| *** NM has quit IRC | 00:00 | |
| *** Guest59433 has joined #openstack-swift | 00:09 | |
| *** catintheroof has quit IRC | 00:29 | |
| *** catintheroof has joined #openstack-swift | 00:30 | |
| *** catintheroof has quit IRC | 00:30 | |
| *** Guest59433 has quit IRC | 00:36 | |
| *** jamielennox is now known as jamielennox|away | 00:36 | |
| *** jamielennox|away is now known as jamielennox | 00:37 | |
| *** catintheroof has joined #openstack-swift | 00:42 | |
| tone_z | Good morning! :) | 00:52 |
|---|---|---|
| mattoliverau | tone_z: morning | 00:55 |
| tone_z | mattoliverau: morning :) | 00:56 |
| *** zhurong has joined #openstack-swift | 00:56 | |
| *** vint_bra has quit IRC | 00:57 | |
| tone_z | mattoliverau: good evening :-) | 00:58 |
| mattoliverau | tone_z: well it's almost 12pm here so maybe almost good afternoon really | 00:58 |
| tone_z | oh :) | 00:59 |
| *** m_kazuhiro has joined #openstack-swift | 01:04 | |
| m_kazuhiro | good morning. | 01:04 |
| mattoliverau | m_kazuhiro: morning | 01:10 |
| m_kazuhiro | mattoliverau: morning! | 01:19 |
| kota_ | good morning | 01:30 |
| openstackgerrit | gengchc2 proposed openstack/python-swiftclient master: Removes unnecessary utf-8 coding for python-swiftclient https://review.openstack.org/414881 | 01:36 |
| *** sams-gleb has joined #openstack-swift | 02:01 | |
| *** sams-gleb has quit IRC | 02:05 | |
| *** pxwang_ has quit IRC | 02:16 | |
| *** dja has quit IRC | 02:17 | |
| *** dja has joined #openstack-swift | 02:19 | |
| *** dschultz has quit IRC | 02:27 | |
| *** klrmn has quit IRC | 02:28 | |
| mattoliverau | kota_: morning | 02:30 |
| *** vint_bra has joined #openstack-swift | 02:33 | |
| kota_ | mattoliverau: o/ | 02:35 |
| *** winggundamth has joined #openstack-swift | 02:56 | |
| *** vint_bra has quit IRC | 02:57 | |
| charz_ | clayg: RE: community cluster for https://review.openstack.org/#/c/390781 | 02:59 |
| patchbot | patch 390781 - swift - Replace replication_one_per_device by custom count | 02:59 |
| charz_ | clayg: I saw some function tests failed due to time issue between nodes, likes ` AssertionError: 1489033835.30948 not greater than 1489033835.311847`. | 03:03 |
| charz_ | It just a tiny timedelta (0.002). And I try to keep node synced in 1 seconds, but it seems like can't solve the problem. | 03:05 |
| charz_ | It's not always happen. | 03:05 |
| *** asettle has quit IRC | 03:07 | |
| *** andymccr has quit IRC | 03:08 | |
| charz_ | maybe someone has some ideas? | 03:08 |
| *** lespaul has joined #openstack-swift | 03:26 | |
| *** dmorita has quit IRC | 03:28 | |
| *** dmorita has joined #openstack-swift | 03:28 | |
| *** dmorita has quit IRC | 03:30 | |
| *** lespaul has quit IRC | 03:37 | |
| *** links has joined #openstack-swift | 03:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 03:40 |
| *** catintheroof has quit IRC | 03:57 | |
| *** catintheroof has joined #openstack-swift | 03:58 | |
| *** catintheroof has quit IRC | 03:59 | |
| *** sams-gleb has joined #openstack-swift | 04:03 | |
| *** m_kazuhiro has quit IRC | 04:06 | |
| *** sams-gleb has quit IRC | 04:07 | |
| *** gkadam has joined #openstack-swift | 04:22 | |
| openstackgerrit | Merged openstack/swift master: Support EC policy for in process functional tests https://review.openstack.org/442749 | 04:38 |
| *** psachin has joined #openstack-swift | 04:42 | |
| *** janonymous has quit IRC | 04:54 | |
| *** Mujahid has joined #openstack-swift | 04:56 | |
| jrichli | timburke, clayg: I am trying to determine the best way to add tests when fixing https://bugs.launchpad.net/swift/+bug/1657246 | 04:57 |
| openstack | Launchpad bug 1657246 in OpenStack Object Storage (swift) "Proxy logs wrong request method when validating SLO segments" [Critical,Confirmed] - Assigned to Janie Richling (jrichli) | 04:57 |
| jrichli | For unit testing, maybe just add a mw to the right and verify that the orig_req_method is correct when it is called for the different request types? | 04:58 |
| jrichli | If we want to go even farther, have that mw make a sub-request, and then invoke the real proxy_logging.log_request by having the real proxy_logging involved as well | 04:58 |
| jrichli | but that seems more involved. I don't think we have an example of verifying log output in functional tests, do we? | 04:58 |
| *** dja has quit IRC | 05:23 | |
| *** mahatic_ has quit IRC | 05:24 | |
| *** adriant has quit IRC | 05:41 | |
| *** JimCheung has joined #openstack-swift | 05:50 | |
| *** JimCheung has quit IRC | 05:54 | |
| openstackgerrit | Merged openstack/pyeclib master: Add Phazr.IO libphazr backend support https://review.openstack.org/440781 | 06:03 |
| *** sams-gleb has joined #openstack-swift | 06:05 | |
| *** sams-gleb has quit IRC | 06:10 | |
| mattoliverau | ^^ cool nice one peeps. | 06:16 |
| *** mahatic has joined #openstack-swift | 06:31 | |
| *** janonymous has joined #openstack-swift | 06:34 | |
| *** rcernin has joined #openstack-swift | 06:35 | |
| mahatic | mattoliverau: off for the day? or still around? | 06:42 |
| *** dmorita has joined #openstack-swift | 07:03 | |
| *** hseipp has joined #openstack-swift | 07:04 | |
| *** dmorita has quit IRC | 07:08 | |
| mattoliverau | Off doing baby routine, so pretty much off for the day, but kinda around :) | 07:21 |
| *** PavelK has joined #openstack-swift | 07:26 | |
| *** tesseract has joined #openstack-swift | 07:42 | |
| mahatic | mattoliverau: ah :) no problem, just wanted clarification on your comment on 439572. Can do it later as well | 07:46 |
| *** sams-gleb has joined #openstack-swift | 07:49 | |
| *** ChubYann has quit IRC | 07:51 | |
| *** geaaru has joined #openstack-swift | 07:55 | |
| openstackgerrit | gengchc2 proposed openstack/swift master: Handle readfp deprecation for swift https://review.openstack.org/444144 | 08:12 |
| *** mccabedo has joined #openstack-swift | 08:18 | |
| *** dja has joined #openstack-swift | 08:27 | |
| *** dja has quit IRC | 08:35 | |
| openstackgerrit | gengchc2 proposed openstack/swift-bench master: Handle readfp deprecation for swift-bench https://review.openstack.org/444155 | 08:35 |
| *** bkopilov has quit IRC | 08:36 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Permit to bind object-server on replication_port https://review.openstack.org/337861 | 08:46 |
| *** amoralej|off is now known as amoralej | 08:51 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Small fixes for ec duplciation https://review.openstack.org/438862 | 08:55 |
| acoles | notmyname: jrichli ROFL re p- topics ! There is a reason, but it's lost in history. One day I may switch to q- ;) | 08:58 |
| * acoles is kinda nervous that jrichli *notices* these things :P | 08:59 | |
| acoles | mattoliverau: re patch 442749, thanks for +A, that'll be great if you are able to propose a new gate job, add me as reviewer. I'm not sure whether the jenkins VM will spit out a load of liberasurecode warnings (old version) or worse, so make it non-voting and let's see. | 09:05 |
| patchbot | https://review.openstack.org/#/c/442749/ - swift - Support EC policy for in process functional tests (MERGED) | 09:05 |
| acoles | mattoliverau: I know clayg is working on getting gate to use newer liberasurecode | 09:06 |
| acoles | mattoliverau: BTW, have a great weekend :) | 09:06 |
| kota_ | acoles: o/ just pushed a new patch set for patch 438862 | 09:07 |
| patchbot | https://review.openstack.org/#/c/438862/ - swift - Small fixes for ec duplciation | 09:07 |
| kota_ | acoles: and i found, perhaps, you should update .mailmap to address current your e-mail address? | 09:08 |
| kota_ | acoles: I saw gmail in the gerrit but the mail_map in the swift master looks different (probably stale ones), maybe? | 09:09 |
| acoles | kota_: I saw the new patch, and that md5 optimisation landed :) | 09:10 |
| acoles | kota_: notmyname told me he'd fix the mailmap ;) thanks for the reminder | 09:11 |
| acoles | kota_: apart from composite rings, and the duplication follow up patches, are there any other global EC related patches? | 09:12 |
| kota_ | acoles: yeah, thanks for reviews all, the reason i found the mail address is because I was looking for your new e-mail address to add the co-authored-by line. | 09:12 |
| kota_ | and Test Cleanup I have now. | 09:14 |
| kota_ | acoles:^^ | 09:14 |
| kota_ | the patch, https://review.openstack.org/#/c/440466/ | 09:15 |
| patchbot | patch 440466 - swift - TestObjectController refactoring | 09:15 |
| kota_ | Both effective node election and cross region reconstruction are not yet pushed (still need to think more for me) | 09:16 |
| acoles | oh yes, the test refactor. we need to pick the best time to merge that to try to avoid conflicts with the other follow up patches. I'll take a look, maybe it is ok right now we have the optimisation patch done. | 09:16 |
| acoles | kota_: ok, I have been thinking about "node election" a little...if you mean selecting the best nodes when getting, and when putting, to a composite ring | 09:17 |
| kota_ | ah, and patch 440936 is realted to 440466 | 09:17 |
| patchbot | https://review.openstack.org/#/c/440936/ - swift - Fix (un)patch_policies | 09:17 |
| acoles | got it. | 09:18 |
| kota_ | 440466 is only for replace/moves the structure of Mixins, and then 440936 is for removing unnecessary unpatch_policies | 09:18 |
| acoles | I'm glad you kept those separate :) | 09:19 |
| kota_ | acoles: yeah, to be easy for reviews is much important I learned in the ec_duplication patch ;-) | 09:19 |
| *** tone_z has quit IRC | 09:22 | |
| kota_ | acoles: for "node election", if you have good idea for that, it's super cool because actually I've not yet reach a good idea how to setup the call back with a few changes... | 09:23 |
| kota_ | current call back can be updated by the result of get_responses but the ec_duplciation need both before/after the first get_response tries. | 09:24 |
| *** janonymous has quit IRC | 09:24 | |
| acoles | kota_: yep. I will think some more on that topic. | 09:27 |
| acoles | kota_: also, in the PUT path, I am wondering how we guarantee (in event of failed backend connection) that handoff frags go to handoff nodes in the desired region/ring? | 09:29 |
| acoles | primary nodes have indexes so we can map frags to them, then lack list prioritises which missing frags to PUT, but the node iter produces handoff nodes and how do we know where each handoff is in global cluster? do we need to know? | 09:30 |
| kota_ | acoles: good point | 09:31 |
| acoles | kota_: at this moment in time, I need to get up to speed on your ring zipper patch, plus think some more. I'm sure we'll figure it out :) | 09:32 |
| kota_ | I added that to my task list "what we should think of" now. | 09:32 |
| kota_ | yeah | 09:32 |
| kota_ | acoles: how do you feel the clay's suggestion on the ring zipper? add new command 'zip' to buider vs making a new ring-zipper binary, i mean | 09:34 |
| acoles | kota_: I haven't got that far yet :) | 09:34 |
| acoles | kota_: from your comment ^^ alone, it sounds reasonable, but I should review the patch today | 09:35 |
| kota_ | acoles: ok, you can look at the comments how clayg is thinking at the patch in the gerrit. | 09:35 |
| kota_ | acoles: yeah, i think the basic routine in the current pach can be used for both ways but it affects the interface for users (in paticular operators) | 09:36 |
| kota_ | so now I'm carefully thinking of that while fixising some existing patches :P | 09:36 |
| acoles | kota_: you are *busy*! | 09:37 |
| kota_ | :P it's fun! | 09:37 |
| acoles | jrichli: it crossed my mind to support ec+encryption in process tests, but felt like that would be a separate next step. in this case I don't think there would be any interactions or conflicts between the loaders, so in principle accepting a list of loaders might be sufficient. | 09:38 |
| acoles | kota_: "better busy than bored" | 09:39 |
| kota_ | true | 09:39 |
| acoles | jrichli: but as you say, we'd need to be careful not to get some inconsistent config. that said, this is not a general plugin mechanism, we write this code. | 09:40 |
| *** donagh_ has joined #openstack-swift | 09:41 | |
| *** mccabedo has quit IRC | 09:41 | |
| acoles | ntata: jrichli mentioned you on this patch - does your experience match what I report there? https://review.openstack.org/443162 | 09:43 |
| patchbot | patch 443162 - swift - Document SAIO rsync service setup for ubuntu 16 | 09:43 |
| *** dmorita has joined #openstack-swift | 09:44 | |
| *** dmorita has quit IRC | 09:49 | |
| *** andymccr has joined #openstack-swift | 10:03 | |
| *** asettle has joined #openstack-swift | 10:03 | |
| *** kjz has joined #openstack-swift | 10:09 | |
| *** zhurong has quit IRC | 10:10 | |
| *** openstackgerrit has quit IRC | 10:33 | |
| *** openstackgerrit has joined #openstack-swift | 10:50 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Update .mailmap https://review.openstack.org/444213 | 10:50 |
| acoles | kota_: notmyname ^^ | 10:51 |
| *** kei_yama has quit IRC | 10:51 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Permit to bind object-server on replication_port https://review.openstack.org/337861 | 10:55 |
| kota_ | acoles: that mailmap change looks good to me, I confirmed git shortlog identify you as a person. In the master I can find you twice with current e-mail and the old one. | 11:05 |
| kota_ | acoles: i think i can add +2 +A for that by only myself review? | 11:05 |
| kota_ | single +2 I mean | 11:06 |
| acoles | kota_: you are asking me to give you permission to +A my own patch? :D | 11:06 |
| acoles | kota_: I think you can do that. notmyname can correct us later if we are wrong. | 11:07 |
| kota_ | acoles: sure, just for confirmation | 11:07 |
| kota_ | done | 11:08 |
| acoles | kota_: thanks! have a great weekend | 11:09 |
| kota_ | acoles: thanks too, just preparing to leave my office ;-) | 11:09 |
| kota_ | acoles: and have a nice weekend too | 11:09 |
| *** vint_bra has joined #openstack-swift | 11:15 | |
| Mujahid | while running openstack container create mycontainer getting following error: openstack container create mycontainer | 11:15 |
| Mujahid | Password: | 11:15 |
| Mujahid | Discovering versions from the identity service failed when creating the password plugin. Attempting to determine version from URL. | 11:15 |
| Mujahid | SSL exception connecting to https://9.3.89.178:5000/v3/auth/tokens: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:590) | 11:15 |
| *** oshritf has joined #openstack-swift | 11:16 | |
| *** Mujahid has quit IRC | 11:24 | |
| *** links has quit IRC | 11:37 | |
| *** links has joined #openstack-swift | 11:54 | |
| *** NM has joined #openstack-swift | 11:57 | |
| *** vint_bra has quit IRC | 12:00 | |
| *** SkyRocknRoll has joined #openstack-swift | 12:03 | |
| *** donagh_ has quit IRC | 12:07 | |
| *** dmorita has joined #openstack-swift | 12:07 | |
| *** donagh_ has joined #openstack-swift | 12:11 | |
| *** dmorita has quit IRC | 12:12 | |
| *** dmorita_ has joined #openstack-swift | 12:12 | |
| *** dmorita has joined #openstack-swift | 12:13 | |
| *** dmorita_ has quit IRC | 12:13 | |
| *** sileht has quit IRC | 12:17 | |
| *** dmorita has quit IRC | 12:18 | |
| *** sileht has joined #openstack-swift | 12:18 | |
| *** sileht has quit IRC | 12:18 | |
| *** sileht has joined #openstack-swift | 12:18 | |
| acoles | Mujahid: you left ... but that error looks like it is from openstack client's interaction with keystone (on port 5000) before it even attempts to make a request to swift | 12:19 |
| *** catintheroof has joined #openstack-swift | 12:34 | |
| *** dmorita has joined #openstack-swift | 12:35 | |
| *** dmorita has quit IRC | 12:40 | |
| *** gkadam has quit IRC | 12:44 | |
| *** acoles_ has joined #openstack-swift | 12:45 | |
| *** ChanServ sets mode: +v acoles_ | 12:45 | |
| *** sams-gleb has quit IRC | 12:51 | |
| *** SkyRocknRoll has quit IRC | 12:52 | |
| *** sileht has quit IRC | 13:03 | |
| *** sileht has joined #openstack-swift | 13:03 | |
| *** sileht has quit IRC | 13:03 | |
| *** sileht has joined #openstack-swift | 13:03 | |
| *** sileht has quit IRC | 13:08 | |
| *** sileht has joined #openstack-swift | 13:08 | |
| *** sileht has quit IRC | 13:08 | |
| *** sileht has joined #openstack-swift | 13:10 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Small fixes for ec duplication https://review.openstack.org/438862 | 13:10 |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Include received frag_index in reconstructor log warnings https://review.openstack.org/444281 | 13:10 |
| *** kjz has left #openstack-swift | 13:11 | |
| *** winggundamth has quit IRC | 13:12 | |
| *** links has quit IRC | 13:17 | |
| *** vint_bra has joined #openstack-swift | 13:19 | |
| *** sams-gleb has joined #openstack-swift | 13:20 | |
| *** vint_bra has quit IRC | 13:21 | |
| *** disaster has joined #openstack-swift | 13:22 | |
| disaster | hi! am i suppose to use specific ACL to put on tempurl? or configure something more than generate the url with "swift tempurl PUT 60 url key" and upload with curl --upload-file file url | 13:25 |
| *** links has joined #openstack-swift | 13:31 | |
| *** ujjain has quit IRC | 13:39 | |
| *** links has quit IRC | 13:39 | |
| openstackgerrit | Merged openstack/swift master: Update .mailmap https://review.openstack.org/444213 | 13:48 |
| *** ujjain has joined #openstack-swift | 13:50 | |
| *** ujjain has joined #openstack-swift | 13:50 | |
| rledisez | disaster: you don't need ACL. | 14:01 |
| rledisez | first, setting a key on your container : | 14:01 |
| rledisez | curl -X POST -H "X-Container-Meta-Temp-URL-Key: a_secret_key" -H 'X-Auth-Token: ****' https://example.com/v1/AUTH_3e649efb369040259d5c118fb8fefe7e/container | 14:01 |
| rledisez | second, generating the tempurl: | 14:01 |
| rledisez | swift tempurl PUT 60 /v1/AUTH_3e649efb369040259d5c118fb8fefe7e/container/tempurl_object a_secret_key | 14:01 |
| rledisez | third, use the tempurl (in the next 60 seconds) | 14:01 |
| rledisez | curl -X PUT -T my_file "https://example.com/v1/AUTH_3e649efb369040259d5c118fb8fefe7e/container/tempurl_object?temp_url_sig=d2fb476ebe175f46277ead09f72e3ad78f2b879e&temp_url_expires=1489154271" | 14:01 |
| disaster | rledisez: i come back in 5mn to tell you if it work | 14:03 |
| disaster | rledisez: container need to be public? | 14:06 |
| rledisez | no | 14:06 |
| *** McMurlock1 has joined #openstack-swift | 14:12 | |
| disaster | rledisez: i'm sorry but i don't find out what is x-auth-token | 14:13 |
| rledisez | it's the authentication token you got from keystone | 14:15 |
| rledisez | disaster: if you don't know how to get a token, you can set the meta with the swift client | 14:16 |
| rledisez | you will just need project, user, password | 14:16 |
| rledisez | or run a simple command with swift client, with --debug flag, you'll see the token in the debug logs | 14:17 |
| *** PavelK has quit IRC | 14:21 | |
| *** amoralej is now known as amoralej|lunch | 14:21 | |
| *** gkadam has joined #openstack-swift | 14:24 | |
| *** dmorita has joined #openstack-swift | 14:25 | |
| *** dmorita has quit IRC | 14:30 | |
| disaster | rledisez: 401 unauthorized: Temp URL invalid | 14:30 |
| tdasilva | disaster: if you testing with curl, make sure you have the " around, i always forget and get that error | 14:32 |
| patchbot | Error: No closing quotation | 14:32 |
| disaster | tdasilva: i try with and without both error | 14:34 |
| disaster | rledisez: when u generate the tempurl you put tempurl_object what is this? | 14:37 |
| rledisez | disaster: the name of the object you want your tempurl to point to | 14:39 |
| disaster | rledisez: ok that's the point i have to define the name before upload? I can't send in full container? | 14:40 |
| *** silor has joined #openstack-swift | 14:44 | |
| disaster | rledisez: thank you it work | 14:44 |
| rledisez | disaster: for container, you want to used prefix based tempurl : | 14:46 |
| rledisez | swift tempurl --prefix-based PUT 60 /v1/AUTH_3e649efb369040259d5c118fb8fefe7e/container/ a_secret_key | 14:46 |
| rledisez | i never tried it yet | 14:46 |
| disaster | rledisez: ok i try it | 14:47 |
| disaster | no such option: --prefix-based | 14:48 |
| *** _gleblanc has quit IRC | 14:50 | |
| *** glb has joined #openstack-swift | 14:51 | |
| *** silor1 has joined #openstack-swift | 14:51 | |
| *** silor has quit IRC | 14:52 | |
| *** silor1 is now known as silor | 14:52 | |
| rledisez | disaster: upgrade your swift client, it's a recent feature | 14:53 |
| disaster | rledisez: is it in stable release? | 14:54 |
| tdasilva | disaster: yes, it's in ocata: https://github.com/openstack/python-swiftclient/commits/stable/ocata | 14:54 |
| *** glb has quit IRC | 14:56 | |
| *** glb has joined #openstack-swift | 14:56 | |
| disaster | tdasilva: probably not in apt cause he don't want upgrade | 14:56 |
| rledisez | tdasilva: did you already use it? prefix-based works for a GET, but i can't make a PUT request. is it normal? | 14:56 |
| *** sams-gleb has quit IRC | 14:57 | |
| *** dmorita has joined #openstack-swift | 14:58 | |
| *** sams-gleb has joined #openstack-swift | 14:58 | |
| tdasilva | rledisez: played with it a while back, don't remember the specifics (I thought it would work with PUT), let me test.... | 14:58 |
| tdasilva | too bad cbartz is not around... | 14:58 |
| rledisez | tdasilva: ignore, it works. i probaly mixed up some commands :) | 14:58 |
| tdasilva | ok | 14:58 |
| disaster | so i have to install this from custom package thank you for the help | 15:00 |
| *** catintheroof has quit IRC | 15:00 | |
| disaster | a last thing, is it possible to put kind of checksum in upload? Just to get an error for corrupt send and send again? | 15:01 |
| *** catintheroof has joined #openstack-swift | 15:01 | |
| tdasilva | disaster: not sure about deb. packages, but there are rpm packages already available for rhel based platforms | 15:01 |
| *** dmorita has quit IRC | 15:02 | |
| tdasilva | onovy: do you package swift client too? | 15:02 |
| *** sams-gleb has quit IRC | 15:02 | |
| rledisez | disaster: set a header "Etag" with the md5 of the file you're uploading as value | 15:04 |
| rledisez | curl ... -H 'Etag: b6d81b360a5672d80c27430f39153e2c' | 15:04 |
| disaster | rledisez: that's wonderful thank you | 15:09 |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Test EC chunk_transformer with larger input chunks https://review.openstack.org/442108 | 15:10 |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Small fixes for ec duplication https://review.openstack.org/438862 | 15:10 |
| *** gkadam has quit IRC | 15:11 | |
| *** NM has quit IRC | 15:14 | |
| *** chlong_ has joined #openstack-swift | 15:14 | |
| *** sams-gleb has joined #openstack-swift | 15:17 | |
| onovy | tdasilva: yep | 15:19 |
| tdasilva | onovy: cool! was just checking, didn't know if it was you or somebody else... | 15:20 |
| disaster | tdasilva: i have the 1:2.3.1-1+deb8u1 version seems quite old no? | 15:20 |
| *** dmorita has joined #openstack-swift | 15:21 | |
| onovy | ok | 15:23 |
| notmyname | good morning. I'm taking today off. ping me if you need something, but I might not respond right away | 15:24 |
| *** dmorita has quit IRC | 15:25 | |
| disaster | tdasilva: wow my version is from 2014 if i understand the github correctly | 15:36 |
| tdasilva | disaster: what does 'swift --version' tell you? | 15:37 |
| disaster | 2.3.1 | 15:37 |
| *** chsc has joined #openstack-swift | 15:39 | |
| *** chsc has joined #openstack-swift | 15:39 | |
| tdasilva | disaster: yep, definitely worth an update if you can... | 15:39 |
| disaster | tdasilva: i install python-swiftclient but i see a python3-swiftclient | 15:39 |
| *** amoralej|lunch is now known as amoralej | 15:39 | |
| *** oshritf has quit IRC | 15:41 | |
| disaster | tdasilva: i'm on ovh mirror and ftp.debian.org but he don't want to upgrade the swiftclient | 15:42 |
| disaster | onovy: which version am i suppose to have on debian stable? | 15:42 |
| onovy | current stable/jessie? | 15:43 |
| disaster | onovy: yes | 15:43 |
| onovy | 1:2.3.1-1+deb8u1 or from jessie-backports 1:3.0.0-3~bpo8+1 | 15:43 |
| disaster | onovy: ho ok so it's from 2014 | 15:44 |
| *** vint_bra has joined #openstack-swift | 15:44 | |
| *** vint_bra has quit IRC | 15:45 | |
| disaster | onovy: i saw a python3-swiftclient is he more up to date? | 15:45 |
| onovy | same version | 15:47 |
| disaster | onovy: ok thank you | 15:51 |
| *** rcernin has quit IRC | 15:59 | |
| *** chsc has quit IRC | 16:00 | |
| disaster | tdasilva: as i'm not a real system admin i think i will stay with the old version to avoid any problem ^^ | 16:09 |
| *** klrmn has joined #openstack-swift | 16:10 | |
| *** gabor_antal has quit IRC | 16:13 | |
| *** zackf has joined #openstack-swift | 16:13 | |
| zackf | Anyone have any tips on how to get swift-dispersion-populate to look at a custom cacert? | 16:14 |
| *** chlong_ has quit IRC | 16:19 | |
| timburke | good morning | 16:26 |
| *** Mujahid has joined #openstack-swift | 16:38 | |
| *** dims_ has quit IRC | 16:45 | |
| *** dmorita has joined #openstack-swift | 16:46 | |
| *** joeljwright has joined #openstack-swift | 16:49 | |
| *** ChanServ sets mode: +v joeljwright | 16:49 | |
| *** disaster has quit IRC | 16:53 | |
| *** logan_ has joined #openstack-swift | 16:56 | |
| *** logan_ is now known as Guest29589 | 16:57 | |
| *** dmorita has quit IRC | 16:57 | |
| zackf | Hmm, ok, looks like i found out the answer, it doesn't seem possible to have swift-dispersion-populate work with a custome cafile | 16:57 |
| *** logan- has quit IRC | 16:59 | |
| *** Guest29589 is now known as logan- | 17:03 | |
| *** catintheroof has quit IRC | 17:07 | |
| *** klrmn has quit IRC | 17:07 | |
| *** catintheroof has joined #openstack-swift | 17:08 | |
| *** McMurlock1 has quit IRC | 17:21 | |
| *** dmorita has joined #openstack-swift | 17:21 | |
| *** dmorita has quit IRC | 17:22 | |
| *** dmorita_ has joined #openstack-swift | 17:22 | |
| *** vint_bra has joined #openstack-swift | 17:29 | |
| *** SkyRocknRoll has joined #openstack-swift | 17:33 | |
| *** JimCheung has joined #openstack-swift | 17:40 | |
| *** chsc has joined #openstack-swift | 17:43 | |
| *** chsc has joined #openstack-swift | 17:43 | |
| *** psachin has quit IRC | 17:48 | |
| *** Mujahid has quit IRC | 17:58 | |
| *** klrmn has joined #openstack-swift | 18:01 | |
| *** cbartz has joined #openstack-swift | 18:20 | |
| *** cbartz has quit IRC | 18:26 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: More port number cleanup https://review.openstack.org/444448 | 18:28 |
| timburke | we should probably consider backporting ^^ and https://review.openstack.org/#/c/440694/ as far back as newton... | 18:31 |
| patchbot | patch 440694 - swift - Fix port ranges in install guide (MERGED) | 18:31 |
| *** cbartz has joined #openstack-swift | 18:36 | |
| *** vint_bra has quit IRC | 18:57 | |
| *** cbartz has left #openstack-swift | 19:05 | |
| *** joeljwright has quit IRC | 19:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 19:15 |
| *** acoles_ has quit IRC | 19:23 | |
| *** silor has quit IRC | 19:38 | |
| *** joeljwright has joined #openstack-swift | 19:38 | |
| *** ChanServ sets mode: +v joeljwright | 19:38 | |
| *** chlong_ has joined #openstack-swift | 19:42 | |
| *** dims has joined #openstack-swift | 19:49 | |
| *** ChubYann has joined #openstack-swift | 20:08 | |
| *** hseipp has quit IRC | 20:24 | |
| *** zul is now known as zulVacation | 20:37 | |
| *** amoralej is now known as amoralej|off | 20:44 | |
| *** SkyRocknRoll has quit IRC | 21:00 | |
| ntata | hello there! Has anyone tried to distribute swift services across multiple nodes inside of a multinode devstack setup? | 21:08 |
| ntata | by distributing swift services, I mean deploying "s-proxy" on one node and "s-account", "s-container", "s-object" on the other nodes in the context of 2 node devstack setup | 21:10 |
| timburke | ntata: i seem to recall cschwede looking into it for rolling-upgrade testing | 21:12 |
| ntata | timburke, yes, I'm trying to review some of the grenade tests and see if I can fill gaps. | 21:13 |
| timburke | i think the patch was https://review.openstack.org/#/c/304465/ ? | 21:14 |
| patchbot | patch 304465 - openstack-infra/devstack-gate - Use subnodes for Swift storage nodes in a multinod... | 21:14 |
| ntata | there are two reviews that are kind of alternative approaches. I'm trying to understand both of them https://review.openstack.org/#/c/411982/ and https://review.openstack.org/#/c/411980/10 | 21:15 |
| patchbot | patch 411982 - openstack-infra/project-config - Swift experimental multinode grenade job | 21:15 |
| patchbot | patch 411980 - openstack-infra/devstack-gate - Enable Swift multinode grenade testing | 21:15 |
| ntata | a general process of testing rolling upgrades is to upgrade services bottom up i.e. upgrade storage services (a/c/o) and then upgrade proxy | 21:16 |
| ntata | thinking along those lines, we need to have devstack split services across multiple nodes at the gate when its trying to test rolling upgrade using grenade.. | 21:18 |
| ntata | which means, I need to make devstack-vm-gate.sh script prepare two separate sets of localrc that deploy swift services across multiple nodes | 21:20 |
| ntata | I'm trying to think out loud here so who ever is getting the context I'm building, can chime in and correct me if I'm wrong | 21:20 |
| ntata | I meant cschwede's one approach and other approach is the patch sets I pasted | 21:22 |
| ntata | I looked around for some material on how to deploy multinode devstack and found this - https://docs.openstack.org/developer/devstack/guides/multinode-lab.html | 21:22 |
| ntata | but even this lab installs all of swift on one node.. I'm trying to think is there a current way to split swift services across node? | 21:25 |
| timburke | i think as long as you still configure the rings properly (and i think that's what the SWIFT_STORAGE_IPS var is for?), its fine if things are *installed* on both, just so long as you only *start* services on the nodes that you want -- and then the changes to features.yaml seem to handle that part? | 21:31 |
| timburke | i'm not real familiar with which versions get installed where/when with the grenade/multi-node tests though | 21:34 |
| *** Jeffrey4l__ has joined #openstack-swift | 21:35 | |
| *** Jeffrey4l_ has quit IRC | 21:35 | |
| ntata | well, that's a good point. I never thought of *not starting* services on a node (sometimes, I need to ask question sooner than exploring by myself)! | 21:36 |
| ntata | Thanks timburke, I'll try that out.. I was looking at STORAGE_IPS but wasn't really thinking I could play with it | 21:36 |
| clayg | timburke: how'd you get a bee in your bonnet about default port documentation? You think the *docs* need backporting? Is there some publishing on the internet that somehow gets built from stable branches? | 21:48 |
| timburke | clayg: apparently: https://docs.openstack.org/project-install-guide/object-storage/newton/initial-rings.html | 21:49 |
| timburke | these just came up from the swift bug mailer. they're like 10 minute fixes where leaving them around makes us all look bad | 21:50 |
| clayg | timburke: speak for youself - i look amazing | 21:51 |
| timburke | (notice that https://docs.openstack.org/project-install-guide/object-storage/draft/initial-rings.html has the fixes from 440694) | 21:52 |
| timburke | clayg: amazing -- for publishing crappy, misleading docs? | 21:52 |
| timburke | i'm not saying docs absolutely must be spectacular -- just that they shouldn't lie to you, and when someone points out that they're lying, we should fix them | 21:54 |
| clayg | timburke: I'm not sure what to say - I still use 600X ports on like all my deployments - i guess if the docs were crappy and misleading that's a bummer for whoever uses them which is maybe not zero people? | 21:54 |
| clayg | yeah... there's definately broke things that need fixing - here here | 21:55 |
| timburke | at least two in the past 10 days -- https://bugs.launchpad.net/swift/+bug/1669389 and https://bugs.launchpad.net/swift/+bug/1671896 | 21:56 |
| openstack | Launchpad bug 1669389 in OpenStack Object Storage (swift) "Wrong ports in swift documentation" [Undecided,Fix released] | 21:56 |
| openstack | Launchpad bug 1671896 in OpenStack Object Storage (swift) "Install and configure the storage nodes for Ubuntu and Debian in Installation Guide for Object Storage Service: Ports Are Inconsistent and Incorrect Based on Later Steps in Process" [Undecided,Confirmed] - Assigned to Tim Burke (1-tim-z) | 21:56 |
| clayg | timburke: that report a bug link is amazing! | 21:59 |
| timburke | right? it's great! so i want to encourage people to use it by being responsive | 22:00 |
| clayg | so going back to my initial questions - "a couple of bugs came through my inbox" "yeah we need to back port to fix the published stable docs" | 22:00 |
| timburke | yup. | 22:00 |
| clayg | coolio - did you already purpose the backports for notmyname? | 22:00 |
| timburke | not yet | 22:00 |
| clayg | timburke: so I guess fixing these ports is next? https://docs.openstack.org/project-install-guide/object-storage/draft/storage-install-ubuntu-debian.html | 22:03 |
| timburke | clayg: k, proposed cherry-picks for the one that landed; will hold off on the follow-up until it's on master | 22:03 |
| clayg | ... in the configs I mean? | 22:04 |
| notmyname | timburke: ack | 22:04 |
| clayg | I mean the honest truth I've never tried to go through those docs? | 22:04 |
| clayg | I have no idea if they build a good cluster or now? | 22:04 |
| clayg | I noticed they don't say "use rsync module per disk" - so that's wrong | 22:04 |
| clayg | do they say servers-per-port? | 22:04 |
| clayg | do they get into monitoring? | 22:04 |
| timburke | clayg: no idea. but i *do know* the port numbers are wrong, and they'll be fixed with https://review.openstack.org/#/c/444448/ | 22:05 |
| patchbot | patch 444448 - swift - More port number cleanup | 22:05 |
| clayg | I just don't feel like I have good grasp on the *audience* of those docs - their use-case - the customer - what is the best way to help them? what is the thing they need most? I *guess* those guys need the ports to match the rdo packages... I wonder what ports ubuntu's default configs use... opensuse? | 22:06 |
| clayg | maybe they should just use the ports that are in docs for those configs? | 22:06 |
| clayg | "wrong"? I mean as long as they are internally consistent they're not "wrong"? | 22:07 |
| timburke | well, they're certainly inconsistent *now*. take a look at https://docs.openstack.org/project-install-guide/object-storage/draft/initial-rings.html | 22:08 |
| clayg | "there's no point in trying - better to be part of the problem -- @clayg" | 22:10 |
| clayg | rofl | 22:10 |
| clayg | so maybe fix draft before backporting? I have no idea. Maybe i'm sorry I asked! :P | 22:10 |
| timburke | maybe everyone was happy using the old ports before i went and did that (unwittingly) partial fix. but it seems better to try to bring them in line with our example configs etc. than leave subtle landmines scattered about | 22:11 |
| clayg | I can't think of anything helpful to contribute :'( | 22:13 |
| clayg | heheh - so... pretty much normal for ME! lol | 22:14 |
| timburke | clayg: go +2/+A https://review.openstack.org/#/c/444448/ ! that's helping! | 22:14 |
| patchbot | patch 444448 - swift - More port number cleanup | 22:14 |
| openstackgerrit | Merged openstack/swift master: Add another failing test for 412 logging https://review.openstack.org/443954 | 22:15 |
| *** joeljwright has quit IRC | 22:17 | |
| timburke | clayg: thank you :-) | 22:19 |
| clayg | timburke: maybe next time I should just start with "timburke: I can see you're rage patching - what can I do to help?" | 22:19 |
| * timburke is not actually in a rage; he's just learned that there's only one way to talk to clayg, and that's with an energy that's comparable to what clayg normally outputs *every day* | 22:20 | |
| timburke | it must be exhausting to be clayg | 22:20 |
| clayg | try and keep up people | 22:21 |
| *** chlong_ has quit IRC | 22:21 | |
| *** catintheroof has quit IRC | 22:23 | |
| *** sams-gleb has quit IRC | 22:24 | |
| *** dja has joined #openstack-swift | 22:35 | |
| *** dja has quit IRC | 22:56 | |
| *** dja has joined #openstack-swift | 23:09 | |
| *** tesseract has quit IRC | 23:29 | |
| openstackgerrit | Merged openstack/swift master: More port number cleanup https://review.openstack.org/444448 | 23:32 |
| *** chsc has quit IRC | 23:39 | |
| *** chsc has joined #openstack-swift | 23:53 | |
| *** chsc has joined #openstack-swift | 23:53 | |
| *** chsc has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!