` done | 20:11 |
| notmyname | timburke: is patch 285514 what needs to land in this release? | 20:13 |
| patchbot | notmyname: https://review.openstack.org/#/c/285514/ - python-swiftclient - Force header keys/values to bytes/unicode before c... | 20:13 |
| *** AJaeger has left #openstack-swift | 20:13 | |
| timburke | yeah. otherwise, we risk breaking things for API users | 20:13 |
| notmyname | right. just making sure I had the right patch :-) | 20:14 |
| *** mmcardle has joined #openstack-swift | 20:15 | |
| *** alejandrito has joined #openstack-swift | 20:20 | |
| *** yarkot_ has joined #openstack-swift | 20:22 | |
| *** yarkot_ has quit IRC | 20:26 | |
| *** silor has quit IRC | 20:26 | |
| *** mmcardle has quit IRC | 20:28 | |
| *** loth1 has left #openstack-swift | 20:30 | |
| *** proteusguy__ has quit IRC | 20:46 | |
| *** stantonnet has quit IRC | 20:47 | |
| *** silor has joined #openstack-swift | 20:49 | |
| *** stantonnet has joined #openstack-swift | 20:49 | |
| *** silor1 has joined #openstack-swift | 20:52 | |
| *** silor has quit IRC | 20:53 | |
| *** silor1 is now known as silor | 20:53 | |
| *** fesp has joined #openstack-swift | 20:58 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** petertr7 is now known as petertr7_away | 21:01 | |
| *** haomaiwang has joined #openstack-swift | 21:01 | |
| tdasilva | gotta run, see you guys in bristol | 21:02 |
| *** petertr7_away is now known as petertr7 | 21:02 | |
| *** davideagle_ has quit IRC | 21:03 | |
| *** proteusguy__ has joined #openstack-swift | 21:03 | |
| *** silor has quit IRC | 21:07 | |
| *** pgbridge has quit IRC | 21:08 | |
| *** silor has joined #openstack-swift | 21:08 | |
| *** yarkot_ has joined #openstack-swift | 21:13 | |
| *** rickyrem has joined #openstack-swift | 21:13 | |
| *** delattec has joined #openstack-swift | 21:15 | |
| *** alejandrito has quit IRC | 21:16 | |
| *** delatte has quit IRC | 21:17 | |
| clayg | what is this icon for gerrit? http://imgur.com/PFs1a34 | 21:19 |
| clayg | ^ I thought my display had dead pixels or something - i guess it's suppose to be a tiny little pincel? | 21:19 |
| notmyname | yeah. looks like a hand with a pencil. maybe you can edit something? | 21:20 |
| *** delattec has quit IRC | 21:20 | |
| notmyname | it feels like gerrit has been signing me out more often lately | 21:20 |
| notmyname | clayg: where'd you see that? I can seem to find it and I don't rememebr it | 21:21 |
| *** delattec has joined #openstack-swift | 21:22 | |
| *** davideagle has joined #openstack-swift | 21:22 | |
| *** stantonnet has quit IRC | 21:23 | |
| *** stantonnet has joined #openstack-swift | 21:23 | |
| *** garthb_ has joined #openstack-swift | 21:28 | |
| *** yarkot_ has quit IRC | 21:29 | |
| *** stantonnet has quit IRC | 21:30 | |
| clayg | notmyname: idk, i was looking at a side by side diff and it was just chillin' in the middle of my screen | 21:30 |
| *** garthb has quit IRC | 21:30 | |
| *** stantonnet has joined #openstack-swift | 21:32 | |
| clayg | notmyname: creepy - it's gone now! | 21:35 |
| *** trifon has joined #openstack-swift | 21:36 | |
| *** fesp has quit IRC | 21:37 | |
| *** NM has quit IRC | 21:38 | |
| *** fesp has joined #openstack-swift | 21:39 | |
| openstackgerrit | Merged openstack/swift: Fix StatsD tests to not use real DNS https://review.openstack.org/283828 | 21:39 |
| *** AndreiaKumpera has quit IRC | 21:42 | |
| *** jamielennox is now known as jamielennox|away | 21:43 | |
| *** porunov has quit IRC | 21:45 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 21:46 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:47 |
| *** haomaiwang has quit IRC | 22:01 | |
| *** haomaiwang has joined #openstack-swift | 22:01 | |
| *** petertr7 is now known as petertr7_away | 22:03 | |
| *** davideagle has quit IRC | 22:17 | |
| *** davideagle_ has joined #openstack-swift | 22:23 | |
| *** davideagle_ has quit IRC | 22:32 | |
| *** silor has quit IRC | 22:34 | |
| *** bjkeller has left #openstack-swift | 22:41 | |
| *** mwheckma1n has quit IRC | 22:41 | |
| *** zaitcev has joined #openstack-swift | 22:42 | |
| *** ChanServ sets mode: +v zaitcev | 22:42 | |
| *** marcusvrn_ has quit IRC | 22:47 | |
| *** trifon has quit IRC | 22:49 | |
| clayg | the thing I like about reviewing acoles_ patches is that it only takes 5-6 hours *tops* [1] -> patch 267788 | 22:54 |
| patchbot | clayg: https://review.openstack.org/#/c/267788/ - swift - Fix inconsistent suffix hashes after ssync of tomb... | 22:54 |
| clayg | 1. total like - some times it can take days to do a decent review - but this one was small :sigh: | 22:54 |
| notmyname | at that speed, you can do 5 every day! | 22:54 |
| *** fesp has quit IRC | 22:59 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** haomaiwa_ has joined #openstack-swift | 23:01 | |
| *** trifon has joined #openstack-swift | 23:03 | |
| *** dmorita has quit IRC | 23:05 | |
| clayg | oh good point - and we have a week! Does that mean we get to make progress!? | 23:05 |
| clayg | i'm going to get ready to leave for the airport | 23:06 |
| clayg | notmyname: see you there probably? | 23:06 |
| notmyname | ya | 23:06 |
| notmyname | in a meeting now | 23:06 |
| *** mmuffinman has quit IRC | 23:06 | |
| *** dmorita has joined #openstack-swift | 23:07 | |
| *** trifon has quit IRC | 23:09 | |
| *** daemontool_ has joined #openstack-swift | 23:10 | |
| *** daemontool has quit IRC | 23:13 | |
| *** chsc has quit IRC | 23:18 | |
| *** awelleck has quit IRC | 23:21 | |
| *** felixcantournet has quit IRC | 23:26 | |
| *** mmuffinman has joined #openstack-swift | 23:28 | |
| *** dmorita has quit IRC | 23:31 | |
| openstackgerrit | Merged openstack/swift: Fix inconsistent suffix hashes after ssync of tombstone https://review.openstack.org/267788 | 23:37 |
| *** dmorita has joined #openstack-swift | 23:42 | |
| *** dmorita_ has joined #openstack-swift | 23:44 | |
| *** dmorita has quit IRC | 23:44 | |
| *** dmorita_ has quit IRC | 23:47 | |
| *** dmorita has joined #openstack-swift | 23:47 | |
| *** dmorita has quit IRC | 23:57 | |
| *** dmorita has joined #openstack-swift | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!