| *** dmorita has joined #openstack-swift | 00:29 | |
| *** haigang has joined #openstack-swift | 00:56 | |
| *** sandywalsh has joined #openstack-swift | 01:15 | |
| *** Trozz_ has quit IRC | 01:16 | |
| *** Trozz has joined #openstack-swift | 01:17 | |
| *** haigang has quit IRC | 01:18 | |
| *** sandywalsh_ has quit IRC | 01:19 | |
| *** imkarrer has joined #openstack-swift | 01:26 | |
| *** imkarrer_ has quit IRC | 01:28 | |
| *** imkarrer_ has joined #openstack-swift | 01:37 | |
| *** haigang has joined #openstack-swift | 01:38 | |
| *** haigang has quit IRC | 01:40 | |
| *** haigang has joined #openstack-swift | 01:41 | |
| *** imkarrer has quit IRC | 01:41 | |
| *** imkarrer_ has quit IRC | 01:42 | |
| *** haigang has quit IRC | 01:49 | |
| *** haigang has joined #openstack-swift | 01:53 | |
| *** asettle is now known as asettle-afk | 02:02 | |
| *** breitz has quit IRC | 02:05 | |
| *** breitz has joined #openstack-swift | 02:05 | |
| *** haigang has quit IRC | 02:56 | |
| *** haigang has joined #openstack-swift | 02:57 | |
| *** asettle-afk is now known as asettle | 02:59 | |
| mattoliverau | Wow, quiet channel today. | 03:00 |
|---|---|---|
| notmyname | hello mandarin1 | 03:12 |
| notmyname | mattoliverau: | 03:12 |
| notmyname | tab-complete fail. | 03:12 |
| mattoliverau | notmyname: hey man, having a good weekend? | 03:23 |
| notmyname | meh. been fighting a cold or something | 03:24 |
| mattoliverau | notmyname: :( I guesss that happens to us all, even PTLs | 03:25 |
| *** sudorando has quit IRC | 03:48 | |
| *** sudorandom has joined #openstack-swift | 03:50 | |
| notmyname | wheee...setting up EC tests and reviewing content for a book. fun sunday night | 04:10 |
| mattoliverau | notmyname: sounds like it :) I've submitted a patch to add EC support to os-ansible-deployment, next is to trick rackspace private cloud to test it in the lab so I'll have a bigger test cluster to test EC with ;) | 04:14 |
| *** kota_ has joined #openstack-swift | 04:47 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:06 | |
| *** Trozz has quit IRC | 05:11 | |
| *** Trozz has joined #openstack-swift | 05:12 | |
| *** ppai has joined #openstack-swift | 05:28 | |
| *** ppai has quit IRC | 05:33 | |
| *** wolsen_ is now known as wolsen | 05:33 | |
| *** proteusguy has quit IRC | 05:40 | |
| *** early` has quit IRC | 05:44 | |
| *** bkopilov has quit IRC | 05:44 | |
| *** early has joined #openstack-swift | 05:47 | |
| *** remix_auei has joined #openstack-swift | 05:53 | |
| *** remix_tj has quit IRC | 05:56 | |
| *** silor has joined #openstack-swift | 05:59 | |
| *** ktsuyuzaki has joined #openstack-swift | 06:03 | |
| *** bkopilov has joined #openstack-swift | 06:03 | |
| *** kota_ has quit IRC | 06:04 | |
| *** zul has joined #openstack-swift | 06:11 | |
| *** haomaiwang has quit IRC | 06:16 | |
| *** haigang has quit IRC | 06:21 | |
| *** zul has quit IRC | 06:40 | |
| *** foexle has quit IRC | 06:40 | |
| *** haomaiwa_ has joined #openstack-swift | 06:51 | |
| *** haomaiw__ has joined #openstack-swift | 06:54 | |
| *** haomaiwa_ has quit IRC | 06:54 | |
| *** Bsony has joined #openstack-swift | 06:56 | |
| *** haomaiwa_ has joined #openstack-swift | 06:59 | |
| *** haomaiw__ has quit IRC | 07:02 | |
| *** CrackerJackMack has quit IRC | 07:15 | |
| *** joeljwright has joined #openstack-swift | 07:16 | |
| *** CrackerJackMack has joined #openstack-swift | 07:18 | |
| *** krykowski has joined #openstack-swift | 07:24 | |
| *** mmcardle has joined #openstack-swift | 07:26 | |
| *** Bsony has quit IRC | 07:46 | |
| *** geaaru has joined #openstack-swift | 07:47 | |
| *** SkyRocknRoll has quit IRC | 07:57 | |
| *** jistr has joined #openstack-swift | 07:57 | |
| *** zul has joined #openstack-swift | 08:00 | |
| *** SkyRocknRoll has joined #openstack-swift | 08:05 | |
| *** openstackstatus has joined #openstack-swift | 08:07 | |
| *** ChanServ sets mode: +v openstackstatus | 08:07 | |
| *** CrackerJackMack has quit IRC | 08:08 | |
| *** zul has quit IRC | 08:08 | |
| -openstackstatus- NOTICE: Restarting gerrit because it stopped sending events (ETA 15 mins) | 08:10 | |
| *** CrackerJackMack has joined #openstack-swift | 08:11 | |
| *** jordanP has joined #openstack-swift | 08:12 | |
| *** openstackgerrit has quit IRC | 08:13 | |
| *** acoles_away is now known as acoles | 08:16 | |
| *** openstackgerrit has joined #openstack-swift | 08:16 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: WIP - working on the encryption feature. https://review.openstack.org/157907 | 08:19 |
| *** zul has joined #openstack-swift | 08:38 | |
| *** zul has quit IRC | 08:41 | |
| *** zul has joined #openstack-swift | 08:41 | |
| *** aix has joined #openstack-swift | 08:58 | |
| *** fanyaohong has joined #openstack-swift | 09:22 | |
| *** zul has quit IRC | 09:30 | |
| *** zul has joined #openstack-swift | 09:33 | |
| *** mandarin1 is now known as mandarine | 09:39 | |
| *** foexle has joined #openstack-swift | 09:43 | |
| *** haomaiwa_ has quit IRC | 10:31 | |
| *** zul has quit IRC | 10:41 | |
| *** ujjain- has quit IRC | 10:52 | |
| *** ujjain- has joined #openstack-swift | 10:52 | |
| *** Anticimex has quit IRC | 11:13 | |
| *** a1|away is now known as AbyssOne__ | 11:22 | |
| *** zul has joined #openstack-swift | 11:32 | |
| *** dencaval has joined #openstack-swift | 11:36 | |
| *** chmouel has quit IRC | 11:48 | |
| *** dmorita has quit IRC | 11:49 | |
| *** chmouel has joined #openstack-swift | 11:50 | |
| *** chmouel_ has joined #openstack-swift | 11:57 | |
| *** chmouel has quit IRC | 11:57 | |
| *** sileht has quit IRC | 11:59 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Don't show the query string for inner usage https://review.openstack.org/177132 | 12:00 |
| *** krykowski has quit IRC | 12:02 | |
| *** krykowski has joined #openstack-swift | 12:03 | |
| *** krykowski has quit IRC | 12:05 | |
| *** krykowski has joined #openstack-swift | 12:05 | |
| *** proteusguy has joined #openstack-swift | 12:05 | |
| *** openstackgerrit has quit IRC | 12:06 | |
| *** openstackgerrit has joined #openstack-swift | 12:06 | |
| *** sileht has joined #openstack-swift | 12:09 | |
| *** fanyaohong has quit IRC | 12:09 | |
| *** haomaiwang has joined #openstack-swift | 12:13 | |
| *** kei_yama has quit IRC | 12:13 | |
| *** km has quit IRC | 12:14 | |
| *** esker has quit IRC | 12:15 | |
| *** SkyRocknRoll has quit IRC | 12:20 | |
| *** chmouel_ is now known as chmouel | 12:33 | |
| *** SkyRocknRoll has joined #openstack-swift | 12:33 | |
| *** openstackgerrit has quit IRC | 12:37 | |
| *** openstackgerrit has joined #openstack-swift | 12:37 | |
| *** SkyRocknRoll has quit IRC | 12:50 | |
| *** sileht has quit IRC | 12:52 | |
| *** Anticimex has joined #openstack-swift | 13:00 | |
| *** Guest__ has joined #openstack-swift | 13:04 | |
| *** openstackgerrit has quit IRC | 13:21 | |
| *** openstackgerrit has joined #openstack-swift | 13:22 | |
| *** esker has joined #openstack-swift | 13:22 | |
| *** krykowski has quit IRC | 13:29 | |
| *** sileht has joined #openstack-swift | 13:32 | |
| *** krykowski has joined #openstack-swift | 13:38 | |
| *** morganfainberg has quit IRC | 13:44 | |
| *** jroll has quit IRC | 13:44 | |
| *** jroll has joined #openstack-swift | 13:44 | |
| *** xianghui has quit IRC | 13:45 | |
| *** xianghui has joined #openstack-swift | 13:46 | |
| *** morganfainberg has joined #openstack-swift | 13:47 | |
| -openstackstatus- NOTICE: gerrit has been restarted to clear a problem with its event stream. change events between 13:09 and 13:36 utc should be rechecked or have approval votes reapplied as needed to trigger jobs | 13:47 | |
| *** mahatic has joined #openstack-swift | 13:48 | |
| *** vishy has quit IRC | 13:55 | |
| *** vishy has joined #openstack-swift | 13:58 | |
| *** themadcanudist has joined #openstack-swift | 14:02 | |
| themadcanudist | Hey guys, what causes container.dbs to grow? I assume the more entries for objects in there the larger it will be. I have one that's a couple gig in size and it seems to be causing lock timeouts a lot | 14:03 |
| themadcanudist | and is there any way to split/compact these? | 14:04 |
| *** NM has joined #openstack-swift | 14:06 | |
| *** mahatic has quit IRC | 14:16 | |
| *** thumpba has joined #openstack-swift | 14:18 | |
| ctennis | themadcanudist: any idea how many objects are in the container? | 14:18 |
| themadcanudist | how can i resolve container -> account based on filesystem hash of the container.db in question | 14:19 |
| themadcanudist | i can find that number out quick if I can do that | 14:19 |
| glange | themadcanudist: do you delete a lot of objects from that container? | 14:19 |
| themadcanudist | glange: likely | 14:19 |
| *** mahatic has joined #openstack-swift | 14:19 | |
| themadcanudist | i don't, it's another user, but i know there app behaviour | 14:19 |
| glange | http://sqlite.org/lang_vacuum.html <-- maybe it's not getting vacuumed | 14:20 |
| *** vinsh_ has quit IRC | 14:20 | |
| *** thumpba has quit IRC | 14:23 | |
| hurricanerix | morning | 14:26 |
| *** jrichli has joined #openstack-swift | 14:28 | |
| themadcanudist | glange: shouldn't swift automatically take care of the vacuuming, etc? | 14:28 |
| glange | I don't think auto_vacuum is on for container db's and I can't find the word "vacuum" anywhere in the code base | 14:32 |
| glange | http://www.sqlite.org/faq.html <-- look at 12 -- the free space is reused on the next insert, but say you insert way fewer objects than you've deleted, you'll still have that extra space hanging around | 14:33 |
| glange | if you copy your db somewhere and run vacuum on it, you can see if this is what you are seeing -- just check the size of the file after vacuuming | 14:34 |
| themadcanudist | yep | 14:35 |
| themadcanudist | glange: can you tell me how to find out what account a container.db belongs to? | 14:35 |
| *** vinsh has joined #openstack-swift | 14:40 | |
| glange | do sqlite3 cont.db | 14:40 |
| glange | and then select account from container_info; | 14:40 |
| themadcanudist | ah brilliant =) | 14:40 |
| themadcanudist | thx | 14:40 |
| glange | np | 14:41 |
| *** foexle has quit IRC | 14:41 | |
| *** krykowski has quit IRC | 14:48 | |
| *** proteusguy has quit IRC | 14:48 | |
| *** Guest__ has quit IRC | 14:48 | |
| *** Guest__ has joined #openstack-swift | 14:50 | |
| *** annegentle has joined #openstack-swift | 14:51 | |
| *** zul has quit IRC | 14:59 | |
| *** marcusvrn_ has joined #openstack-swift | 14:59 | |
| *** zul has joined #openstack-swift | 14:59 | |
| *** Guest__ has quit IRC | 15:04 | |
| *** breitz has quit IRC | 15:04 | |
| *** Guest__ has joined #openstack-swift | 15:05 | |
| *** breitz has joined #openstack-swift | 15:05 | |
| *** proteusguy has joined #openstack-swift | 15:07 | |
| *** aix has quit IRC | 15:09 | |
| *** annegentle has quit IRC | 15:12 | |
| *** timburke_ is now known as timburke | 15:14 | |
| *** minwoob has joined #openstack-swift | 15:27 | |
| *** foexle has joined #openstack-swift | 15:37 | |
| *** thumpba has joined #openstack-swift | 15:40 | |
| *** pberis has joined #openstack-swift | 15:41 | |
| *** jroll has quit IRC | 15:44 | |
| *** jroll has joined #openstack-swift | 15:46 | |
| *** Guest__ has quit IRC | 15:48 | |
| *** thumpba has quit IRC | 15:49 | |
| *** Guest__ has joined #openstack-swift | 15:50 | |
| *** bkopilov has quit IRC | 15:52 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: WIP Revert to another handoff when primary is unavailable https://review.openstack.org/176403 | 15:52 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Make SSYNC receiver return 5xx when initial checks fail https://review.openstack.org/177836 | 15:52 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Remove _ensure_flush() from SSYNC receiver https://review.openstack.org/177837 | 15:52 |
| *** jroll has quit IRC | 15:53 | |
| *** jroll has joined #openstack-swift | 15:53 | |
| acoles | dfg: if you have chance could you take a quick look at that review 177837 ^^ ? - i saw you did the eventlet patch for minimum_write_chunk_size so I'm guessing you know that ssync code better than me | 15:55 |
| *** gyee has joined #openstack-swift | 15:55 | |
| *** acoles is now known as acoles_away | 15:57 | |
| *** bkopilov has joined #openstack-swift | 15:58 | |
| *** jordanP has quit IRC | 15:59 | |
| *** david-ly_ is now known as david-lyle | 16:01 | |
| *** Guest__ has quit IRC | 16:02 | |
| dfg_ | acoles_away: i actually added that for bulk requests because eventlet buffers stuff and the white spaces that bulk requsts yields to keep the connection alive wouldn't get through otherwise | 16:04 |
| themadcanudist | ctennis: glange: # of objects in container is 8.5 million, the vacuum didn't do much, trimmed 70 megs from 2.2G, so it's efficient | 16:07 |
| themadcanudist | I'm thinking of putting the container ring on SSD only, that should allow me to scale significantly more, right? | 16:07 |
| ctennis | themadcanudist: we generally recommend only about 1 million objects in a contaienr unless your container DB is backed by an SSD | 16:07 |
| ctennis | yes exactly | 16:07 |
| themadcanudist | what's teh recommended upper limit for SSD? | 16:07 |
| ctennis | in the 10ish million range | 16:07 |
| ctennis | it depends on how frequently you do PUTs or DELETEs | 16:08 |
| themadcanudist | aye | 16:08 |
| ctennis | or you can shard any use multiple containers | 16:09 |
| themadcanudist | yeah, we'd need to do this behind the scenes | 16:12 |
| themadcanudist | Is there any fix in future versions of swift to do internal sharding? | 16:13 |
| notmyname | good morning | 16:13 |
| *** zul has quit IRC | 16:16 | |
| ctennis | themadcanudist: it's something that will probably be looked at again once the erasure code features are done, later this year perhaps | 16:17 |
| notmyname | themadcanudist: yup. see mattoliverau's work at https://review.openstack.org/#/c/139921/ (/cc ctennis) | 16:18 |
| *** jistr has quit IRC | 16:21 | |
| mattoliverau | I've been working on sharding a lot since EC release. I hope to have a POC very soon.. And I really need to update that spec. But first... Back to sleep ;) | 16:26 |
| notmyname | I'm thinking mattoliverau must have his alarm tied to IRC notifications | 16:52 |
| notmyname | in which case, I should stop using mattoliverau's nick so that his IRC client doesn't keep beeing | 16:52 |
| notmyname | *beeping | 16:53 |
| notmyname | mattoliverau would probably appreciate that | 16:53 |
| *** Guest___ has joined #openstack-swift | 16:56 | |
| mattoliverau | Lol, may have to disconnect my phone from my irc forwarder, while sleeping ;) | 16:57 |
| *** erlon has joined #openstack-swift | 16:58 | |
| *** Guest___ has quit IRC | 16:59 | |
| *** Guest___ has joined #openstack-swift | 16:59 | |
| *** foexle has quit IRC | 17:01 | |
| themadcanudist | =D | 17:03 |
| *** logan2 has quit IRC | 17:05 | |
| *** Guest___ has quit IRC | 17:07 | |
| *** mahatic has quit IRC | 17:08 | |
| *** mahatic has joined #openstack-swift | 17:10 | |
| *** Guest__ has joined #openstack-swift | 17:12 | |
| *** EmilienM is now known as EmilienM|afk | 17:17 | |
| *** zhill has joined #openstack-swift | 17:40 | |
| *** marcusvrn_ has quit IRC | 17:49 | |
| *** zaitcev has joined #openstack-swift | 17:54 | |
| *** ChanServ sets mode: +v zaitcev | 17:54 | |
| *** peluse has joined #openstack-swift | 18:03 | |
| *** ChanServ sets mode: +v peluse | 18:03 | |
| notmyname | I just got the shipments of swift shirts for the summit | 18:03 |
| notmyname | thye look good | 18:03 |
| peluse | hey, good morning/afternoonish | 18:03 |
| notmyname | hello | 18:04 |
| *** EmilienM|afk is now known as EmilienM | 18:06 | |
| *** cdelatte has joined #openstack-swift | 18:07 | |
| jrichli | notmyname: do we get to hear what's on them now, or is it a surprise? | 18:12 |
| notmyname | http://d.not.mn/vancouver_T_shirt_mockup.png | 18:15 |
| *** geaaru has quit IRC | 18:24 | |
| *** bkopilov has quit IRC | 18:25 | |
| peluse | sweet | 18:25 |
| *** bkopilov has joined #openstack-swift | 18:26 | |
| jrichli | cool! I will take a small red ;-) | 18:28 |
| *** jlk has joined #openstack-swift | 18:31 | |
| jlk | Hey all, I need some help figuring out why swift-proxy is 401ing my tempurl requests. These requests are being generated by tempest. I'm seeing the 401 in the swift-proxy logs but not really any reason behind the 401, and I could really use some help. | 18:32 |
| notmyname | jlk: how are you generating the tempurl? | 18:33 |
| jlk | notmyname: tempest is doing it, I"d have to read up on that I suppose | 18:34 |
| *** thumpba has joined #openstack-swift | 18:34 | |
| jlk | https://github.com/openstack/tempest/blob/master/tempest/api/object_storage/test_object_temp_url.py#L78 | 18:34 |
| *** thumpba has quit IRC | 18:39 | |
| jlk | notmyname: looks like tempest sets metadata for the temp url key, and then uses that key to construct a temp url to post to, and attempts the post | 18:41 |
| notmyname | jlk: that looks correct | 18:41 |
| jlk | swift logs show a 401 on that post, without reason. | 18:41 |
| jlk | swift-proxy logs that is | 18:41 |
| jlk | the tempest client does get this back in the body: | 18:42 |
| jlk | Details: 401 Unauthorized: Temp URL invalid | 18:42 |
| *** themadcanudist has left #openstack-swift | 18:42 | |
| notmyname | jlk: which test is failing? | 18:43 |
| jlk | https://github.com/openstack/tempest/blob/master/tempest/api/object_storage/test_object_temp_url.py#L143 | 18:43 |
| jlk | put_object_using_temp_url | 18:44 |
| jlk | get_object_using_temp_url fails too | 18:44 |
| jlk | which is https://github.com/openstack/tempest/blob/master/tempest/api/object_storage/test_object_temp_url.py#L98 | 18:45 |
| jlk | I can see in the swift proxy logs the object it's trying to get is created | 18:46 |
| jlk | before the attempt to get it | 18:46 |
| clayg | morning | 18:46 |
| peluse | hey what up man? | 18:47 |
| *** cdelatte has quit IRC | 18:49 | |
| *** cdelatte has joined #openstack-swift | 18:49 | |
| glange | clayg: nice video on EC :) | 19:04 |
| *** proteusguy has quit IRC | 19:04 | |
| notmyname | jlk: since _get_expiry_date() is returning something 1000 seconds in the future, I wouldn't think that would be the problem, but time issues are possible | 19:05 |
| notmyname | jlk: check to make sure the local clock is the same as the clock for swift | 19:06 |
| jlk | notmyname: yeah, I just ntp synced both of them. They're within tiny seconds of each other | 19:06 |
| notmyname | jlk: next is to check that the account tempurl key is set to the value you expect | 19:07 |
| jlk | that may be... interesting. The key gets created as set up for the test, and then deleted after the test is done | 19:07 |
| notmyname | well the setUp() seems to check that it's there | 19:08 |
| notmyname | but I don't know the order of resource_setup and resource_cleanup (when they are called) | 19:08 |
| jlk | setup at first, cleanup when existing | 19:09 |
| notmyname | jlk: do you also see failures for the tempurl2 tests? are you seeing errors for every test that's run here? | 19:10 |
| jlk | I don't think the second one ever runs | 19:10 |
| jlk | the only ones that run are the GET and the PUT | 19:10 |
| notmyname | jlk: is this a new test? or has it been there and just now started failing? | 19:12 |
| jlk | well, it's a new cluster, and I"ve never seen this test pass in this cluster, but it could absolutely be something wrong with my cluster | 19:12 |
| jlk | this test passes in the gate I think | 19:12 |
| notmyname | hmmm | 19:12 |
| notmyname | any chance you have odd things in the URL path? I'm comparing it to https://github.com/openstack/swift/blob/master/test/functional/tests.py#L2693 | 19:14 |
| *** proteusguy has joined #openstack-swift | 19:17 | |
| notmyname | jlk: I've got to step away for an appointment | 19:19 |
| jlk | alright, thanks for the help so far | 19:19 |
| notmyname | jlk: I'm hoping you can make some progress, or maybe someone else can help | 19:19 |
| *** ChanServ sets mode: +v torgomatic | 19:21 | |
| *** annegentle has joined #openstack-swift | 19:27 | |
| *** remix_auei has quit IRC | 19:31 | |
| *** remix_tj has joined #openstack-swift | 19:31 | |
| *** remix_tj has quit IRC | 19:35 | |
| *** remix_tj has joined #openstack-swift | 19:36 | |
| MooingLemur | how does object-expirer use memcache? What does it store/retrieve? | 19:45 |
| *** geaaru has joined #openstack-swift | 19:47 | |
| *** Guest__ has quit IRC | 19:55 | |
| *** silor has quit IRC | 19:57 | |
| *** dencaval has quit IRC | 20:01 | |
| *** Guest__ has joined #openstack-swift | 20:05 | |
| clayg | MooingLemur: just normal account/container existance check stuff | 20:06 |
| MooingLemur | clayg: Separate question: I'm not breaking anything by having separate memcache pools for different groups of proxies, right? Since the Swift cache code appears to only support IPv4 but the proxies are only globally reachable to each other on IPv6, I can in theory point each proxy to its own memcache and things will still work as expected, albeit less efficiently. | 20:10 |
| MooingLemur | Considering I've been running this way for quite some time, I suspect it's not detrimental :P But I just want to make sure I understand the ramifications. | 20:12 |
| clayg | heh - are you using keystone or some other authoritive backing auth store? | 20:13 |
| MooingLemur | swauth | 20:13 |
| clayg | yeah you're fine :) seperate memcache pools for proxies will work fine in that case | 20:14 |
| clayg | at least... I also am not aware of any issues with that topology | 20:14 |
| MooingLemur | I think swauth actually PUTs tokens as objects | 20:15 |
| MooingLemur | with X-Delete-At | 20:15 |
| clayg | MooingLemur: that's true about how swauth stores tokens - yes | 20:15 |
| swifterdarrell | kudos to python-swiftclient for auto-creating the *_segments container in the same SP as the manifest for DLO uploads! | 20:32 |
| *** mahatic has quit IRC | 20:35 | |
| *** annegentle has quit IRC | 20:36 | |
| notmyname | clayg: MooingLemur: there is one possibility of a pathological case where separate memcache pools are bad | 20:55 |
| notmyname | if you have a load balancer in front of 2 proxies. auth to proxy A, get token. request with token to proxy B. proxy B returns 401, so you reauth to proxy A. repeat | 20:56 |
| clayg | notmyname: well that's why an authoritive backing store for the auth request is a win | 20:56 |
| clayg | notmyname: in swauth's case the token will be stored in swift - so even if it's not in proxy B's memcache pool - it will fall back to asking swift - then set it in the memcache pool for the next guy | 20:57 |
| clayg | notmyname: same basically goes for keystone | 20:57 |
| notmyname | right | 20:57 |
| notmyname | it's a weird edge case | 20:57 |
| morganfainberg | notmyname: hm. i actually need to make sure i catch you in vancouver | 20:58 |
| notmyname | morganfainberg: great. looking at my calendar now, I think I have a slot from 1:45am to 2:30am on tuesday. ;-) | 20:58 |
| morganfainberg | notmyname: hehe | 20:58 |
| notmyname | (I'm sure your schedule is the same) | 20:58 |
| morganfainberg | notmyname: yeah | 20:58 |
| morganfainberg | notmyname: maybe just grab a lunch sometime | 20:59 |
| * notmyname hopes he gets a chance to eat lunch... | 20:59 | |
| morganfainberg | notmyname: if you don't get a chance to eat lunch, you need to block off lunch and tell people to fly kites when they harass you | 20:59 |
| morganfainberg | ;) | 20:59 |
| notmyname | heh. yeah, I've done that before | 20:59 |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Fix account replication during pre-storage-policy upgrade https://review.openstack.org/177964 | 21:01 |
| *** Guest__ has quit IRC | 21:01 | |
| *** annegent_ has joined #openstack-swift | 21:07 | |
| *** esker has quit IRC | 21:12 | |
| *** foexle has joined #openstack-swift | 21:18 | |
| *** esker has joined #openstack-swift | 21:18 | |
| *** gyee has quit IRC | 21:20 | |
| *** zhill has quit IRC | 21:50 | |
| *** jrichli has quit IRC | 21:58 | |
| mattoliverau | Morning | 22:03 |
| *** delattec has joined #openstack-swift | 22:08 | |
| *** vishy has quit IRC | 22:17 | |
| *** vishy has joined #openstack-swift | 22:19 | |
| *** esker has quit IRC | 22:23 | |
| torgomatic | does anyone have any thoughts on python 2.6 support after the kilo release? | 22:27 |
| torgomatic | just to kick things off, my thoughts are "kill it with fire" | 22:27 |
| * clayg likes the way torgomatic thinks | 22:28 | |
| swifterdarrell | torgomatic: +2, but if someone was running Swift on, like, whatever that old Ubuntu is, Lucid? Or CentOS system python (2.6)... they might... not like that | 22:28 |
| torgomatic | swifterdarrell: true, but Lucid goes EOL on Thursday, so I don't really care about it any more | 22:28 |
| swifterdarrell | torgomatic: but I have enough trouble thinking for myself, I'd hate to try thinking for others... | 22:29 |
| swifterdarrell | torgomatic: hahahahahaha, EOL... | 22:29 |
| torgomatic | swifterdarrell: technically correct is the best kind of correct ;) | 22:29 |
| swifterdarrell | torgomatic: (of course that doesn't mean all the systems running it magically upgrade if their operators haven't done that yet) | 22:29 |
| torgomatic | and it's not like existing clusters will fall over, just that the Llama release won't run on them until they get up to python 2.7 | 22:30 |
| swifterdarrell | torgomatic: I'll give it +2.7 then | 22:31 |
| clayg | torgomatic: not true - the moment *you* say you don't care about python 2.6 the entire world falls over | 22:31 |
| torgomatic | swifterdarrell: 👍 | 22:31 |
| swifterdarrell | torgomatic: haha, my font's real small and I thought that was a middle finger | 22:32 |
| *** foexle has quit IRC | 22:32 | |
| * torgomatic wishes there were an emoji for that | 22:32 | |
| torgomatic | Unicode 7.0 to the rescue! http://emojipedia.org/reversed-hand-with-middle-finger-extended/ | 22:32 |
| openstackgerrit | Denis Cavalcante proposed openstack/swift: Add storage policy support for sorting method https://review.openstack.org/160877 | 22:34 |
| *** km has joined #openstack-swift | 22:37 | |
| *** annegent_ has quit IRC | 22:51 | |
| *** mmcardle has quit IRC | 23:15 | |
| *** NM has quit IRC | 23:19 | |
| *** AbyssOne__ is now known as a1|away | 23:23 | |
| *** kei_yama has joined #openstack-swift | 23:27 | |
| *** geaaru has quit IRC | 23:28 | |
| *** annegentle has joined #openstack-swift | 23:33 | |
| *** annegentle has quit IRC | 23:35 | |
| *** erlon has quit IRC | 23:41 | |
| *** theanalyst has quit IRC | 23:50 | |
| *** ho has joined #openstack-swift | 23:52 | |
| ho | good morning! | 23:53 |
| *** theanalyst has joined #openstack-swift | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
