| kota_ | good morning | 00:07 |
|---|---|---|
| kota_ | nice to see encryption with barbican landed! | 00:08 |
| mattoliverau | kota_: morning | 00:08 |
| kota_ | mattoliverau: morning | 00:09 |
| kota_ | notmyname: ack for https://review.openstack.org/#/c/495805/, it looks like not a big patch (I'm just opening the URL though) so I think I can do it today. | 00:09 |
| patchbot | patch 495805 - swift - Drop reconstructor stats when worker has no devices | 00:09 |
| *** baojg has quit IRC | 00:16 | |
| *** baojg has joined #openstack-swift | 00:17 | |
| *** baojg has quit IRC | 00:21 | |
| *** catintheroof has quit IRC | 00:25 | |
| *** kiennt has joined #openstack-swift | 00:36 | |
| *** shuyingya has joined #openstack-swift | 00:43 | |
| *** shuyingya has quit IRC | 00:45 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 00:48 |
| *** mat128 has joined #openstack-swift | 00:51 | |
| *** cshastri has joined #openstack-swift | 00:59 | |
| *** klrmn has quit IRC | 01:12 | |
| *** shuyingya has joined #openstack-swift | 01:16 | |
| *** bkopilov has quit IRC | 02:13 | |
| *** shuyingya has quit IRC | 02:14 | |
| *** shuyingya has joined #openstack-swift | 02:14 | |
| *** baojg has joined #openstack-swift | 02:23 | |
| *** mvk_ has quit IRC | 02:25 | |
| *** baojg has quit IRC | 02:28 | |
| *** catintheroof has joined #openstack-swift | 02:30 | |
| *** shuyingya has quit IRC | 02:31 | |
| *** mvk_ has joined #openstack-swift | 02:36 | |
| *** catintheroof has quit IRC | 02:57 | |
| *** itlinux has joined #openstack-swift | 03:04 | |
| *** itlinux has quit IRC | 03:21 | |
| *** Tgrv has joined #openstack-swift | 03:21 | |
| *** bkopilov has joined #openstack-swift | 03:31 | |
| *** gkadam has joined #openstack-swift | 03:33 | |
| *** itlinux has joined #openstack-swift | 03:35 | |
| *** itlinux has quit IRC | 03:36 | |
| *** links has joined #openstack-swift | 03:51 | |
| *** mat128 has quit IRC | 03:53 | |
| *** itlinux has joined #openstack-swift | 04:16 | |
| *** itlinux has quit IRC | 04:16 | |
| *** itlinux has joined #openstack-swift | 04:49 | |
| *** itlinux has quit IRC | 05:17 | |
| *** tovin07_ has joined #openstack-swift | 05:18 | |
| *** shuyingya has joined #openstack-swift | 05:19 | |
| *** psachin has joined #openstack-swift | 05:24 | |
| *** mkrcmari__ has joined #openstack-swift | 05:28 | |
| *** links has quit IRC | 05:30 | |
| *** mvk_ has quit IRC | 05:31 | |
| *** links has joined #openstack-swift | 05:33 | |
| *** hoonetorg has quit IRC | 05:34 | |
| *** hoonetorg has joined #openstack-swift | 05:49 | |
| *** kiennt has quit IRC | 05:58 | |
| *** kiennt has joined #openstack-swift | 06:09 | |
| *** hoonetorg has quit IRC | 06:14 | |
| *** Tgrv has quit IRC | 06:15 | |
| *** alenavolk has joined #openstack-swift | 06:20 | |
| *** alenavolk has quit IRC | 06:20 | |
| *** hoonetorg has joined #openstack-swift | 06:21 | |
| *** kiennt has quit IRC | 06:22 | |
| mathiasb | tdasilva: thanks for your comprehensive testing and +2! | 06:29 |
| *** ChubYann has quit IRC | 06:35 | |
| *** rcernin has joined #openstack-swift | 06:45 | |
| *** pcaruana has joined #openstack-swift | 06:45 | |
| *** hseipp has joined #openstack-swift | 06:47 | |
| *** jistr is now known as jistr|trng | 06:53 | |
| *** kiennt has joined #openstack-swift | 06:59 | |
| *** xrb has joined #openstack-swift | 07:04 | |
| *** xrb has quit IRC | 07:04 | |
| *** xrb has joined #openstack-swift | 07:05 | |
| *** cshastri has quit IRC | 07:19 | |
| *** tesseract has joined #openstack-swift | 07:20 | |
| *** cshastri has joined #openstack-swift | 07:21 | |
| *** shuyingya has quit IRC | 07:25 | |
| *** shuyingya has joined #openstack-swift | 07:27 | |
| *** psachin has quit IRC | 07:34 | |
| -openstackstatus- NOTICE: Gerrit is going to be restarted due to slow performance | 07:37 | |
| *** oshritf has joined #openstack-swift | 07:37 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Imported Translations from Zanata https://review.openstack.org/496137 | 07:38 |
| *** geaaru has joined #openstack-swift | 07:41 | |
| -openstackstatus- NOTICE: Gerrit has been restarted successfully | 07:42 | |
| *** skudlik has joined #openstack-swift | 07:57 | |
| *** skudlik has left #openstack-swift | 07:58 | |
| acoles | mathiasb: congratulations! glad your patch has finally merged | 08:03 |
| acoles | good morning | 08:04 |
| mathiasb | good morning | 08:04 |
| mathiasb | acoles: thanks for your feedback and reviews too! | 08:04 |
| acoles | mathiasb: you're welcome - I'm only sorry I wasn't able to join in the final push | 08:05 |
| *** kei_yama has quit IRC | 08:13 | |
| *** kei_yama has joined #openstack-swift | 08:14 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Socket errors don't warrant tracebacks when talking to memcached https://review.openstack.org/496150 | 08:26 |
| *** mvk_ has joined #openstack-swift | 08:39 | |
| *** cbartz has joined #openstack-swift | 08:41 | |
| *** mkrcmari__ has quit IRC | 08:43 | |
| kota_ | acoles: morning | 08:45 |
| kota_ | and I'm on https://review.openstack.org/#/c/495805/ | 08:45 |
| patchbot | patch 495805 - swift - Drop reconstructor stats when worker has no devices | 08:45 |
| kota_ | fyi | 08:45 |
| *** mkrcmari__ has joined #openstack-swift | 08:54 | |
| *** mvk_ has quit IRC | 08:57 | |
| *** mahatic has joined #openstack-swift | 08:59 | |
| *** mvk has joined #openstack-swift | 09:03 | |
| *** mkrcmari__ has quit IRC | 09:06 | |
| *** mahatic has quit IRC | 09:07 | |
| *** mahatic has joined #openstack-swift | 09:08 | |
| acoles | kota_: thanks | 09:22 |
| kota_ | acoles: just pushing my comments :-) | 09:27 |
| *** baojg has joined #openstack-swift | 09:30 | |
| *** baojg has quit IRC | 09:34 | |
| acoles | kota_: looking... | 09:34 |
| kota_ | perhaps, my comments may be overlook rather than just the no device situation... | 09:35 |
| *** cshastri has quit IRC | 09:46 | |
| acoles | kota_: I'm replying on gerrit so others can read too | 09:50 |
| kota_ | ok, thanks acoles | 09:50 |
| *** kei_yama has quit IRC | 09:50 | |
| kota_ | I'll circle back to the patch tomorrow morning again | 09:53 |
| acoles | kota_: posted reply | 09:54 |
| kota_ | so soon! | 09:54 |
| acoles | kota_: have a good evening | 09:54 |
| acoles | kota_: I'm not sure if I perfectly understood your concern so let me know if I did not answer your question | 09:55 |
| acoles | zaitcev: ditto ^^ | 09:55 |
| *** tovin07_ has quit IRC | 10:00 | |
| mahatic | good evening | 10:01 |
| mahatic | I'm back from a vacation and fixed up my bouncer issues | 10:01 |
| kota_ | acoles: quick reading for your worth comment, it would make sense to me that "parent process aggregates the per_disk stats and workers updates per_disk info only (basically)" | 10:05 |
| acoles | mahatic: welcome back! | 10:05 |
| *** kiennt has quit IRC | 10:05 | |
| mahatic | acoles: hello! thanks | 10:06 |
| acoles | kota_: does it make sense to you to have the special case (no devices) when the worker updates the root? | 10:06 |
| acoles | kota_: or are you arguing that should also be done by the parent aggregate_recon_update? | 10:06 |
| kota_ | but, I'd like to have time for a while (tonight), mainly, I'm wondering how can i finalize the reconstruction_time/last in the root node in recon_update with final_recon_dump when run_once case... | 10:07 |
| kota_ | it's definately "with devices" case though | 10:08 |
| acoles | kota_: sure. the case of updating the root when using run_once in worker is an issue, but separate IMHO :) | 10:08 |
| kota_ | with devices, self.all_local_devices will be a list contains some devices | 10:09 |
| kota_ | i see | 10:09 |
| kota_ | tentatively, i pulled out my -1 from the patch but anyway, I'd like to dig the behavior (probably it's from my lack of knowledge on multiple reconstructor process) more to give my +2 | 10:12 |
| kota_ | if zaitcev gets it and feels to add his +2, I'm ok with it. | 10:13 |
| kota_ | while I'm asleep | 10:13 |
| *** silor has joined #openstack-swift | 10:39 | |
| *** bkopilov has quit IRC | 10:56 | |
| *** dancn has quit IRC | 11:00 | |
| *** htruta has quit IRC | 11:00 | |
| *** neonpastor has quit IRC | 11:00 | |
| *** vint_bra has quit IRC | 11:00 | |
| *** StevenK has quit IRC | 11:00 | |
| *** htruta has joined #openstack-swift | 11:01 | |
| *** geaaru has quit IRC | 11:01 | |
| *** Dinesh_Bhor has quit IRC | 11:01 | |
| *** dja_ has quit IRC | 11:01 | |
| *** nikivi has quit IRC | 11:01 | |
| *** etienneme has quit IRC | 11:01 | |
| *** StevenK has joined #openstack-swift | 11:02 | |
| *** vint_bra has joined #openstack-swift | 11:04 | |
| *** Dinesh_Bhor has joined #openstack-swift | 11:05 | |
| *** nikivi has joined #openstack-swift | 11:05 | |
| *** neonpastor has joined #openstack-swift | 11:05 | |
| *** etienneme has joined #openstack-swift | 11:05 | |
| *** geaaru has joined #openstack-swift | 11:06 | |
| *** dja has joined #openstack-swift | 11:06 | |
| *** dancn has joined #openstack-swift | 11:53 | |
| *** mat128 has joined #openstack-swift | 12:07 | |
| *** SkyRocknRoll has quit IRC | 12:29 | |
| *** oshritf has quit IRC | 12:31 | |
| *** oshritf has joined #openstack-swift | 12:32 | |
| *** SkyRocknRoll has joined #openstack-swift | 12:32 | |
| *** psachin has joined #openstack-swift | 12:35 | |
| *** gkadam has quit IRC | 12:36 | |
| *** xrb has quit IRC | 12:37 | |
| *** shuyingya has quit IRC | 12:37 | |
| *** xrb has joined #openstack-swift | 12:37 | |
| *** xrb has quit IRC | 12:38 | |
| *** shuyingya has joined #openstack-swift | 12:38 | |
| *** sanchitmalhotra has quit IRC | 12:38 | |
| *** psachin has quit IRC | 12:39 | |
| *** psachin has joined #openstack-swift | 12:39 | |
| *** shuyingya has quit IRC | 12:42 | |
| *** bkopilov has joined #openstack-swift | 12:49 | |
| *** baojg has joined #openstack-swift | 12:56 | |
| *** zhurong has joined #openstack-swift | 12:56 | |
| *** oshritf has quit IRC | 13:32 | |
| zaitcev | kota_: thanks for the explanations | 13:36 |
| zaitcev | re. 495005 | 13:37 |
| *** shuyingya has joined #openstack-swift | 13:38 | |
| *** oshritf has joined #openstack-swift | 13:41 | |
| *** lucasxu has joined #openstack-swift | 13:54 | |
| *** SkyRocknRoll has quit IRC | 13:58 | |
| *** zhurong has quit IRC | 14:19 | |
| *** links has quit IRC | 14:20 | |
| *** gyee has joined #openstack-swift | 15:11 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:15 | |
| *** rcernin has quit IRC | 15:15 | |
| *** tovin07 has joined #openstack-swift | 15:22 | |
| *** oshritf has quit IRC | 15:27 | |
| *** jarbod has joined #openstack-swift | 15:28 | |
| *** cbartz has quit IRC | 15:28 | |
| notmyname | good morning | 15:37 |
| *** tesseract has quit IRC | 16:04 | |
| openstackgerrit | Merged openstack/swift master: Make test assertion more holistic https://review.openstack.org/495389 | 16:10 |
| *** catintheroof has joined #openstack-swift | 16:10 | |
| openstackgerrit | Merged openstack/swift master: Fix more X-Delete-At timing issues https://review.openstack.org/494756 | 16:15 |
| *** silor has quit IRC | 16:15 | |
| *** tovin07 has quit IRC | 16:16 | |
| *** silor has joined #openstack-swift | 16:26 | |
| *** hseipp has quit IRC | 16:26 | |
| *** pcaruana has quit IRC | 16:37 | |
| *** chsc has joined #openstack-swift | 16:38 | |
| *** chetna has joined #openstack-swift | 16:44 | |
| *** srb has joined #openstack-swift | 16:44 | |
| *** mvk_ has joined #openstack-swift | 16:45 | |
| *** srb has left #openstack-swift | 16:45 | |
| *** srbgupta has joined #openstack-swift | 16:46 | |
| *** shuyingya has quit IRC | 16:47 | |
| *** mvk has quit IRC | 16:48 | |
| srbgupta | Hi, I am trying to validate MD5 of a Static large object in Swift with the original object on disk. ?multipart-manifest=get" is returning different Etag value when compared to MD5 of the object. Any idea how I can compare these? | 16:49 |
| timburke | srbgupta: it'll be a little complicated... since swift never sees all of the bytes for the object all at once, it only knows the MD5s of the pieces. you can do the ?multipart-manifest=get, then walk through the segments yourself -- read "bytes" from the source file, take the MD5, and compare that against the "hash" | 16:58 |
| timburke | python-swiftclient does something like that around https://github.com/openstack/python-swiftclient/blob/3.4.0/swiftclient/service.py#L1175-L1187 | 17:00 |
| srbgupta | If I get all MD5s of the pieces/segments, does it relate to the original object in any way? | 17:08 |
| srbgupta | because if I do a HEAD with ?multipart-manifest=get , as per documentation, it is sum of all the MD5s of the segments | 17:09 |
| srbgupta | but this is not equating with MD5 of original file on disk. | 17:10 |
| notmyname | srbgupta: I'm not sure which file you're referring to, exactly | 17:12 |
| notmyname | the etag returned when reading the large object (*not* the manifest) will be equal to the md5 of the concatenation of the etags of each segment. the etag for each segment is the md5 of the contents of that segment | 17:12 |
| notmyname | so what timburke said about getting the manifest to find the etags for the component segments is the way to validate the large object download. and yes it's complicated | 17:13 |
| srbgupta | @notmyname by file, I meant the original large file which I have on the disk to compare Etag with. | 17:18 |
| srbgupta | thanks guys for help :) | 17:20 |
| *** SkyRocknRoll has quit IRC | 17:21 | |
| notmyname | srbgupta: ah, right. the md5 of that original large file that you have locally won't match the etag of the downloaded large object from swift | 17:23 |
| notmyname | srbgupta: you could split the local file in the same way it is split in the swift cluster and do the same md5 dance and that should match | 17:23 |
| srbgupta | We used swift client to upload with --use-slo and segment-size | 17:24 |
| notmyname | yeah, so if you split the local file by the same segment size, it should match | 17:24 |
| *** geaaru has quit IRC | 17:24 | |
| srbgupta | yeah that would match, but the data is huge like 45TB, and it will be a big overhead to do this for every file. | 17:26 |
| notmyname | yep | 17:26 |
| notmyname | so .. here's what I'd recommend | 17:26 |
| notmyname | if you upload the large object successfully, then swift has validated that the data is correctly saved | 17:26 |
| notmyname | then HEAD the large object and get the etag. store that someplace and used that saved value to compare against future downloads | 17:27 |
| notmyname | there's 2 things I don't like about that solution, though | 17:27 |
| srbgupta | I passed the Etag in header, it works well for small objects but for large objects it returns 422 as it must be comparing against segments. | 17:27 |
| notmyname | first, you can't really send the etag with the upload to validate the local data is the same as what's saved (swift will validate that is saved what it got from the network) | 17:28 |
| notmyname | second, you have to store something locally and keep track of that state | 17:28 |
| srbgupta | Both of your statements are correct | 17:29 |
| notmyname | yeah, that's what I don't like about what I suggested. however, there's not really a better/different way to validate etags for large objects | 17:30 |
| srbgupta | :( | 17:30 |
| notmyname | if I had a checksum/hash algorithm that would allow for combined checksums to equal the checksum of the combined contents, then I could make a better feature for you | 17:32 |
| srbgupta | yeah.. Was looking for something like that. | 17:33 |
| *** klrmn has joined #openstack-swift | 17:33 | |
| notmyname | yeah, me too :-) | 17:33 |
| srbgupta | Ok.. I will post if I will get some better solution for this. | 17:33 |
| notmyname | our of curiosity, what sort of data are you storing in swift? | 17:34 |
| srbgupta | These are the log.gz files, but are very important for future reference | 17:34 |
| *** baojg has quit IRC | 17:37 | |
| notmyname | torgomatic: any better ideas for srbgupta? | 17:37 |
| *** baojg has joined #openstack-swift | 17:38 | |
| notmyname | joeljwright: how's work on https://review.openstack.org/#/c/365371/ going? | 17:49 |
| patchbot | patch 365371 - swift - Add Preamble and Postamble to SLO and SegmentedIte... | 17:49 |
| notmyname | kota_: mahatic: mattoliverau: acoles: I don't see any agenda updates for the 0700 meeting this week, nor do I know who's chairing | 17:52 |
| torgomatic | notmyname: I'd suggest sticking the real checksum in object metadata and comparing that | 17:52 |
| notmyname | which makes me think we should have the 0700 meeting | 17:52 |
| notmyname | torgomatic: that is a very simple and very good idea :-) | 17:52 |
| notmyname | srbgupta: does that make sense? | 17:52 |
| torgomatic | Large-object ETags aren't really useful; they have some semantic meaning, but it's a real pain in the rear to use them effectively. Basically they meet the HTTP requirements for "ETag", but nothing beyond that. | 17:52 |
| acoles | notmyname: did you mean 'should'? | 17:53 |
| acoles | notmyname: I'm probably the only one on your list that is awake | 17:53 |
| torgomatic | Plus, if you do your own checksum storage, you can use whatever checksum(s) you want and you're not bound by what Swift has support for. Depending on what you're doing, this could be handy. | 17:53 |
| notmyname | which makes me think we *shouldn't* have the 0700 meeting | 17:53 |
| notmyname | acoles: yeah, I'm trusting (1) you know what I mean ;-) and (2) bouncer playback | 17:54 |
| notmyname | I can't count on (1), but maybe tech won't fail me :-) | 17:54 |
| acoles | lol | 17:54 |
| acoles | notmyname: I'm just reading log of the last 0700. great job mattoliverau | 17:55 |
| acoles | notmyname: mattoliverau volunteered himself to chair next meeting http://eavesdrop.openstack.org/meetings/swift/2017/swift.2017-08-09-07.01.log.html#l-58 | 17:56 |
| notmyname | oh, I see that :-) | 17:57 |
| acoles | I'll try to get online in time to keep him company IF it happens | 17:58 |
| *** srbgupta has quit IRC | 18:05 | |
| *** itlinux has joined #openstack-swift | 18:06 | |
| *** mvk_ has quit IRC | 18:18 | |
| *** rmcall has joined #openstack-swift | 18:21 | |
| *** itlinux has quit IRC | 18:26 | |
| *** itlinux has joined #openstack-swift | 18:42 | |
| *** ChubYann has joined #openstack-swift | 18:49 | |
| timburke | thinking more about checksums and validations -- i suppose what people ultimately want is a way to verify "all the bytes sent by one client are properly received on a subsequent download" -- a separate checksum in metadata *could* help, but (1) we'll need something like p 336323 before we can *really* trust it and (2) you won't know that your data's bad until someone tries to access it | 19:05 |
| patchbot | https://review.openstack.org/#/c/336323/ - swift - Add checksum to object extended attributes | 19:05 |
| timburke | because ultimately, ^^^ requires three different "validations". there's "swift stored all the bytes the client sent correctly", "over time, the stored representation in swift is still correct", and "when a client downloads this, they can verify that it matches what swift sent" | 19:05 |
| timburke | that middle one's important! | 19:05 |
| *** itlinux has quit IRC | 19:21 | |
| timburke | p 212824 might be able to help with it -- maybe put a bunch of (SLO, segment data) pairs in a queue with a bunch of workers that head the segments and quarantine (maybe? maybe that's too heavy handed?) the SLO if there's a failure | 19:25 |
| patchbot | https://review.openstack.org/#/c/212824/ - swift - Let developers/operators add watchers to object audit | 19:25 |
| timburke | basically, notmyname's instinct was spot-on -- torgomatic will save us all ;-) | 19:27 |
| timburke | this did lead me to noticing that zlib has a crc32_combine function, though, so that's cool https://github.com/madler/zlib/blob/v1.2.11/crc32.c#L428 | 19:30 |
| *** srbgupta has joined #openstack-swift | 19:32 | |
| *** silor1 has joined #openstack-swift | 19:33 | |
| *** silor has quit IRC | 19:35 | |
| *** silor1 is now known as silor | 19:35 | |
| *** itlinux has joined #openstack-swift | 19:47 | |
| notmyname | timburke: but in your 3 different validations, the first and third can be grouped and the second treated separately | 19:54 |
| timburke | notmyname: kinda. but sending x-object-meta-original-sha512 or some such is only useful for the third one; swift just sees meta to store during the PUT | 19:56 |
| notmyname | yes (today). I'm only stating that we don't have to solve all 3 for every object at the same time. we maybe could support something that would do 1 and 3 but leave 2 for later | 19:57 |
| notmyname | but I still don't have any idea how to solve 1 in the large object case | 19:57 |
| notmyname | that is, in the case of objects that are manifests | 19:58 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Add tests for memcache logging change https://review.openstack.org/496392 | 20:04 |
| *** silor has quit IRC | 20:07 | |
| torgomatic | man, what happened to patch 336323? it needs a rebase, but I don't recognize half the stuff in there, so rebasing gets confusing to me :| | 20:07 |
| patchbot | https://review.openstack.org/#/c/336323/ - swift - Add checksum to object extended attributes | 20:07 |
| *** baojg has quit IRC | 20:09 | |
| timburke | main thing i remember about it is that it prompted us to have the split between gate-swift-python27-ubuntu-xenial and gate-swift-tox-xfs-tmp-py27-ubuntu-xenial | 20:09 |
| notmyname | yeah, and saio should have and xfs tmpdir too | 20:09 |
| notmyname | but all that is working now | 20:09 |
| *** baojg has joined #openstack-swift | 20:09 | |
| *** cschwede_ has quit IRC | 20:27 | |
| *** rmcall has quit IRC | 20:30 | |
| *** rmcall has joined #openstack-swift | 20:31 | |
| *** lucasxu has quit IRC | 20:31 | |
| *** psachin has quit IRC | 20:34 | |
| *** baojg has quit IRC | 20:38 | |
| *** baojg has joined #openstack-swift | 20:38 | |
| *** rcernin has joined #openstack-swift | 20:50 | |
| *** itlinux has quit IRC | 20:59 | |
| clayg | timburke: i don't care what the tests say - just that they blow up when I break the change - have at it | 21:00 |
| clayg | FWIW the debug logger doesn't include tracebacks - just the message that goes with the traceback | 21:01 |
| clayg | the memcache vs memcached was me trying to validate it broke and getting bored/careless | 21:01 |
| clayg | also $%^& virtualbox networking | 21:01 |
| *** itlinux has joined #openstack-swift | 21:06 | |
| tdasilva | heh, was just looking at those patches | 21:06 |
| tdasilva | guess will wait for timburke | 21:06 |
| *** mat128 has quit IRC | 21:15 | |
| *** mat128 has joined #openstack-swift | 21:16 | |
| *** mat128 has quit IRC | 21:17 | |
| timburke | `FAILED (SKIP=14, failures=61)` oh man, what a yak! it's gonna be great though. looking forward to being able to disambiguate between logging.error and logging.exception in tests :D | 21:28 |
| clayg | don't do it | 21:29 |
| clayg | don't do it | 21:29 |
| clayg | don't do it | 21:29 |
| timburke | clayg: hey, you wanted testts... coulda just +A'ed :P | 21:29 |
| clayg | ok | 21:30 |
| clayg | try to keep orthogonal changes separate - adding infrastructure to test logging in the tests for the module where you changed logging; great - reworking a bunch of unrelated test infrastructure can hopefully be on a different non-blocking review/priority cycle | 21:32 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Socket errors don't warrant tracebacks when talking to memcached https://review.openstack.org/496150 | 21:35 |
| *** catintheroof has quit IRC | 22:01 | |
| *** itlinux has quit IRC | 22:03 | |
| *** baojg has quit IRC | 22:11 | |
| *** baojg has joined #openstack-swift | 22:12 | |
| *** baojg has quit IRC | 22:20 | |
| *** baojg has joined #openstack-swift | 22:20 | |
| *** baojg has quit IRC | 22:25 | |
| *** baojg has joined #openstack-swift | 22:25 | |
| notmyname | FYI this is my response to the openstack marketing people about "what is new in pike": https://gist.github.com/notmyname/5fa8ca9cf0857ec2f9bdedc44c348e68 | 22:26 |
| *** baojg has quit IRC | 22:30 | |
| *** baojg has joined #openstack-swift | 22:30 | |
| *** rcernin has quit IRC | 22:32 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Replicator timeouts don't deserve tracebacks https://review.openstack.org/496433 | 22:48 |
| notmyname | timburke: traceback are for replicators that don't time out? | 22:49 |
| timburke | gd right they are | 22:49 |
| *** itlinux has joined #openstack-swift | 22:54 | |
| *** srbgupta has quit IRC | 22:58 | |
| *** baojg has quit IRC | 23:05 | |
| mattoliverau | morning | 23:06 |
| mattoliverau | acoles: thanks, please come it was lonely last time :P | 23:06 |
| *** baojg has joined #openstack-swift | 23:06 | |
| *** vint_bra has quit IRC | 23:18 | |
| *** mat128 has joined #openstack-swift | 23:19 | |
| *** baojg has quit IRC | 23:22 | |
| notmyname | mattoliverau: so are you gonna chair? is there a need for a meeting? nothings been added to the agenda | 23:22 |
| notmyname | I'd propose skipping this one, and then there is one more 0700 meeting right before the ptg | 23:22 |
| *** chetna has quit IRC | 23:23 | |
| *** baojg has joined #openstack-swift | 23:23 | |
| mattoliverau | I'm fine with that.. I added info to the agenda but nothing more interesting then letting people know things | 23:23 |
| *** mat128 has quit IRC | 23:24 | |
| notmyname | there's only one thing that'd I'd like some follow-up on, and that's joeljwright's patch 365371 | 23:24 |
| patchbot | https://review.openstack.org/#/c/365371/ - swift - Add Preamble and Postamble to SLO and SegmentedIte... | 23:24 |
| notmyname | and aside from that, the only reason to have an 0700 or 2100 meeting is for info update. so I'd propose skipping | 23:25 |
| notmyname | (but I'm waiting to hear from joeljwright ) | 23:25 |
| mattoliverau | kk, I'll wait until and see if mahatic and kota_ turn up and see if there ok with skipping it then.. But yeah I'm fine with it. I just gathered info from emails over the last week, so really everyone should have already read em ;) | 23:27 |
| *** kei_yama has joined #openstack-swift | 23:27 | |
| notmyname | yep. sounds good. thanks | 23:27 |
| mattoliverau | tdasilva: I've got version 1 of statsd/collectd/influxdb/grafana playbook working (well should work, I'll test it somemore today), working for setting up the statsd server in swift-ansible. Once I test it again, I'll PR over to your repo. | 23:31 |
| mattoliverau | notmyname: the playbook probably work great for runway as well :) ^^ | 23:31 |
| mattoliverau | I guess my next step is to see if I can even run runway on OpenSuse, and if not port it. The nodes can happily be whatever, so ubuntu. Just the running host. | 23:34 |
| *** chsc has quit IRC | 23:38 | |
| *** baojg has quit IRC | 23:40 | |
| notmyname | mattoliverau: cool! | 23:40 |
| *** baojg has joined #openstack-swift | 23:41 | |
| *** baojg has quit IRC | 23:45 | |
| *** baojg has joined #openstack-swift | 23:46 | |
| *** itlinux has quit IRC | 23:47 | |
| *** baojg has quit IRC | 23:50 | |
| *** baojg has joined #openstack-swift | 23:52 | |
| *** gyee has quit IRC | 23:53 | |
| *** gyee has joined #openstack-swift | 23:57 | |
| *** baojg has quit IRC | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!