| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Reduce memory usage for download/delete and add --no-shuffle option to st_download https://review.openstack.org/130339 | 00:07 |
|---|---|---|
| ho | good morning guys! | 00:07 |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: This patch fixes downloading files to stdout, allows _SwiftReader to operate as an iterator as well as a context manager. https://review.openstack.org/144899 | 00:09 |
| notmyname | hihi! | 00:09 |
| notmyname | ho: hi! | 00:09 |
| mattoliverau | ho morning | 00:12 |
| ho | notmyname: hi, where do you live? you are always in here so I would like to know your time zone. | 00:12 |
| *** sigmavirus24 has left #openstack-swift | 00:13 | |
| notmyname | ho: I'm in San Francisco. it's currently 4:13pm for me | 00:13 |
| ho | notmyname: I see. you work so hard :) | 00:14 |
| *** echevemaster has quit IRC | 00:16 | |
| ho | Is there any good way to know time zones of swift members? | 00:29 |
| notmyname | nope. just ask. most are in North America. some in EU. mattoliverau is in Australia | 00:30 |
| notmyname | some in India | 00:30 |
| notmyname | all over | 00:30 |
| mattoliverau | woo, go team Australia!! | 00:31 |
| mattoliverau | ho: so you and I are almost the same timezone :) | 00:31 |
| ho | notmyname: thanks. I will ask to each member. | 00:32 |
| notmyname | peluse: https://gist.github.com/notmyname/68874a5c7b5c0cb6d79e | 00:33 |
| *** Masahiro has joined #openstack-swift | 00:33 | |
| remix_tj | a small question: i'm authenticating via keystone and i found that if i create a tenant called i.e. SOMETHING and i create a user SOMETHING with role user (non admin) on that tenant the user can still do admin operations. Is this correct? | 00:35 |
| ho | mattoliverau: morning! yeah, if there is no member in same time zone, it's bit difficult to develop patches for understand basic rules. :) | 00:36 |
| *** david-ly_ has joined #openstack-swift | 00:36 | |
| *** dmorita has joined #openstack-swift | 00:36 | |
| *** Masahiro has quit IRC | 00:38 | |
| ho | remix_tj: can i ask your swift version? | 00:38 |
| *** david-ly_ has quit IRC | 00:41 | |
| mattoliverau | ho: it's not too bad in the swift community, but its slower thanks to reviews and that the dev community are happy to be pinged. We are lucky that the US overlaps at this time of year and notmyname is nice enough to hang around afterhours to help :) | 00:41 |
| mattoliverau | remix_tj: it depends on what you set your operator_roles and reseller_admin_role to in the keystoneauth section of proxy-server.conf | 00:42 |
| ho | mattoliverau: swift dev community is great for me. notmyname, you and other members have great knowledge and skills so I'm very exiting! | 00:45 |
| notmyname | ho: we're glad to have you! | 00:46 |
| notmyname | ho: you're in Japan, right? | 00:46 |
| ho | notmyname: yeah! i'm in japan | 00:47 |
| ho | remix_tj: mattoliverau: it might be related to 'allow_names_in_acls' in the conf | 00:47 |
| notmyname | ho: we've had several other contributors from NTT (eg dmorita). you're at fujitsu, right (based on your email address) | 00:49 |
| ho | nomyname: right. OpenStack Days in Tokyo will be held in Japan from Feb2 to Feb4. I may have a chance to meet other contributers in japan. | 00:51 |
| notmyname | nice! | 00:53 |
| *** Masahiro has joined #openstack-swift | 00:56 | |
| *** addnull has joined #openstack-swift | 01:09 | |
| *** addnull has quit IRC | 01:25 | |
| *** nellysmitt has joined #openstack-swift | 01:37 | |
| *** nellysmitt has quit IRC | 01:42 | |
| *** kota_ has joined #openstack-swift | 01:49 | |
| *** nosnos has joined #openstack-swift | 01:50 | |
| *** fandi has quit IRC | 01:55 | |
| *** haomaiwa_ has joined #openstack-swift | 02:15 | |
| *** gyee has quit IRC | 02:25 | |
| *** occup4nt is now known as occupant | 02:47 | |
| *** addnull has joined #openstack-swift | 02:59 | |
| *** addnull has quit IRC | 03:16 | |
| *** SkyRocknRoll has joined #openstack-swift | 03:25 | |
| *** nosnos has quit IRC | 03:34 | |
| *** bkopilov has quit IRC | 03:37 | |
| *** nellysmitt has joined #openstack-swift | 03:38 | |
| *** nellysmitt has quit IRC | 03:42 | |
| *** ppai has joined #openstack-swift | 03:46 | |
| *** david-lyle has joined #openstack-swift | 03:48 | |
| *** ho__ has joined #openstack-swift | 04:06 | |
| *** ho has quit IRC | 04:09 | |
| *** km has quit IRC | 04:11 | |
| *** david-lyle is now known as david-lyle_afk | 04:25 | |
| *** nosnos has joined #openstack-swift | 04:29 | |
| *** addnull has joined #openstack-swift | 05:03 | |
| *** bkopilov has joined #openstack-swift | 05:11 | |
| *** addnull has quit IRC | 05:11 | |
| *** dmorita has quit IRC | 05:20 | |
| *** zul has quit IRC | 05:21 | |
| *** k4n0 has joined #openstack-swift | 05:24 | |
| *** k4n0 has quit IRC | 05:25 | |
| *** addnull has joined #openstack-swift | 05:36 | |
| *** nellysmitt has joined #openstack-swift | 05:38 | |
| *** nellysmitt has quit IRC | 05:43 | |
| *** nshaikh has joined #openstack-swift | 05:47 | |
| *** sandywalsh has quit IRC | 05:55 | |
| *** sandywalsh has joined #openstack-swift | 05:57 | |
| *** zaitcev has quit IRC | 06:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Imported Translations from Transifex https://review.openstack.org/145138 | 06:14 |
| *** ho__ has quit IRC | 06:14 | |
| openstackgerrit | Madhuri Kumari proposed openstack/swift: This patch lets GET/HEAD fail on object in deleted account. https://review.openstack.org/145141 | 06:24 |
| mattoliverau | well I'm calling it a day, night all. | 06:29 |
| *** abhirc has quit IRC | 06:37 | |
| *** ho has joined #openstack-swift | 06:52 | |
| *** ppai has quit IRC | 06:58 | |
| *** ppai has joined #openstack-swift | 07:12 | |
| *** nellysmitt has joined #openstack-swift | 07:39 | |
| *** nellysmitt has quit IRC | 07:43 | |
| *** ppai has quit IRC | 07:46 | |
| *** chlong has quit IRC | 07:49 | |
| *** chlong has joined #openstack-swift | 07:49 | |
| *** chlong has quit IRC | 07:51 | |
| *** chlong has joined #openstack-swift | 07:53 | |
| *** ppai has joined #openstack-swift | 08:00 | |
| *** fandi has joined #openstack-swift | 08:08 | |
| *** ppai has quit IRC | 08:26 | |
| *** nellysmitt has joined #openstack-swift | 08:33 | |
| *** occupant has quit IRC | 08:34 | |
| *** ppai has joined #openstack-swift | 08:40 | |
| *** occupant has joined #openstack-swift | 08:42 | |
| *** jistr has joined #openstack-swift | 08:58 | |
| *** nshaikh has quit IRC | 09:01 | |
| *** addnull has quit IRC | 09:02 | |
| *** addnull has joined #openstack-swift | 09:05 | |
| openstackgerrit | Merged openstack/swift: Add node/pair index patch back into feature/EC https://review.openstack.org/134065 | 09:05 |
| *** jordanP has joined #openstack-swift | 09:06 | |
| *** nshaikh has joined #openstack-swift | 09:06 | |
| *** nellysmitt has quit IRC | 09:08 | |
| openstackgerrit | Prashanth Pai proposed openstack/swift: Avoid unnecessary unlink() on every successful PUT https://review.openstack.org/143943 | 09:09 |
| *** nellysmitt has joined #openstack-swift | 09:12 | |
| *** ajay_ has quit IRC | 09:21 | |
| *** addnull has quit IRC | 09:22 | |
| *** Masahiro has quit IRC | 09:27 | |
| *** Masahiro has joined #openstack-swift | 09:27 | |
| *** Masahiro has quit IRC | 09:31 | |
| *** addnull has joined #openstack-swift | 09:32 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Fix slo constraint according to the error message https://review.openstack.org/145177 | 09:33 |
| *** Masahiro has joined #openstack-swift | 09:37 | |
| *** Masahiro has quit IRC | 09:39 | |
| acoles | remix_tj: did you get an answer to your question? it looks like there is a feature in keystoneauth that permits access when username==tenantname, but that feature is disabled by default. It is enabled by setting is_admin = True in keystoneauth config, and would log a warning if used. | 09:40 |
| acoles | remix_tj: otherwise, as mattoliverau said, check if 'user' is included in the roles you have configured in operator_roles | 09:41 |
| *** Masahiro has joined #openstack-swift | 09:50 | |
| ho | remix_tj: acoles: sorry. "allow_names_in_acls" is wrong. the info from acole "is_admin" is what i wanted to say. sorry. | 09:53 |
| *** Masahiro has quit IRC | 09:53 | |
| acoles | ho: hi! | 09:53 |
| ho | acoles: hi, thank you for your correction. | 09:54 |
| ho | acoles: i should check the code before i wrote it. | 09:56 |
| acoles | ho: no problem :) | 09:56 |
| *** addnull has quit IRC | 09:56 | |
| *** Masahiro has joined #openstack-swift | 09:57 | |
| *** Masahiro has quit IRC | 10:00 | |
| *** nshaikh has left #openstack-swift | 10:06 | |
| *** Masahiro has joined #openstack-swift | 10:09 | |
| remix_tj | acoles: no, operator roles are by default admin and SwiftOperator, so maybe is this is_admin = True | 10:13 |
| remix_tj | ho: is the version included in icehouse | 10:13 |
| *** Masahiro has quit IRC | 10:14 | |
| *** kota_ has quit IRC | 10:29 | |
| remix_tj | confirm, is_admin = true. But wasn't set by me, is a default option? | 10:29 |
| acoles | remix_tj: no, the default is false, and i have just noticed that it is not even included in the sample config https://github.com/openstack/swift/blob/master/etc/proxy-server.conf-sample#L281 | 10:38 |
| remix_tj | acoles: maybe is a rdo choice | 10:40 |
| acoles | remix_tj: could be, i dont know answer to that sorry | 10:41 |
| remix_tj | acoles: no problem, the real answer was why. It's suggested to set to false? Could lead to a security issue? | 10:42 |
| acoles | remix_tj: there is a note in the code to say it is deprecated, written before my time on project. IMHO it seems like a convenient shortcut that could lead to surprises (as you have found). | 10:45 |
| acoles | remix_tj: Also, in general it is better to us IDs for auth and not names because names may not be unique since keystone v3 API. | 10:45 |
| acoles | remix_tj: there can be users/tenants with same names in different keystone domains. | 10:45 |
| acoles | but IDs are globally unique. | 10:46 |
| *** Masahiro has joined #openstack-swift | 10:54 | |
| *** Masahiro has quit IRC | 10:58 | |
| *** Masahiro has joined #openstack-swift | 11:00 | |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Fix cross account upload using --os-storage-url https://review.openstack.org/125759 | 11:03 |
| remix_tj | acoles: ok, thank you for the explanation. I'll write down your notes and tomorrow i'll fix my dev cluster | 11:05 |
| *** addnull has joined #openstack-swift | 11:07 | |
| *** addnull has quit IRC | 11:11 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Enable Object Replicator's failure count in recon https://review.openstack.org/138342 | 11:13 |
| *** haomaiwa_ has quit IRC | 11:14 | |
| mattoliverau | acoles: thanks for heads up, I didn't know the option existed! Maybe it should be in the sample config with a deprecation and warning comment. I'll throw together a patch in the morning if no one else does. Is it mentioned in the doc (I'm too lazy and its too late for me to check from my phone) :p | 11:16 |
| mattoliverau | acoles: oh and happy new year :) | 11:16 |
| acoles | mattoliverau: hi! happy new year to you too. yeah, i may (or may not) get that done today, i don't think it is in the doc | 11:17 |
| mattoliverau | Nah I just had a quick look and couldn't find it. No stress, I'll either whip it up or review it in the morning ;p | 11:21 |
| *** nshaikh has joined #openstack-swift | 11:38 | |
| *** abhirc has joined #openstack-swift | 11:44 | |
| *** ho has quit IRC | 11:48 | |
| *** Masahiro has quit IRC | 11:48 | |
| *** nellysmitt has quit IRC | 11:50 | |
| *** kei_yama has joined #openstack-swift | 11:54 | |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Reduce memory usage for download/delete and add --no-shuffle option to st_download https://review.openstack.org/130339 | 11:55 |
| *** fandi has quit IRC | 11:56 | |
| *** kei_yama has quit IRC | 11:57 | |
| *** NM has joined #openstack-swift | 12:06 | |
| *** nellysmitt has joined #openstack-swift | 12:16 | |
| *** ppai has quit IRC | 12:40 | |
| *** zul has joined #openstack-swift | 12:42 | |
| *** Masahiro has joined #openstack-swift | 12:49 | |
| *** ppai has joined #openstack-swift | 12:54 | |
| *** Masahiro has quit IRC | 12:54 | |
| *** addnull has joined #openstack-swift | 12:59 | |
| *** fandi has joined #openstack-swift | 13:11 | |
| *** ppai has quit IRC | 13:13 | |
| *** lpabon has joined #openstack-swift | 13:47 | |
| *** tdasilva has joined #openstack-swift | 13:50 | |
| *** ajay_ has joined #openstack-swift | 13:57 | |
| *** guest123 has quit IRC | 13:58 | |
| ajay_ | In swift-init, I get this error. "cannot import name make_subrequest". I followed: http://docs.openstack.org/developer/swift/development_saio.html. I use fedora 20. | 13:59 |
| *** silor has joined #openstack-swift | 13:59 | |
| ajay_ | Can anyone help me? | 13:59 |
| *** Masahiro has joined #openstack-swift | 14:00 | |
| NM | ajay_: Can you show the full commando you are using? | 14:03 |
| *** nosnos has quit IRC | 14:04 | |
| *** Masahiro has quit IRC | 14:05 | |
| *** mahatic has joined #openstack-swift | 14:07 | |
| ajay_ | NM: swift-init main start | 14:08 |
| ajay_ | http://pastebin.com/nEWDANfr | 14:10 |
| ajay_ | I have python2.7 | 14:10 |
| ajay_ | 2.7.5 to be precise | 14:10 |
| *** kragniz has quit IRC | 14:22 | |
| *** kragniz has joined #openstack-swift | 14:22 | |
| NM | from your python shell, can you import the lib? | 14:25 |
| NM | $python | 14:25 |
| NM | >>> from swift.common.wsgi import make_subrequest | 14:25 |
| NM | Does it work? | 14:25 |
| *** addnull has quit IRC | 14:29 | |
| *** SkyRocknRoll has quit IRC | 14:40 | |
| *** key has joined #openstack-swift | 14:51 | |
| *** silor has quit IRC | 14:51 | |
| *** SkyRocknRoll has joined #openstack-swift | 14:53 | |
| *** key has quit IRC | 14:53 | |
| *** nellysmitt has quit IRC | 14:54 | |
| *** jistr has quit IRC | 14:56 | |
| *** jistr has joined #openstack-swift | 14:57 | |
| *** rebelshr_ has joined #openstack-swift | 15:07 | |
| *** torgomatic has quit IRC | 15:10 | |
| *** torgomatic has joined #openstack-swift | 15:10 | |
| *** ChanServ sets mode: +v torgomatic | 15:10 | |
| *** nshaikh has quit IRC | 15:10 | |
| *** dmsimard_away is now known as dmsimard | 15:22 | |
| *** bkopilov has quit IRC | 15:33 | |
| *** Nadeem has joined #openstack-swift | 15:38 | |
| openstackgerrit | Merged openstack/swift: Build up reconstructor with correct node selection https://review.openstack.org/129361 | 15:43 |
| *** bkopilov has joined #openstack-swift | 15:48 | |
| *** Masahiro has joined #openstack-swift | 15:49 | |
| *** barra204_ has joined #openstack-swift | 15:51 | |
| *** barra204_ is now known as shakamunyi | 15:53 | |
| *** Masahiro has quit IRC | 15:54 | |
| openstackgerrit | paul luse proposed openstack/swift: New hashes.pkl format and other reconstructor changes https://review.openstack.org/131872 | 15:55 |
| clayg | peluse: you *did* seemed sorta jazzed about the going socket to socket for the fragment reconstructor - I think you really nailed it on the class name "DiskFileLikeThing" | 16:03 |
| peluse | clayg, sweet, thanks | 16:06 |
| peluse | clayg, didn't want to have to rebuild the whole thing in mem though but the send.put() needs all the metadata up front so I had to calc content len and etag before sending | 16:08 |
| *** ajay_ has quit IRC | 16:12 | |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Fix cross account upload using --os-storage-url https://review.openstack.org/125759 | 16:14 |
| *** jistr has quit IRC | 16:15 | |
| *** jistr has joined #openstack-swift | 16:16 | |
| *** jistr has quit IRC | 16:17 | |
| *** jistr has joined #openstack-swift | 16:17 | |
| mordred | notmyname: (or anyone) if I'm using swiftclient put_object with a chunk size of 8192, I'm assuming that's reading the file in 8192 byte chunks, right? I've got code where I'd _ALSO_ like to get an md5 of the file I'm uploading which I could do by hashlib update-ing each chunk ... | 16:30 |
| mordred | so the questions are - 1) is there a better way to do "please upload this file and also tell me its md5" 2) is there a way to hook in to the current chunking code | 16:31 |
| mordred | or do I just need to go lower level myself? | 16:31 |
| *** bkopilov has quit IRC | 16:34 | |
| *** SkyRocknRoll has quit IRC | 16:35 | |
| mordred | ACTUALLY | 16:39 |
| mordred | nevermind | 16:39 |
| mordred | I'm a complete moron | 16:39 |
| * mordred goes back to his hole | 16:40 | |
| mahatic | torgomatic, hi, around? | 16:40 |
| *** gyee has joined #openstack-swift | 16:40 | |
| notmyname | mordred: glad you got it figured out ;-) | 16:48 |
| *** nellysmitt has joined #openstack-swift | 16:50 | |
| mordred | notmyname: 5 more minutes of digging before I open my fat mouth next time | 16:51 |
| NM | mordred: Have you read "Time management for sysadmins" ? That is pretty common. Just talking about a problem helps to solve it :) | 16:52 |
| mordred | NM: no - I should clearly read that though ... | 16:55 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add undocumented options to keystoneauth sample config https://review.openstack.org/145266 | 17:01 |
| acoles | mattoliverau: ^^ | 17:01 |
| mahatic | notmyname, hello! | 17:15 |
| notmyname | mahatic: good morning. how are you? | 17:15 |
| mahatic | notmyname, i'm good. how are you? Happy 2015! | 17:16 |
| notmyname | you too | 17:16 |
| mahatic | notmyname, thanks, https://review.openstack.org/#/c/140103/ | 17:16 |
| mahatic | notmyname, torgomatic mentioned X-swift-server-type, not sure what that is | 17:17 |
| notmyname | mahatic: that's what we've talked abut before. you'll need to respond with some info about what kind of server it is. that way recon will be able to tell if it's the right kind or not | 17:17 |
| mahatic | notmyname, okay. that comes in a different patch then, correct? | 17:19 |
| *** jistr has quit IRC | 17:22 | |
| notmyname | mahatic: maybe. it could be either way. | 17:24 |
| notmyname | it would be a pretty small follow-on patch | 17:24 |
| notmyname | hmm...actually, yeah, probably makes sense to do it as a separate patch | 17:24 |
| mahatic | okay | 17:26 |
| notmyname | mahatic: from your perspective, is there anything else you want to add or have outstanding on the current patch? or is it ready for full review? | 17:28 |
| mahatic | notmyname, it is ready I believe, but there is one comment on using sort/sorted in python, I felt sorted would be more efficient. Can you take a look at the comment? | 17:30 |
| *** Masahiro has joined #openstack-swift | 17:38 | |
| *** Masahiro has quit IRC | 17:43 | |
| notmyname | mahatic: the comment is correct. since sorted would be called on every call to allowed_methods, then sorting just one time in the if block would be more efficient | 17:56 |
| *** jordanP has quit IRC | 18:00 | |
| mahatic | notmyname, okay, will modify that. And i think that will be all for the full review | 18:21 |
| notmyname | ok, thanks | 18:21 |
| mahatic | will send it right away | 18:21 |
| notmyname | mahatic: how's your understanding of recon going? | 18:22 |
| mahatic | notmyname, not so good, i'm afraid | 18:23 |
| notmyname | mahatic: ok. because you haven't had time yet, or because you're blocked somewhere? | 18:23 |
| *** bkopilov has joined #openstack-swift | 18:25 | |
| notmyname | acoles: thank for being the "Translations Approve Bot" for today :-) | 18:25 |
| torgomatic | mahatic: I'm around now | 18:25 |
| mahatic | notmyname, partly both. I couldn't get ahead because of some blockade, I hung around in IRC for a while and when I saw no one around much, that's when I started wasting time ;) | 18:26 |
| mahatic | torgomatic, cool, Happy 2015! And it was about your comment on a patch. notmyname clarified. Will modify and submit | 18:26 |
| notmyname | mahatic: what have you tried so far, on the recon side, to understand | 18:27 |
| mahatic | notmyname, I will get back with my questions and more clarity by tomorrow | 18:27 |
| notmyname | ok, great | 18:27 |
| acoles | notmyname: i needed something easy to finish the day | 18:28 |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Add notion of overload to swift-ring-builder https://review.openstack.org/141452 | 18:28 |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Only move too-close-together replicas when they can spread out. https://review.openstack.org/140879 | 18:28 |
| notmyname | :-) | 18:28 |
| *** jistr has joined #openstack-swift | 18:30 | |
| *** jistr has quit IRC | 18:31 | |
| *** nellysmitt has quit IRC | 18:31 | |
| *** jistr has joined #openstack-swift | 18:32 | |
| *** reed has joined #openstack-swift | 18:34 | |
| *** acoles is now known as acoles_away | 18:35 | |
| *** nellysmitt has joined #openstack-swift | 18:42 | |
| notmyname | redbo: around? | 18:42 |
| openstackgerrit | Dhriti Shikhar proposed openstack/swift: Substituted object storage paragraph with simple definition https://review.openstack.org/145294 | 18:45 |
| *** nellysmitt has quit IRC | 18:46 | |
| *** reed has quit IRC | 18:49 | |
| notmyname | redbo: acoles_away: cschwede: dfg: glange: we were looking at the replication code while doing some EC work and found this. needs confirmation but it looks wrong https://bugs.launchpad.net/swift/+bug/1408061 | 18:58 |
| redbo | If it's handled by update_deleted, it shouldn't really be getting a lot of objects added to it. But yeah, it's a little racy. I'm not sure how we'd fix it. I mean other than long-term, make ssync work and remove objects as they're replicated instead of in bulk. | 19:05 |
| *** shakamunyi has quit IRC | 19:07 | |
| notmyname | redbo: we (myself, torgomatic, clayg, peluse) are sitting in a room together looking at EC stuff. we're using replication as a pattern for the EC rebuilding stuff (the reconstructor), and that's what we saw. we might have a way to sidestep it with EC, but probably needs something on the replicated policies too | 19:08 |
| *** dkingshott has joined #openstack-swift | 19:11 | |
| notmyname | redbo: ahale_: how is ssync working for you these days? | 19:12 |
| notmyname | there's some new patches that are dependent on it (global replication optimizations and erasure codes), but I'd also love to be able to start recommending it to use in prod instead of rsync | 19:12 |
| redbo | we've never been able to deploy it wide-scale because replication traffic interferes with user requests too much | 19:14 |
| notmyname | have you used a separate replication network yet or are you still running with one physical network (not that "just install a new physical network" is possible) | 19:16 |
| redbo | all on one network | 19:16 |
| notmyname | ok | 19:16 |
| notmyname | so it's chattier that rsync? it's not like you aren't doing anything today | 19:17 |
| *** reed has joined #openstack-swift | 19:17 | |
| redbo | I don't think it's necessarily chattier. We have performance problems with the object server normally, and then ssync dumps a ton more traffic on there. | 19:21 |
| notmyname | ah, ok. so cpu bound, not bandwidth | 19:21 |
| notmyname | ya, I can understand that | 19:21 |
| notmyname | heh. just wait for EC. fun times. | 19:22 |
| *** Masahiro has joined #openstack-swift | 19:27 | |
| redbo | We're looking at running a second object server on a different port for replication so it's not competing with users, at least as far as GIL and threadpools and stuff are concerned. | 19:28 |
| notmyname | ya, IIRC that's what we do for our customers | 19:29 |
| glange | notmyname: I bet not everybody will be running EC, it's optional, right? | 19:29 |
| notmyname | glange: "optional" until your product manager and sales guy hears about it ;-) | 19:29 |
| glange | we don't have those, so we are good! :) | 19:29 |
| notmyname | glange: but ya. it's "just" another storage policy | 19:29 |
| redbo | RIP swagner | 19:30 |
| notmyname | nick isnt' there anymore? | 19:30 |
| redbo | he's the product manager for amazon prime video or something now | 19:31 |
| notmyname | oh. cool. good for him | 19:31 |
| redbo | I need to get his phone number so I can nag him about getting more seasons of Justified on prime | 19:31 |
| *** Masahiro has quit IRC | 19:31 | |
| glange | he's on the http://amazon.com/leadership page now for some reason | 19:32 |
| *** zaitcev has joined #openstack-swift | 19:32 | |
| *** ChanServ sets mode: +v zaitcev | 19:32 | |
| notmyname | glange: I don't think I have the same view of that page as you | 19:34 |
| glange | it's eventually consistent, keep trying :) | 19:35 |
| *** david-lyle_afk is now known as david-lyle | 19:39 | |
| openstackgerrit | Mahati proposed openstack/swift: Implement OPTIONS verb for storage nodes. https://review.openstack.org/140103 | 19:49 |
| *** shakamunyi has joined #openstack-swift | 19:51 | |
| notmyname | peluse: https://gist.github.com/notmyname/907ed26a9d9db22d9de8 | 19:52 |
| clayg | glange & redbo !!!! | 19:53 |
| notmyname | peluse: https://gist.github.com/notmyname/317fdb2c50a2fdb1f3e2 | 19:54 |
| *** barra204_ has joined #openstack-swift | 19:54 | |
| clayg | glange: here? http://www.amazon.com/s/?field-keywords=nick+swanger | 19:55 |
| clayg | oh swagner? | 19:56 |
| clayg | that didn't work either... | 19:56 |
| glange | https://flic.kr/p/oCdVhS <-- that blury picture is the only known picture of him, he's at the end of the table | 19:57 |
| * clayg can't wait to be trolled | 19:58 | |
| clayg | so dissapoint | 19:58 |
| glange | no, that's really him, sorry :) | 19:58 |
| notmyname | glange: I like how dfg is hiding from the camera | 19:58 |
| clayg | who's that you're making out with? | 19:58 |
| glange | it was a reverse troll, the worse kind | 19:58 |
| *** shakamunyi has quit IRC | 19:58 | |
| clayg | this flicker stream is great! | 19:59 |
| clayg | nom nom pickles | 19:59 |
| glange | haha | 19:59 |
| clayg | hey that's muir woods!? | 20:00 |
| glange | yeah, I visited my brother a while back, he lives just south of San Fran and is going to law school at Berkley | 20:00 |
| clayg | i thought gholt's brother lived in south san fran - are you guys brothers! | 20:01 |
| glange | https://flic.kr/p/pLdWEH <-- there is a dfg picture for you :) | 20:01 |
| clayg | glange: which park is this -> https://www.flickr.com/photos/greg_lange/14686186409/in/photostream/ | 20:02 |
| clayg | is it a good course? | 20:03 |
| glange | that is golden gate park, it's a fairly decent course, it has epic trees | 20:03 |
| glange | I was surprised there was a course in San Fran because space is at such a premium | 20:05 |
| ahale_ | i had just managed to get over swagner now all the memories are back :( | 20:09 |
| *** jistr has quit IRC | 20:25 | |
| *** tellesnobrega has quit IRC | 20:28 | |
| *** tellesnobrega has joined #openstack-swift | 20:41 | |
| *** nellysmitt has joined #openstack-swift | 20:43 | |
| openstackgerrit | Merged openstack/swift: Imported Translations from Transifex https://review.openstack.org/145138 | 20:44 |
| *** pcaruana is now known as pcaruana|afk| | 20:45 | |
| *** nellysmitt has quit IRC | 20:48 | |
| *** gyee has quit IRC | 21:00 | |
| *** NM has quit IRC | 21:13 | |
| *** Masahiro has joined #openstack-swift | 21:16 | |
| mattoliverau | Morning | 21:17 |
| notmyname | hi mattoliverau | 21:17 |
| *** Masahiro has quit IRC | 21:20 | |
| *** lpabon has quit IRC | 21:32 | |
| *** ajiang has joined #openstack-swift | 21:35 | |
| *** ajiang has quit IRC | 21:37 | |
| *** barra204_ has quit IRC | 21:39 | |
| *** barra204_ has joined #openstack-swift | 21:39 | |
| *** lpabon has joined #openstack-swift | 21:41 | |
| *** lpabon has quit IRC | 21:41 | |
| *** mahatic has quit IRC | 22:10 | |
| *** barra204_ has quit IRC | 22:12 | |
| *** barra204 has joined #openstack-swift | 22:13 | |
| *** gyee has joined #openstack-swift | 22:14 | |
| *** dmsimard is now known as dmsimard_away | 22:15 | |
| *** elambert has joined #openstack-swift | 22:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 22:52 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 22:52 |
| *** lcurtis has joined #openstack-swift | 22:56 | |
| lcurtis | hello all...continuing my swift journey | 22:58 |
| lcurtis | could someone remind me what config file or action pairs storage nodes to a proxy node? | 22:59 |
| lcurtis | ie, how to set which nodes are paired with a proxy? | 22:59 |
| notmyname | lcurtis: a proxy is a proxy if it's running swift-proxy-server | 23:04 |
| *** Masahiro has joined #openstack-swift | 23:04 | |
| notmyname | it's just the processes that are running. there's no other "designation" in swift. whatever management tool you're using might have something additional | 23:05 |
| lcurtis | hello again notmyname! okay thank you | 23:05 |
| lcurtis | so if i wanted flexibility i could run proxy on every node and perhaps just not use that proxy? | 23:06 |
| notmyname | correct | 23:06 |
| notmyname | in that case you'd add or remove it from your load balance pool | 23:07 |
| lcurtis | and not much overhead otherwise? | 23:07 |
| *** Masahiro has quit IRC | 23:09 | |
| notmyname | lcurtis: "much" is such a relative term ;-) | 23:10 |
| lcurtis | lolol | 23:10 |
| notmyname | lcurtis: if you've got a proxy server process sitting there not getting and requests, it won't consume many resources aside from loading the ring into memory | 23:11 |
| lcurtis | great...okay...good to know | 23:12 |
| notmyname | you've already got 30 servers, you said, so at that size you'll probably want to start thinking about optimizing some SKUs for different tasts. eg having dedicated proxy nodes or separate account+container servers | 23:12 |
| lcurtis | still figuring things out as i go | 23:12 |
| lcurtis | yes | 23:12 |
| lcurtis | and waiting for network equipment and things needed yesterday, etc | 23:12 |
| lcurtis | so trying to build in some flexibility so i can move proxy nodes to different servers | 23:13 |
| lcurtis | deploying through puppet | 23:13 |
| notmyname | that seems smart (building the deployment flexibility in) | 23:13 |
| lcurtis | running scared is all...lol | 23:13 |
| notmyname | lcurtis: really, it depends on your workload. low request rates for a lot of storage (eg backups) has a much different hardware footprint than high request rates to storage (eg public web/mobile content) | 23:14 |
| *** nottrobin has quit IRC | 23:20 | |
| *** miqui has quit IRC | 23:20 | |
| *** notmyname has quit IRC | 23:21 | |
| *** Tyger_ has quit IRC | 23:21 | |
| *** esmute has quit IRC | 23:21 | |
| *** ctennis has quit IRC | 23:21 | |
| *** NobodyCam has quit IRC | 23:21 | |
| *** wolsen has quit IRC | 23:21 | |
| *** dosaboy has quit IRC | 23:21 | |
| *** tanee2 has quit IRC | 23:21 | |
| *** reed has quit IRC | 23:21 | |
| *** Nadeem has quit IRC | 23:21 | |
| *** IRTermite has quit IRC | 23:21 | |
| *** hurricanerix has quit IRC | 23:21 | |
| *** dvorkbjel has quit IRC | 23:21 | |
| *** bobby2 has quit IRC | 23:21 | |
| *** ekarlso- has quit IRC | 23:21 | |
| *** btorch has quit IRC | 23:21 | |
| *** evanjfraser has quit IRC | 23:21 | |
| *** Anticimex has quit IRC | 23:21 | |
| *** mandarine has quit IRC | 23:21 | |
| *** redbo has quit IRC | 23:21 | |
| *** dfg has quit IRC | 23:21 | |
| *** esmute has joined #openstack-swift | 23:21 | |
| *** NobodyCam has joined #openstack-swift | 23:21 | |
| *** tanee has joined #openstack-swift | 23:21 | |
| *** wolsen has joined #openstack-swift | 23:21 | |
| *** Tyger has joined #openstack-swift | 23:21 | |
| *** nottrobin has joined #openstack-swift | 23:21 | |
| *** notmyname has joined #openstack-swift | 23:22 | |
| *** ChanServ sets mode: +v notmyname | 23:22 | |
| *** reed has joined #openstack-swift | 23:22 | |
| *** Nadeem has joined #openstack-swift | 23:22 | |
| *** IRTermite has joined #openstack-swift | 23:22 | |
| *** hurricanerix has joined #openstack-swift | 23:22 | |
| *** dvorkbjel has joined #openstack-swift | 23:22 | |
| *** bobby2 has joined #openstack-swift | 23:22 | |
| *** ekarlso- has joined #openstack-swift | 23:22 | |
| *** btorch has joined #openstack-swift | 23:22 | |
| *** evanjfraser has joined #openstack-swift | 23:22 | |
| *** mandarine has joined #openstack-swift | 23:22 | |
| *** Anticimex has joined #openstack-swift | 23:22 | |
| *** redbo has joined #openstack-swift | 23:22 | |
| *** dfg has joined #openstack-swift | 23:22 | |
| *** sendak.freenode.net sets mode: +v redbo | 23:22 | |
| *** ctennis has joined #openstack-swift | 23:24 | |
| *** dvorkbjel has quit IRC | 23:24 | |
| *** miqui has joined #openstack-swift | 23:25 | |
| *** dosaboy has joined #openstack-swift | 23:26 | |
| *** dvorkbjel has joined #openstack-swift | 23:27 | |
| *** remix_auei has joined #openstack-swift | 23:33 | |
| *** remix_tj has quit IRC | 23:33 | |
| *** Nadeem has quit IRC | 23:38 | |
| *** barra204 has quit IRC | 23:43 | |
| *** ho has joined #openstack-swift | 23:53 | |
| *** dkingshott has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!