| *** gyee has quit IRC | 00:06 | |
| mattoliverau | timur: cool, add me as a reviewer when you do, cause I'd love to follow up and join in on discussions there, especially after discussions we had at summit. | 00:11 |
|---|---|---|
| *** chlong has joined #openstack-swift | 00:12 | |
| timur | mattoliverau: will do! | 00:18 |
| *** jerrygb has joined #openstack-swift | 00:34 | |
| *** dmorita has quit IRC | 00:37 | |
| *** catintheroof has quit IRC | 00:38 | |
| *** jerrygb has quit IRC | 00:38 | |
| *** catintheroof has joined #openstack-swift | 00:39 | |
| *** dmorita has joined #openstack-swift | 00:39 | |
| *** catintheroof has quit IRC | 00:43 | |
| *** jamielennox is now known as jamielennox|away | 00:44 | |
| openstackgerrit | John Dickinson proposed openstack/swift: Add checksum to object extended attributes https://review.openstack.org/336323 | 00:53 |
| notmyname | torgomatic: so many funny stories in there ^ | 00:53 |
| notmyname | good news is that everythign is working in my SAIO for having an xfs TMPDIR in my home directory and /tmp using tmpfs. so I /think/ I can get https://review.openstack.org/#/c/394600/ updated to do something similar and we'll be good to go | 00:55 |
| patchbot | patch 394600 - openstack-infra/project-config - enable xfs for swift in-process functests | 00:55 |
| timburke | notmyname: no Depends-On: I3339de0d57726b339dfc6638d225e49d097f7b63 ? | 00:55 |
| notmyname | timburke: no. I linked it instead of Depends-On | 00:55 |
| notmyname | why? because...umm... | 00:55 |
| notmyname | seemed like a good idea at the time? (ie 30 seconds ago) | 00:55 |
| notmyname | it works both ways, just skips a lot without the infra test | 00:56 |
| notmyname | and I want to see what happens with the infra change before I make this one depend on it | 00:56 |
| *** clu_ has quit IRC | 00:59 | |
| *** ChubYann has quit IRC | 01:02 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Add checksum to object extended attributes https://review.openstack.org/336323 | 01:10 |
| *** dmorita_ has joined #openstack-swift | 01:11 | |
| notmyname | torgomatic: thanks | 01:11 |
| *** dmorita has quit IRC | 01:11 | |
| *** dmorita has joined #openstack-swift | 01:25 | |
| *** dmorita_ has quit IRC | 01:26 | |
| *** dmorita_ has joined #openstack-swift | 01:33 | |
| *** dmorita has quit IRC | 01:33 | |
| *** dmorita_ has quit IRC | 01:38 | |
| *** dmorita has joined #openstack-swift | 01:40 | |
| *** dmorita_ has joined #openstack-swift | 01:45 | |
| *** dmorita has quit IRC | 01:46 | |
| *** takashi has joined #openstack-swift | 01:50 | |
| *** chlong has quit IRC | 02:11 | |
| *** jerrygb has joined #openstack-swift | 02:34 | |
| *** jerrygb_ has joined #openstack-swift | 02:37 | |
| *** jerrygb has quit IRC | 02:39 | |
| *** winggundamth has joined #openstack-swift | 02:43 | |
| *** david_c_ has quit IRC | 02:45 | |
| kota_ | hello world | 02:45 |
| mattoliverau | kota_: hola o/ | 02:53 |
| kota_ | mattoliverau: hola! glad you are getting back :) | 02:54 |
| *** jerrygb has joined #openstack-swift | 02:54 | |
| mattoliverau | kota_: yeah it's good to be back :) | 02:54 |
| *** jerrygb_ has quit IRC | 02:56 | |
| *** jerrygb has quit IRC | 03:00 | |
| *** diogogmt has quit IRC | 03:15 | |
| *** dikonoor has joined #openstack-swift | 03:20 | |
| *** ppai has joined #openstack-swift | 03:35 | |
| *** ppai has quit IRC | 03:43 | |
| *** takashi has quit IRC | 03:46 | |
| *** takashi has joined #openstack-swift | 03:46 | |
| *** dmorita_ has quit IRC | 03:47 | |
| jrichli | timburke has patchbot-foo | 03:48 |
| jrichli | or is that fu? | 03:48 |
| jrichli | mattoliverau: so glad you are feeling better | 03:49 |
| mattoliverau | jrichli: thanks, yeah I am | 03:49 |
| mattoliverau | if its like mad kung-fu skills then fu.. but maybe as devs foo makes more sense :) | 03:50 |
| *** diogogmt has joined #openstack-swift | 04:18 | |
| *** jerrygb has joined #openstack-swift | 04:20 | |
| *** jerrygb has quit IRC | 04:23 | |
| *** jerrygb has joined #openstack-swift | 04:23 | |
| jrichli | mattoliverau: lol, yes ... i meant fu. but just used to writing foo | 04:23 |
| *** takashi has quit IRC | 04:26 | |
| *** jerrygb has quit IRC | 04:28 | |
| *** gmmaha has quit IRC | 05:00 | |
| *** jerrygb has joined #openstack-swift | 05:01 | |
| *** jerrygb has quit IRC | 05:02 | |
| *** gmmaha has joined #openstack-swift | 05:03 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:05 | |
| *** chirag has joined #openstack-swift | 06:03 | |
| chirag | Hello, can someone help me, what is the difference between starting all swift services via "swift-init all start" & "systemctl start" | 06:06 |
| chirag | In both the cases I start service successfully but I still get an error of 503 service not available :( | 06:07 |
| chirag | Using release Mitaka | 06:07 |
| *** jlwhite has quit IRC | 06:10 | |
| mattoliverau | chirag: what did you use to install swift? | 06:11 |
| chirag | redhat OSP release | 06:11 |
| mattoliverau | ie, I assume you use some system package | 06:11 |
| mattoliverau | Swift by default uses swift-init to start stop services and it manages keeping track of pid files etc. | 06:11 |
| chirag | Yes I used OSP 9 package | 06:12 |
| *** jlwhite has joined #openstack-swift | 06:12 | |
| mattoliverau | Systemd configuration was added by the package mantainer, and I don't know what they do. If they just wrap up swift-init, there I guess there isn't much difference, however if they manage the pid files etc systemd then that will collide with swift-init and you shouldn't use both. | 06:13 |
| chirag | mattoliverau, i started all service successfully useing swift-init, but still when I try uploading objects I get an 503 error | 06:13 |
| mattoliverau | but again, I don't know how the systemd OSP packages were written | 06:14 |
| mattoliverau | what error message are you getting in the logs? | 06:14 |
| * mattoliverau notes I'm about to end for the day as it's after 5pm on a Friday here and the wife is waiting for me. | 06:15 | |
| chirag | just 503 on OUT and additional [Errno 13] Permission denied: '/var/cache/swift/object.recon | 06:15 |
| chirag | *on PUT | 06:16 |
| mattoliverau | well the recon thing just means swift doesn't have permission to dump recon data there, that doesn't stop running. Was that the proxy log or the object servers? sorry gotta go, boss (wife) is now calling me. | 06:22 |
| chirag | no issues, will manage :) | 06:24 |
| *** klrmn has quit IRC | 06:25 | |
| *** Jeffrey4l has quit IRC | 06:35 | |
| *** Jeffrey4l has joined #openstack-swift | 06:48 | |
| *** tesseract has joined #openstack-swift | 07:03 | |
| openstackgerrit | Drew Balfour proposed openstack/swift: add byteorder information and logic to ring files https://review.openstack.org/395237 | 07:03 |
| *** tesseract is now known as Guest7605 | 07:03 | |
| zaitcev | It's a very fresh bug, too | 07:04 |
| zaitcev | chirag: see https://review.openstack.org/387591 | 07:06 |
| patchbot | patch 387591 - swift - Set owner of drive-audit recon cache to swift user (MERGED) | 07:06 |
| *** sams-gleb has joined #openstack-swift | 07:10 | |
| *** sams-gleb has quit IRC | 07:10 | |
| *** sams-gleb has joined #openstack-swift | 07:11 | |
| *** rcernin has joined #openstack-swift | 07:11 | |
| *** sams-gleb has quit IRC | 07:57 | |
| *** sams-gleb has joined #openstack-swift | 07:58 | |
| *** sams-gleb has quit IRC | 08:02 | |
| *** geaaru has joined #openstack-swift | 08:23 | |
| chirag | I am configuring ceilometer with swift facing an error "LookupError: Entry point 'swift' not found in egg 'ceilometer'" | 08:32 |
| chirag | Do I need to specify anything manually other than in proxy.conf? | 08:33 |
| *** amoralej|off is now known as amoralej | 08:34 | |
| *** wuhg has joined #openstack-swift | 08:39 | |
| *** dmorita has joined #openstack-swift | 08:47 | |
| *** dmorita has quit IRC | 08:52 | |
| *** raman has joined #openstack-swift | 09:02 | |
| *** raman has left #openstack-swift | 09:03 | |
| *** asettle has joined #openstack-swift | 09:14 | |
| *** cbartz has joined #openstack-swift | 09:19 | |
| *** cbartz has left #openstack-swift | 09:24 | |
| *** cbartz has joined #openstack-swift | 09:26 | |
| *** sams-gleb has joined #openstack-swift | 09:33 | |
| *** Guess456787654 has joined #openstack-swift | 09:43 | |
| *** openstackgerrit has quit IRC | 09:48 | |
| *** openstackgerrit has joined #openstack-swift | 09:49 | |
| kota_ | tdasilva: hello, can you add links for github code base to https://launchpad.net/pyeclib and https://launchpad.net/liberasurecode description like as https://launchpad.net/swift | 09:50 |
| *** Guess456787654 has quit IRC | 09:50 | |
| kota_ | tdasilva: that's worthful to make the launchpad is starting point of the project (tracking issues, knowing code, etc...) | 09:51 |
| *** ChubYann has joined #openstack-swift | 09:51 | |
| openstackgerrit | Merged openstack/liberasurecode: Change link to pyeclib in README.md https://review.openstack.org/325132 | 09:52 |
| *** Trixboxer has joined #openstack-swift | 10:09 | |
| *** chirag has quit IRC | 10:09 | |
| *** suresh_ has joined #openstack-swift | 10:11 | |
| suresh_ | hii all, how can increase swift upload speed is there any parameter kind of thing in swift configuration | 10:13 |
| *** acoles_ is now known as acoles | 10:13 | |
| suresh_ | please someone help? | 10:13 |
| suresh_ | hii all, Is there any way to increase swift upload speed? is there any parameter kind of thing in swift configuration | 10:16 |
| suresh_ | please someone help? | 10:16 |
| *** suresh_ has quit IRC | 10:36 | |
| *** silor has joined #openstack-swift | 10:40 | |
| *** silor1 has joined #openstack-swift | 10:45 | |
| *** silor has quit IRC | 10:47 | |
| *** silor1 is now known as silor | 10:47 | |
| *** wuhg has quit IRC | 10:51 | |
| *** d0ugal has quit IRC | 11:07 | |
| *** Guess456787654 has joined #openstack-swift | 11:17 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Include debug message for rsync tempfiles https://review.openstack.org/391960 | 11:52 |
| acoles | clayg: timburke I removed some stale code ^^ one of you should probably double check me before +A | 11:55 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: WIP: Allowing copy objects which has special content type https://review.openstack.org/396603 | 11:57 |
| kota_ | timburke: how do you think of patch 396603? | 11:57 |
| patchbot | https://review.openstack.org/#/c/396603/ - swift - WIP: Allowing copy objects which has special conte... | 11:57 |
| kota_ | timburke: I found the behavior when I was trying to re-put all objects in my containers for a reason | 11:58 |
| kota_ | timburke: to consider the context of the delete-marker, it seems ok to disallow to copy the delete-marker to anywhere but not sure right now any case we don't have similar situation (i cannot remenber all use case ';' in the swift middlewares) | 12:00 |
| -openstackstatus- NOTICE: Our OpenStack CI systems are stuck and no new jobs are submitted. Please do not recheck - and do not approve changes until this is fixed. | 12:01 | |
| kota_ | oh, notice says 'do not approve changes' | 12:16 |
| kota_ | so it's good time to be heading home? | 12:17 |
| kota_ | heading for | 12:17 |
| *** vint_bra has joined #openstack-swift | 12:21 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/pyeclib: Make hard-coded dependency for liberasurecode>=1.3.1 https://review.openstack.org/395998 | 12:26 |
| kota_ | got new idea for less code but still in thinking of how we can resolve the gate setup liberasurecode from source. | 12:27 |
| *** catintheroof has joined #openstack-swift | 12:27 | |
| *** catintheroof has quit IRC | 12:31 | |
| *** Guess456787654 has quit IRC | 12:35 | |
| *** Guess456787654 has joined #openstack-swift | 12:36 | |
| *** catintheroof has joined #openstack-swift | 12:40 | |
| *** Guess456787654 has quit IRC | 12:42 | |
| *** Guess456787654 has joined #openstack-swift | 12:50 | |
| *** Guess456787654 has quit IRC | 12:55 | |
| -openstackstatus- NOTICE: Our OpenStack CI system is coming back online again. Thanks for your patience. | 13:00 | |
| *** ma9 has joined #openstack-swift | 13:25 | |
| ma9 | Hi, when I use Swift by exposing its S3 interface, I need to create EC2 credentials. Given that the line is something like this | 13:27 |
| ma9 | openstack credential create --type ec2 --project s3test testuser1 '{"access":"testuser1","secret":"****"}' | 13:27 |
| ma9 | is there a way to make the secret be the same as the keystone password? | 13:27 |
| ma9 | I mean, I would like the user to be able to use S3, and he only should know his password | 13:28 |
| ma9 | and it should be the same he uses to access keystone anyway | 13:28 |
| openstackgerrit | Merged openstack/swift: Py3: Fixes header key dict https://review.openstack.org/348322 | 13:32 |
| *** kei_yama has quit IRC | 13:38 | |
| *** SkyRocknRoll has quit IRC | 13:48 | |
| *** amoralej is now known as amoralej|lunch | 13:54 | |
| tdasilva | kota_: hi, just saw your msg, let me update that | 13:55 |
| kota_ | hi tdasilva, thanks | 13:55 |
| tdasilva | kota_: you are still around! it is late in Tokyo | 13:56 |
| kota_ | tdasilva: just playing with my laptop in night time ;-) | 13:57 |
| *** asettle has quit IRC | 13:59 | |
| tdasilva | kota_: changed it, i think you should now also be able to change ;) | 14:03 |
| *** catinthe_ has joined #openstack-swift | 14:05 | |
| kota_ | tdasilva: thanks! And i fixed small mistake in launchpad liberasurecode (pyeclib -> liberasurecode) | 14:06 |
| tdasilva | kota_: left it there just to see if you were paying attention ;) | 14:06 |
| tdasilva | hehe | 14:07 |
| tdasilva | jk | 14:07 |
| kota_ | lol | 14:07 |
| *** catintheroof has quit IRC | 14:08 | |
| kota_ | tdasilva: btw, do you know how to fix, https://bugs.launchpad.net/pyeclib/+bug/1641074 ? | 14:15 |
| openstack | Launchpad bug 1641074 in PyECLib "Pypi's bug tracker url point to bitbucket" [Undecided,New] | 14:15 |
| *** StraubTW has joined #openstack-swift | 14:17 | |
| kota_ | it looks like no project-config setting and no info about https://bitbucket.org/kmgreen2/pyeclib/issues in the code from my egrep result. | 14:17 |
| *** Renich has joined #openstack-swift | 14:17 | |
| tdasilva | kota_: yeah, i can't really find it where to change that yet | 14:20 |
| tdasilva | lookng | 14:20 |
| *** amoralej|lunch is now known as amoralej | 14:21 | |
| kota_ | tdasilva: thx. not urgent and I'm going to offline for asleep | 14:21 |
| tdasilva | kota_: have a good night! | 14:21 |
| kota_ | g'night | 14:21 |
| *** dikonoor has quit IRC | 14:28 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Unset random seed after rebalancing ring https://review.openstack.org/371564 | 14:30 |
| *** d0ugal has joined #openstack-swift | 14:40 | |
| *** Renich has quit IRC | 14:46 | |
| *** dmorita has joined #openstack-swift | 14:48 | |
| *** dmorita has quit IRC | 14:53 | |
| *** StraubTW_ has joined #openstack-swift | 15:00 | |
| *** sams-gleb has quit IRC | 15:00 | |
| *** StraubTW has quit IRC | 15:01 | |
| *** sams-gleb has joined #openstack-swift | 15:01 | |
| *** sams-gleb has quit IRC | 15:06 | |
| *** sams-gleb has joined #openstack-swift | 15:28 | |
| *** klamath has joined #openstack-swift | 15:35 | |
| *** sgundur_ has joined #openstack-swift | 15:36 | |
| *** klamath has quit IRC | 15:43 | |
| *** sgundur_ has quit IRC | 15:45 | |
| *** asettle has joined #openstack-swift | 15:48 | |
| openstackgerrit | Janie Richling proposed openstack/swift: Automatic refresh of memcache config settings https://review.openstack.org/218490 | 15:48 |
| jrichli | now the default behavior - which is not to enable refresh - is fully documented ^^ | 15:49 |
| *** sgundur_ has joined #openstack-swift | 15:52 | |
| *** rcernin has quit IRC | 16:05 | |
| *** sgundur_ has quit IRC | 16:05 | |
| *** ntata_ has joined #openstack-swift | 16:25 | |
| *** StraubTW_ has quit IRC | 16:27 | |
| *** ma9 has left #openstack-swift | 16:29 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add OPTIONS in manpages https://review.openstack.org/390667 | 16:31 |
| *** ntata_ has quit IRC | 16:34 | |
| *** david_c_ has joined #openstack-swift | 16:34 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Show file parameters as mandatory in swift-*-info man pages https://review.openstack.org/396691 | 16:37 |
| acoles | trivial ^^ | 16:37 |
| *** Guest7605 has quit IRC | 16:43 | |
| *** cbartz has left #openstack-swift | 16:47 | |
| notmyname | good morning | 16:51 |
| *** sgundur_ has joined #openstack-swift | 16:57 | |
| *** david-lyle has quit IRC | 16:58 | |
| clayg | morning! | 17:00 |
| *** asettle has quit IRC | 17:15 | |
| pdardeau | good morning | 17:15 |
| *** asettle has joined #openstack-swift | 17:16 | |
| *** clu_ has joined #openstack-swift | 17:44 | |
| timburke | good morning | 17:47 |
| *** d0ugal has quit IRC | 17:57 | |
| *** sgundur_ has quit IRC | 17:57 | |
| *** david-lyle has joined #openstack-swift | 17:58 | |
| *** sgundur_ has joined #openstack-swift | 18:00 | |
| *** david-lyle has quit IRC | 18:03 | |
| *** klrmn has joined #openstack-swift | 18:05 | |
| *** rcernin has joined #openstack-swift | 18:10 | |
| *** d0ugal has joined #openstack-swift | 18:12 | |
| *** rcernin has quit IRC | 18:12 | |
| *** asettle has quit IRC | 18:24 | |
| *** acoles is now known as acoles_ | 18:28 | |
| *** Administrator_ has joined #openstack-swift | 18:30 | |
| *** zhugaoxiao has quit IRC | 18:33 | |
| *** david-lyle has joined #openstack-swift | 18:36 | |
| jrichli | bkeller`: thanks for the review of the memcache patch! | 18:43 |
| *** sgundur_ has quit IRC | 18:46 | |
| openstackgerrit | Tim Burke proposed openstack/swift: remove double checks on account/container info https://review.openstack.org/317080 | 18:46 |
| *** sgundur_ has joined #openstack-swift | 18:54 | |
| *** catinthe_ has quit IRC | 18:56 | |
| *** sgundur_ has quit IRC | 18:59 | |
| *** sgundur_ has joined #openstack-swift | 19:00 | |
| *** d0ugal has quit IRC | 19:02 | |
| *** pcaruana has quit IRC | 19:07 | |
| *** stradling has joined #openstack-swift | 19:20 | |
| clayg | jrichli: i'm just not sure that account_autocreate is even really a "security" issue? | 19:22 |
| clayg | jrichli: there's one like "allow_account_mangement" - they're related - but I always forget how | 19:22 |
| *** stradling has quit IRC | 19:23 | |
| *** catintheroof has joined #openstack-swift | 19:25 | |
| *** catintheroof has quit IRC | 19:27 | |
| *** catintheroof has joined #openstack-swift | 19:29 | |
| jrichli | oh, i will look more into it. I think my past work experiences have left me with an awareness of over-cautious hardening and compliance checklists | 19:31 |
| notmyname | did I miss something you asked jrichli? I didn't see anything come up in here about account autocreate | 19:32 |
| jrichli | notmyname: comments from https://bugs.launchpad.net/swift/+bug/1629711 | 19:33 |
| openstack | Launchpad bug 1629711 in OpenStack Object Storage (swift) "change account_autocreate to default to true" [Wishlist,In progress] - Assigned to Cheng Li (shcli) | 19:33 |
| notmyname | ah, got it. thanks | 19:34 |
| *** catintheroof has quit IRC | 19:42 | |
| *** catintheroof has joined #openstack-swift | 19:55 | |
| *** geaaru has quit IRC | 19:55 | |
| *** sgundur_ has quit IRC | 19:57 | |
| *** catintheroof has quit IRC | 19:59 | |
| *** sgundur_ has joined #openstack-swift | 20:01 | |
| *** dmorita has joined #openstack-swift | 20:04 | |
| *** dmorita has quit IRC | 20:09 | |
| *** amoralej is now known as amoralej|off | 20:13 | |
| *** sgundur_ has quit IRC | 20:16 | |
| *** silor has quit IRC | 20:22 | |
| *** portante has quit IRC | 20:26 | |
| *** portante has joined #openstack-swift | 20:31 | |
| *** ndk_ has quit IRC | 20:39 | |
| *** ndk_ has joined #openstack-swift | 20:42 | |
| *** catintheroof has joined #openstack-swift | 20:45 | |
| *** portante has quit IRC | 20:46 | |
| *** portante has joined #openstack-swift | 20:51 | |
| *** portante has quit IRC | 20:55 | |
| *** ndk_ has quit IRC | 20:56 | |
| *** sgundur_ has joined #openstack-swift | 20:58 | |
| tdasilva | jrichli: hello, still around? | 21:14 |
| jrichli | tdasilva: yes. what's up? | 21:15 |
| tdasilva | jrichli: in patch 232162 you mentioned seeing func. test errors when running without symlinks in the pipeline but with fast post enabled | 21:16 |
| patchbot | https://review.openstack.org/#/c/232162/ - swift - Symlink implementation. | 21:16 |
| tdasilva | are you still seeing those errors? | 21:16 |
| *** ndk_ has joined #openstack-swift | 21:17 | |
| jrichli | yes, but it's just that symlink tests themselves are not skipped if symlinks is not in the pipeline. | 21:18 |
| tdasilva | jrichli: oh yeah, sorry forgot about that | 21:18 |
| tdasilva | jrichli: jumping on that now | 21:18 |
| *** portante has joined #openstack-swift | 21:20 | |
| openstackgerrit | Merged openstack/swift: Include debug message for rsync tempfiles https://review.openstack.org/391960 | 21:20 |
| openstackgerrit | Merged openstack/swift: Add OPTIONS in manpages https://review.openstack.org/390667 | 21:22 |
| openstackgerrit | Merged openstack/swift: Show file parameters as mandatory in swift-*-info man pages https://review.openstack.org/396691 | 21:22 |
| jrichli | tdasilva clayg: I am planning to create a first cut of the api-ref docs soon | 21:26 |
| jrichli | tdasilva: I also had a question about whether or not we will allow a POST to a symlink to update symlink meta (using symlink=true) | 21:32 |
| *** Jeffrey4l has quit IRC | 21:33 | |
| tdasilva | jrichli: no, at least that's not the plan today | 21:34 |
| tdasilva | a POST to a symlink will return an error | 21:34 |
| *** catinthe_ has joined #openstack-swift | 21:34 | |
| *** catintheroof has quit IRC | 21:34 | |
| tdasilva | symlink=true is only used for GET/HEAD | 21:34 |
| jrichli | oh, ok. I was confused about that. I guess I was thinking that we needed to support updating the symlink metadata for auto-tiering, but I guess that will be different | 21:35 |
| tdasilva | jrichli: yeah, so that would be handled by the auto-tiering feature itself | 21:37 |
| jrichli | there was a comment "TODO: need to document fast post." I wondered what is different for fast post that needed to be documented | 21:38 |
| tdasilva | jrichli: mm...need to look...loading all this stuff in my head again, haven't looked at symlinks for a while now :/ | 21:38 |
| jrichli | ok, np | 21:38 |
| tdasilva | jrichli: will get back to you on that | 21:39 |
| jrichli | ok, i will update with replies to questions you have answered here | 21:40 |
| *** sams-gleb has quit IRC | 21:42 | |
| *** sams-gleb has joined #openstack-swift | 21:43 | |
| *** sgundur_ has quit IRC | 21:43 | |
| *** portante has quit IRC | 21:43 | |
| *** ndk_ has quit IRC | 21:44 | |
| *** sams-gleb has quit IRC | 21:47 | |
| *** ndk_ has joined #openstack-swift | 21:48 | |
| *** portante has joined #openstack-swift | 21:48 | |
| *** si1ver has joined #openstack-swift | 21:58 | |
| si1ver | Is there a guide for setting up swift3 middleware to support aws4 auth? | 22:00 |
| *** vint_bra has quit IRC | 22:08 | |
| *** portante has quit IRC | 22:18 | |
| *** ndk_ has quit IRC | 22:18 | |
| *** portante has joined #openstack-swift | 22:19 | |
| clayg | timburke: kota_: ^ i can barely spell aws | 22:19 |
| si1ver | The documentation appears to not have been updated since icehouse. | 22:20 |
| si1ver | and it mentiones a swift3 config file, which I cannot find. | 22:21 |
| clayg | si1ver: can you drop a link? maybe we can file a bug. | 22:21 |
| notmyname | are you lookign at https://github.com/openstack/swift3? | 22:21 |
| clayg | si1ver: and thanks - sorry I don't more - try to hang around - someone might be ... like notmyname! | 22:22 |
| clayg | *don't know | 22:22 |
| timburke | si1ver: is swift3 already configured? you should get v4 auth support for free with swift3 v1.11 + keystone 9.0 | 22:22 |
| timburke | the config file it's refering to is almost certainly proxy-server.conf | 22:22 |
| si1ver | I have the directives for s3token to reach keystone in proxy-server.conf | 22:23 |
| notmyname | timburke: is the swift3 built docs tree published anywhere? I didn't find it on docs.openstack.org | 22:23 |
| si1ver | Looking for that page again still | 22:23 |
| timburke | notmyname: no idea | 22:23 |
| *** ndk_ has joined #openstack-swift | 22:24 | |
| si1ver | http://docs.openstack.org/juno/config-reference/content/configuring-openstack-object-storage-with-s3_api.html | 22:24 |
| si1ver | I can't find it for mitaka but I swear I have seen it. :p | 22:24 |
| si1ver | I'll check my keystone version | 22:24 |
| si1ver | 2:8.0.0-0ubuntu1~cloud0 | 22:25 |
| si1ver | So that is too old? | 22:25 |
| timburke | si1ver: can you make requests with v2 signatures? | 22:25 |
| si1ver | Yes | 22:25 |
| si1ver | I'm trying to support newer clients that are not v2 capable | 22:26 |
| timburke | ah, yeah, i think you need a version of keystone that includes https://github.com/openstack/keystone/commit/f11d396 | 22:26 |
| si1ver | Does newton ship with keystone 9.0? | 22:26 |
| si1ver | Seems mitaka is 8.0 | 22:26 |
| si1ver | assuming 2:8.0.0.0 actually means 8.0 | 22:27 |
| timburke | oh, weird. i thought mitaka started with 9.0 :-/ | 22:27 |
| si1ver | hang on, might be a ubuntu issue | 22:27 |
| si1ver | apt-get upgrade didn't pick up the cloud repo, so my mistake. | 22:28 |
| si1ver | I've been trying for a week on liberty, trying today on mitaka. | 22:29 |
| si1ver | still getting ' The request signature we calculated does not match the signature you provided. Check your key and signing method.' | 22:30 |
| si1ver | ii keystone 2:9.2.0-0ubuntu1~cloud0 | 22:33 |
| *** ndk_ has quit IRC | 22:34 | |
| si1ver | ok so how about basics. The v2 creds seem to be access="tenant:user" and secret="password". Does v4 use ec2 credentials generated by keystone or does it use the same format as v2? We use v2 with tempauth. | 22:35 |
| *** two_tired has joined #openstack-swift | 22:35 | |
| *** ndk_ has joined #openstack-swift | 22:36 | |
| *** Renich has joined #openstack-swift | 22:40 | |
| two_tired | I want to improve the error handling of swiftclient for when there is a connection reset... Is there any known way to simulate a connection reset? | 22:40 |
| two_tired | the problem that exists today is that when a segment fails due to connection reset, swiftclient waits for the rest of the segments to be uploaded before failing | 22:41 |
| notmyname | two_tired: might be possible to mock/monkey-patch a socket in a test and simulate it. woudl that get you what you need? | 22:42 |
| timburke | si1ver: keystone users always use ec2 credentials. tempauth doesn't yet support v4; i've been meaning to fix that... | 22:42 |
| two_tired | notmyname: yeah, mocking could do it | 22:42 |
| si1ver | Keystone users always use ec2 creds... then what's all the admin-rc and demo-rc stuff about? Those allow the swift client to work with keystone without having ec2 creds. | 22:43 |
| two_tired | In the past I've actually killed active TCP connections, but that's not the same as a connection reset | 22:43 |
| si1ver | OS_USERNAME and OS_PASSWORD don't look like ec2 creds. | 22:43 |
| *** portante has quit IRC | 22:44 | |
| timburke | si1ver: for S3 api requests, i mean. keystone doesn't currently give us a way to validate signatures using the user's actual password | 22:44 |
| si1ver | Ok that makes more sense | 22:44 |
| *** portante has joined #openstack-swift | 22:45 | |
| notmyname | two_tired: I think the place I'd start would be in client.py:HTTPConnection._request() | 22:46 |
| notmyname | two_tired: probably try to mock that, maybe find an underlying socket object, etc | 22:47 |
| two_tired | It's _retry() where I need to fix the exception handling | 22:47 |
| two_tired | Connection._retry() | 22:48 |
| two_tired | it catches SSLError, and then raises it | 22:48 |
| two_tired | seems pointless | 22:48 |
| two_tired | and prevents a retry from happening | 22:49 |
| two_tired | so I'm proposing replace raise with an error message and "pass" | 22:49 |
| two_tired | unless there's some reason not to retry after an SSLError | 22:49 |
| timburke | i think the assumption was that it was the result of an invalid cert | 22:50 |
| notmyname | the commit that changed that was a really big one (porting to use requests). yeah. what timburke said | 22:50 |
| timburke | or some other server-side issue where retrying wouldn't help | 22:50 |
| two_tired | in my situation it's not a cert issue | 22:51 |
| *** mariusv has quit IRC | 22:51 | |
| two_tired | especially considering the rest of the segments complete successfully | 22:51 |
| notmyname | that sounds like a great test to add (and bug to fix, then) :-) | 22:51 |
| *** mariusv has joined #openstack-swift | 22:51 | |
| two_tired | but yeah, I saw that was a big commit, and figured the author meant to come back to it at some point | 22:51 |
| timburke | can we get more info out of the exception to determine what went wrong? if we only re-raise on invalid certs, i think i'd be happy | 22:52 |
| *** ndk_ has quit IRC | 22:54 | |
| si1ver | timburke: so when I run keystone ec2-credentials-list it kicks out tenant, access, and secret. Do I need to do anything with the tenant value? | 22:55 |
| *** ndk_ has joined #openstack-swift | 22:56 | |
| timburke | si1ver: nope; you just need access and secret | 23:00 |
| si1ver | getting 401 errors now from keystone in the proxy log. That's progress. I was getting 404s. | 23:00 |
| timburke | what port are you using? i think i remember something about the admin vs public pipelines having different behavior there... | 23:01 |
| si1ver | 8080 | 23:02 |
| si1ver | here are the proxy logs: http://pastebin.com/R3CXCdfx | 23:02 |
| si1ver | Demo cluster, not on internet. Don't care about the pw being exposed. | 23:02 |
| *** vint_bra has joined #openstack-swift | 23:02 | |
| si1ver | account: none and user_id:none do not sound right to me. | 23:03 |
| si1ver | unless keystone fills in those blanks | 23:04 |
| timburke | sorry, i meant the keystone port. fwiw, looks like we test against 35357? | 23:04 |
| si1ver | Would an s3 client work on that port? That seemed to only be for the python swift client. | 23:05 |
| si1ver | oh wait do you mean the port listed i nthe proxy server config, not the client config? | 23:06 |
| timburke | in proxy-server.conf, the s3token section...yeah! that | 23:06 |
| si1ver | auth_port = 35357 | 23:06 |
| si1ver | I confirmed I can get a token with the admin credentials in the swift config, as suggested by https://ask.openstack.org/en/question/43818/keystone-auth-failure-is-what-it-seems/ | 23:18 |
| si1ver | Here's the keystone log that was at the same time that the proxy log above was from. http://pastebin.com/raw/e6c5rRRQ | 23:25 |
| *** Jeffrey4l has joined #openstack-swift | 23:31 | |
| timburke | hmm. maybe try with port 5000? i know someone else was bit by that admin/public pipeline issue earlier, let me go back through my notes... | 23:34 |
| si1ver | trying | 23:38 |
| two_tired | pprint on my exception gives me: SSLError(SSLError(CertificateError("hostname 'abc.def.com' doesn't match '*.top.secret.com'",),),) | 23:41 |
| two_tired | what's the right way to test the object for CertificateError? | 23:41 |
| si1ver | no luck with port 5000. | 23:48 |
| si1ver | I've gotta run, but thanks for looking. | 23:48 |
| zaitcev | there's a way to use openssl to just open a connection and examine a cert, but in practice I think everyone just runs curl. | 23:51 |
| two_tired | My cert is intentionally bad and I want to be able to identify that it's bad in swiftclient | 23:52 |
| *** Renich has quit IRC | 23:52 | |
| clayg | openssl s_client -connect HOSTNAME | 23:56 |
| clayg | although I hear gnutls-cli is the all the rage now that we're out of legacy ip addresses | 23:56 |
| *** catinthe_ has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!