| *** manous has quit IRC | 00:00 | |
| *** darrenc is now known as darren_afk | 00:09 | |
| *** natarej has quit IRC | 00:16 | |
| *** natarej has joined #openstack-swift | 00:16 | |
| *** klamath has joined #openstack-swift | 00:17 | |
| *** darren_afk is now known as darrenc | 00:23 | |
| *** klamath has quit IRC | 00:26 | |
| kota_ | hello, back to online. | 00:47 |
|---|---|---|
| *** itlinux has joined #openstack-swift | 00:48 | |
| kota_ | jrichli: thanks for sharing simlink etherpad, will see. | 00:48 |
| mattoliverau | kota_: morning | 00:48 |
| kota_ | mattoliverau: morning | 00:48 |
| mattoliverau | kota_: hows bub and mum doing? | 00:48 |
| kota_ | mattoliverau: everything is well, my parents is taking care of them in day time instead of me ) | 00:49 |
| kota_ | thanks! | 00:49 |
| mattoliverau | kota_: nice, well glad to have you back then :) | 00:51 |
| kota_ | mattoliverau: thanks :) | 00:51 |
| *** catintheroof has joined #openstack-swift | 00:55 | |
| *** itlinux has quit IRC | 00:57 | |
| *** mingdang1 has joined #openstack-swift | 02:19 | |
| *** itlinux has joined #openstack-swift | 02:33 | |
| *** amit213 has quit IRC | 02:34 | |
| *** amit213 has joined #openstack-swift | 02:34 | |
| *** sorrison_laptop has quit IRC | 03:05 | |
| *** sheel has joined #openstack-swift | 03:14 | |
| *** dmorita has joined #openstack-swift | 03:26 | |
| *** nadeem has joined #openstack-swift | 03:27 | |
| *** furlongm has joined #openstack-swift | 03:28 | |
| *** dmorita has quit IRC | 03:34 | |
| *** links has joined #openstack-swift | 03:44 | |
| *** klrmn has quit IRC | 03:53 | |
| *** catintheroof has quit IRC | 03:56 | |
| *** takashi has joined #openstack-swift | 04:08 | |
| *** silor has joined #openstack-swift | 04:16 | |
| *** psachin has joined #openstack-swift | 04:27 | |
| *** mingdang1 has quit IRC | 04:58 | |
| *** psachin has quit IRC | 05:18 | |
| *** ChubYann has quit IRC | 05:29 | |
| *** itlinux has quit IRC | 05:31 | |
| *** ppai has joined #openstack-swift | 05:34 | |
| *** rcernin has joined #openstack-swift | 05:40 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:47 | |
| *** nadeem has quit IRC | 06:24 | |
| *** jmccarthy has joined #openstack-swift | 06:29 | |
| *** nadeem has joined #openstack-swift | 06:36 | |
| *** mingdang1 has joined #openstack-swift | 06:39 | |
| *** openstackgerrit has quit IRC | 06:47 | |
| *** openstackgerrit has joined #openstack-swift | 06:48 | |
| *** tesseract has joined #openstack-swift | 07:01 | |
| *** mmcardle has joined #openstack-swift | 07:12 | |
| *** rledisez has joined #openstack-swift | 07:12 | |
| *** ntt has joined #openstack-swift | 07:35 | |
| ntt | Hi, I have a test environment with swift (kilo). Actually I have replica=2 with 2 zones and I'm trying wo verify quorum on writes. I turned off one zone and I tried to upload an object .... and it works! how is this possible? | 07:37 |
| ntt | quorum should be 2 because Q = (replicas / 2) + 1 | 07:38 |
| hosanai | ntt: i think it uses a hand-off node | 07:41 |
| ntt | hosanai: ok.... | 07:44 |
| *** daemontool has joined #openstack-swift | 07:45 | |
| ntt | Checked... ok... it seems right. Thank you hosanai | 07:47 |
| mattoliverau | ntt: hosanai is write, swift still wants to keep your data durable so if it can't talk to the other primary node, it'll pick another node in the cluster to store it (a handoff node). And backend replication will place it on the write handoff node once it's back online. | 07:48 |
| mattoliverau | s/write/right | 07:48 |
| ntt | mattoliverau: checked now... and it seems ok. Thank you | 07:49 |
| mattoliverau | My brain says right.. but fingers type write.. I've been programming too long :P | 07:49 |
| ntt | Actually I'm trying to count objects between 2 zones ... I have 464 folders in objects folder in Z1 and 458 in Z2.... Is this normal? | 07:50 |
| ntt | this is due to handoff? | 07:50 |
| ntt | ops... no ... I'm sorry | 07:51 |
| ntt | It's ok | 07:51 |
| ntt | mattoliverau: Count the number of folders in objects is a good or a stupid check? | 07:51 |
| ahale | counting out-of-place partitions might make more sense | 07:55 |
| ntt | ahale I'm using ls -alh /srv/node/sdX1/account | wc -l where X = b,c,d,.... | 07:56 |
| ahale | oh yeah thats not what i was thinking you were asking :) | 07:57 |
| ahale | sometimes i use something like this- https://gist.github.com/racker-andrew-hale/04d78e73f92d71c20cefec71f5561e5a | 07:57 |
| ahale | or finding if handoff happened for a specific put, doesnt proxy log that | 07:59 |
| ntt | interesting.... thank you | 07:59 |
| ntt | I have to add 1 zone with 4 new disks and +1 replica to a production (kilo) cluster. Can I modify the ring and add all using just one rebalance? I know that this will cause a lot of copy on the new server, but data are only 800GB and the new server is fast enough | 08:01 |
| *** jordanP has joined #openstack-swift | 08:01 | |
| ahale | oh i also have a 'handoff analysis' tool , that'll take a ring and a list of down host ips and show you where handoff would go for every part, i can find that if you're interested | 08:01 |
| *** joeljwright has joined #openstack-swift | 08:01 | |
| *** ChanServ sets mode: +v joeljwright | 08:01 | |
| ahale | maybe - i dont know what the implications of adding capacity and replicas at the same time is | 08:03 |
| ntt | ahale: I think that every thing that con simplify the sysadmin life is welcome :) | 08:03 |
| ahale | i dont know - sometimes knowing more about the ring doesnt simplify life ;) | 08:04 |
| ntt | but if I have to add a replica, I have to add one replica and at least one disk in a separate zone.... otherwise I write datas on the same zone and I think that is not so useful | 08:04 |
| ahale | yeah | 08:06 |
| ntt | you're right .... sometimes knowing more abount the ring could be worse :) | 08:06 |
| ntt | This morning I will give a try on a test environment where I have 1 zone , 1 replica and 2 disks. I will try to add 1 zone, 1 replica and 2 disks using one rebalance | 08:07 |
| ntt | The alternative could be to add 1 zone, 1 replica and 1 disk using weight gradually | 08:07 |
| ntt | I think if I forget to add simultaneously 1 zone and 1 replica when I add the disk, datas will be distributed over the new disk in the second zone.... is this right? | 08:08 |
| *** takashi has quit IRC | 08:11 | |
| ahale | ah found that script for looking at handoff, its at the bottom of https://gist.github.com/ahale/b91fc9bc745e710e070a | 08:11 |
| ahale | yeah i would expect it to just add 1 new replica to the new set of disks, that would make sense | 08:12 |
| ahale | and 800G total probably isnt that much to need adding weight gradually, i dunno - i prefer larger single ring changes than lots of small ones anyway | 08:13 |
| *** nadeem has quit IRC | 08:14 | |
| ntt | ahale: thank you..... I'm doing test adding 2 disks using one rebalance | 08:15 |
| *** openstackgerrit has quit IRC | 08:17 | |
| *** Shashikant86 has joined #openstack-swift | 08:18 | |
| *** openstackgerrit has joined #openstack-swift | 08:18 | |
| ntt | ahale: I'm trying to remove 1 zone , 1 replica and 1 disk from the cluster (before trying to add 2 disks). I'm using 1) set_weight 0, set_replicas 1 (was2) and rebalance. For accounts and containers all works well, but with objects it tells me that no partitions could be reassigned | 08:19 |
| ahale | hmm not sure, wasnt there a bug about that recently ? | 08:19 |
| ahale | also - pm'd you another ring thing | 08:20 |
| *** daemontool has quit IRC | 08:21 | |
| ntt | actually I have: 1024 partitions, 1.000000 replicas, 1 regions, 2 zones, 3 devices, 999.99 balance, 0.00 dispersion | 08:21 |
| ntt | but rebalance tells me that not partitions could be reassigned | 08:21 |
| ntt | I used pretend_min_part_hours | 08:21 |
| *** Shashikant86 has quit IRC | 08:32 | |
| *** jistr has joined #openstack-swift | 08:32 | |
| *** natarej has quit IRC | 08:33 | |
| *** natarej has joined #openstack-swift | 08:33 | |
| *** jamielennox is now known as jamielennox|away | 08:40 | |
| *** cbartz has joined #openstack-swift | 09:02 | |
| *** kei_yama_ has joined #openstack-swift | 09:09 | |
| *** kei_yama has quit IRC | 09:09 | |
| *** acoles_ is now known as acoles | 09:12 | |
| *** Jeffrey4l_ has quit IRC | 09:16 | |
| *** takashi has joined #openstack-swift | 09:17 | |
| *** Jeffrey4l_ has joined #openstack-swift | 09:23 | |
| *** mvk has joined #openstack-swift | 09:23 | |
| *** psachin has joined #openstack-swift | 09:29 | |
| *** daemontool has joined #openstack-swift | 09:45 | |
| *** admin6 has joined #openstack-swift | 09:46 | |
| *** asettle has joined #openstack-swift | 09:54 | |
| *** ppai has quit IRC | 10:06 | |
| *** silor has quit IRC | 10:07 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Rename hash_cleanup_listdir tests https://review.openstack.org/290391 | 10:10 |
| *** daemontool_ has joined #openstack-swift | 10:11 | |
| *** daemontool has quit IRC | 10:14 | |
| *** ppai has joined #openstack-swift | 10:15 | |
| *** hosanai has quit IRC | 10:17 | |
| *** asettle_ has joined #openstack-swift | 10:22 | |
| *** asettle has quit IRC | 10:24 | |
| *** daemontool__ has joined #openstack-swift | 10:31 | |
| *** asettle_ has quit IRC | 10:31 | |
| openstackgerrit | janonymous proposed openstack/swift: Unable to read test.conf while executing unit tests https://review.openstack.org/314041 | 10:31 |
| *** daemontool_ has quit IRC | 10:34 | |
| *** daemontool has joined #openstack-swift | 10:36 | |
| *** takashi has quit IRC | 10:37 | |
| *** daemontool__ has quit IRC | 10:37 | |
| acoles | timburke: clayg: thanks for reviews, patch 290391 will need a fresh +A (merge conflicts - see my comment on patch) | 10:39 |
| patchbot | acoles: https://review.openstack.org/#/c/290391/ - swift - Rename hash_cleanup_listdir tests | 10:39 |
| *** asettle has joined #openstack-swift | 10:40 | |
| *** daemontool_ has joined #openstack-swift | 10:40 | |
| *** daemontool__ has joined #openstack-swift | 10:43 | |
| *** daemontool has quit IRC | 10:44 | |
| *** mariusv has joined #openstack-swift | 10:45 | |
| *** daemontool_ has quit IRC | 10:46 | |
| *** dmellado|off is now known as dmellado | 10:47 | |
| *** daemontool_ has joined #openstack-swift | 10:51 | |
| *** daemontool__ has quit IRC | 10:55 | |
| *** asettle has quit IRC | 10:58 | |
| *** daemontool__ has joined #openstack-swift | 10:59 | |
| *** daemontool_ has quit IRC | 11:03 | |
| *** daemontool has joined #openstack-swift | 11:04 | |
| *** ppai has quit IRC | 11:04 | |
| *** daemontool_ has joined #openstack-swift | 11:06 | |
| *** daemontool__ has quit IRC | 11:07 | |
| *** daemontool__ has joined #openstack-swift | 11:08 | |
| *** daemontool has quit IRC | 11:10 | |
| *** daemontool has joined #openstack-swift | 11:11 | |
| *** daemontool_ has quit IRC | 11:12 | |
| *** daemontool_ has joined #openstack-swift | 11:14 | |
| *** daemontool__ has quit IRC | 11:15 | |
| *** daemontool has quit IRC | 11:18 | |
| *** ppai has joined #openstack-swift | 11:18 | |
| *** kei_yama_ has quit IRC | 11:45 | |
| *** kei_yama has joined #openstack-swift | 11:46 | |
| openstackgerrit | janonymous proposed openstack/swift: Using ostestr to run unittests https://review.openstack.org/257053 | 11:49 |
| *** silor has joined #openstack-swift | 12:02 | |
| *** silor has quit IRC | 12:04 | |
| *** silor has joined #openstack-swift | 12:05 | |
| openstackgerrit | janonymous proposed openstack/swift: Using ostestr to run unittests https://review.openstack.org/257053 | 12:07 |
| *** pauloewerton has joined #openstack-swift | 12:08 | |
| *** early` has quit IRC | 12:14 | |
| *** ppai has quit IRC | 12:34 | |
| *** daemontool__ has joined #openstack-swift | 12:41 | |
| *** early has joined #openstack-swift | 12:43 | |
| *** daemontool_ has quit IRC | 12:45 | |
| psachin | Hi, can we re-configure replication time? | 12:45 |
| *** bsdkurt has quit IRC | 12:46 | |
| *** bsdkurt has joined #openstack-swift | 12:49 | |
| *** zul has quit IRC | 12:50 | |
| *** chuck__ has joined #openstack-swift | 12:51 | |
| *** chuck__ has quit IRC | 12:51 | |
| *** chuck__ has joined #openstack-swift | 12:52 | |
| *** km has quit IRC | 12:54 | |
| acoles | psachin: replicator config options are shown in [object-replicator] section here https://github.com/openstack/swift/blob/master/etc/object-server.conf-sample | 12:55 |
| *** kei_yama has quit IRC | 13:02 | |
| *** links has quit IRC | 13:02 | |
| *** SkyRocknRoll has quit IRC | 13:05 | |
| * jrichli will be out for the next 5 hours. | 13:07 | |
| psachin | thanx acoles | 13:14 |
| openstackgerrit | Brian Ober proposed openstack/swift: Container Object Count Quota Functional Test https://review.openstack.org/314099 | 13:29 |
| *** david-lyle_ has joined #openstack-swift | 13:48 | |
| *** david-lyle has quit IRC | 13:50 | |
| *** cdelatte has joined #openstack-swift | 13:53 | |
| *** ametts has joined #openstack-swift | 13:53 | |
| *** klamath has joined #openstack-swift | 13:57 | |
| *** klamath has quit IRC | 13:58 | |
| *** klamath has joined #openstack-swift | 13:59 | |
| *** zaitcev has joined #openstack-swift | 13:59 | |
| *** ChanServ sets mode: +v zaitcev | 13:59 | |
| *** catintheroof has joined #openstack-swift | 14:03 | |
| *** d0ugal has quit IRC | 14:07 | |
| *** links has joined #openstack-swift | 14:08 | |
| *** links has quit IRC | 14:11 | |
| *** vinsh has joined #openstack-swift | 14:13 | |
| *** mingdang1 has quit IRC | 14:21 | |
| *** pgbridge has joined #openstack-swift | 14:38 | |
| *** _JZ_ has joined #openstack-swift | 14:45 | |
| *** natarej has quit IRC | 14:46 | |
| *** psachin has quit IRC | 14:46 | |
| *** natarej has joined #openstack-swift | 14:47 | |
| *** d0ugal has joined #openstack-swift | 14:55 | |
| *** esker has joined #openstack-swift | 15:03 | |
| *** d0ugal has quit IRC | 15:08 | |
| *** klrmn has joined #openstack-swift | 15:11 | |
| *** klamath has quit IRC | 15:15 | |
| openstackgerrit | Brian Ober proposed openstack/swift: Storage Account Object Count Quota Support https://review.openstack.org/312603 | 15:17 |
| *** dmorita has joined #openstack-swift | 15:18 | |
| *** arch-nemesis has joined #openstack-swift | 15:20 | |
| *** klamath has joined #openstack-swift | 15:20 | |
| *** d0ugal has joined #openstack-swift | 15:26 | |
| *** jistr has quit IRC | 15:33 | |
| *** Jeffrey4l_ has quit IRC | 15:34 | |
| *** daemontool__ has quit IRC | 15:41 | |
| *** d0ugal has quit IRC | 15:51 | |
| *** morgan has joined #openstack-swift | 15:52 | |
| morgan | notmyname timburke: https://review.openstack.org/#/c/310892/ as discussed at the summit. just following up on it | 15:52 |
| patchbot | morgan: patch 310892 - swift3 - Add s3_token middleware to the swift3 project | 15:52 |
| *** jistr has joined #openstack-swift | 15:54 | |
| admin6 | Hi there, Is that a normal behavior to have a lot of STDERR: UnicodeDecodeError in the object-replicator logs after rebalancing my replica3 object ring and it last until it is completely rebalanced (more logs in http://pastebin.com/nUprVRrk) | 15:58 |
| *** tesseract has quit IRC | 16:01 | |
| *** rledisez has quit IRC | 16:05 | |
| *** rcernin has quit IRC | 16:06 | |
| *** d0ugal has joined #openstack-swift | 16:06 | |
| notmyname | good morning, world | 16:06 |
| notmyname | morgan: thanks. probably best to put that into the #swift3 channel (there's a channel for everything!) | 16:07 |
| *** ejat is now known as fenris- | 16:07 | |
| *** fenris- is now known as ejat | 16:07 | |
| *** ejat is now known as fenris- | 16:08 | |
| *** fenris- is now known as ejat | 16:08 | |
| *** NM has joined #openstack-swift | 16:19 | |
| *** klrmn has quit IRC | 16:20 | |
| *** gyee has joined #openstack-swift | 16:22 | |
| openstackgerrit | Mohit Motiani proposed openstack/swift: ng chkconfig to sysv-rc-conf in SAIO installation doc: Below step in SAIO installation always throws an error: sudo chkconfig memcached on https://review.openstack.org/314200 | 16:24 |
| *** esker has quit IRC | 16:29 | |
| clayg | lol @ "sometimes knowing more abount the ring could be worse" | 16:29 |
| mmotiani_ | notmyname, I was looking at this patch https://review.openstack.org/#/c/309219/ | 16:30 |
| patchbot | mmotiani_: patch 309219 - swift - created other-requirements.txt | 16:30 |
| notmyname | ok | 16:30 |
| mmotiani_ | and I was wondering if sysv-rc-conf (an alternative of chckconfig) is also a non-python requirement. | 16:30 |
| notmyname | I honestly don't know | 16:32 |
| ahale | mmotiani_: isnt that a rhel command you are trying to run on ubuntu? | 16:32 |
| ahale | like immediately before the chkconfig command it says "On non-Ubuntu distros you need to ensure memcached is running" | 16:32 |
| *** lyrrad has joined #openstack-swift | 16:33 | |
| *** cbartz has quit IRC | 16:34 | |
| clayg | isn't xenial going systemd too? | 16:34 |
| *** SkyRocknRoll has joined #openstack-swift | 16:35 | |
| *** dmorita has quit IRC | 16:36 | |
| *** rcernin has joined #openstack-swift | 16:37 | |
| mmotiani_ | ahale, I thought that patch is for all Linux machines :) | 16:38 |
| *** diogogmt has joined #openstack-swift | 16:38 | |
| mmotiani_ | chkconfig was also an ubuntu command till ubuntu 12.01. Now, its replacement is sysv-rc-conf :) | 16:39 |
| ahale | oh live and learn :) never bothered using it on debuntu, just update-rc | 16:40 |
| notmyname | yeah, we can put non-ubuntu stuff in other-requirements.txt. it's got some DSL for describing that | 16:41 |
| briancline | it's too early on a monday for systemd | 16:42 |
| ahale | yeah - im about to leave work for the day and its still too early for that | 16:42 |
| briancline | lol | 16:43 |
| notmyname | so yeah, if there's something missing from other-requirements, then it should be added | 16:45 |
| notmyname | but there's gotta be a limit somewhere | 16:45 |
| *** david-lyle_ is now known as david-lyle | 16:45 | |
| notmyname | it's not like we're saying "must have libc and a linux kernel" | 16:45 |
| notmyname | so if it's stuff that's in a base install, I don't think we need to add it. but I also don't really know the definition of "base" | 16:46 |
| *** jordanP has quit IRC | 16:51 | |
| *** nadeem has joined #openstack-swift | 16:54 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - cleanup decrypter exception handling https://review.openstack.org/304806 | 17:02 |
| *** jistr has quit IRC | 17:02 | |
| *** daemontool__ has joined #openstack-swift | 17:09 | |
| notmyname | tdasilva: are you working on a follow-up for https://review.openstack.org/#/c/156923/ today? or is timburke wrong and you're looking for another reviewer? ;-) | 17:10 |
| patchbot | notmyname: patch 156923 - swift - Refactor server side copy as middleware | 17:10 |
| *** pgbridge has quit IRC | 17:10 | |
| tdasilva | working for a follow-up for today | 17:13 |
| notmyname | thanks | 17:14 |
| tdasilva | np, just trying to figure out the logging issue | 17:14 |
| notmyname | sgundur-: what's the WIP flag on patch 313150 for? | 17:17 |
| patchbot | notmyname: https://review.openstack.org/#/c/313150/ - swift - To fix probe tests failing from commit cf48e75 | 17:17 |
| *** klrmn has joined #openstack-swift | 17:17 | |
| sgundur- | notmyname: there are still some references to ports in 6000 range , so was wondering should I add all these changes in this patch or not | 17:19 |
| notmyname | at this very moment, I don't really care. what I care very much about is that probetests work. either or both or whatever of patch 313150 and patch 311899 need to land, preferably today | 17:21 |
| patchbot | notmyname: https://review.openstack.org/#/c/313150/ - swift - To fix probe tests failing from commit cf48e75 | 17:21 |
| patchbot | notmyname: https://review.openstack.org/#/c/311899/ - swift - Fix probe tests from commit cf48e75 | 17:21 |
| notmyname | https://wiki.openstack.org/wiki/Swift/PriorityReviews is updated | 17:21 |
| notmyname | acoles: jrichli: ^ I wasn't sure of the current top crypto patch, so I left it as-is (although i removed the reference to the already landed one) | 17:22 |
| sgundur- | notmyname: ok, with either of the above patches the probetests failing because of port change is fixed - but getting this error http://paste.openstack.org/show/496139/ | 17:24 |
| *** klrmn has quit IRC | 17:24 | |
| acoles | notmyname: ok thanks. looking... | 17:25 |
| sgundur- | notmyname: in the test_container_sync:TestContainerSync._test_sync ; put_object returns with 201 but after the container-sync the get_object fails with 404 not found | 17:25 |
| sgundur- | notmyname: I ll try to debug this error today | 17:25 |
| notmyname | sgundur-: ok. please let me know if you need help | 17:25 |
| sgundur- | notmyname: yep, I ll. thanks :) | 17:26 |
| *** klrmn has joined #openstack-swift | 17:26 | |
| acoles | notmyname: I added one more. the next I'd like to land is patch 304806 which is large (lots of cleanup) but timburke has been reviewing so I'm hoping we can get that donean d if there are spare eyes focus them on those listed on priority reviews | 17:29 |
| patchbot | acoles: https://review.openstack.org/#/c/304806/ - swift (feature/crypto) - crypto - cleanup decrypter exception handling | 17:29 |
| *** dmorita has joined #openstack-swift | 17:34 | |
| acoles | timburke: re your comment here https://review.openstack.org/#/c/312662/2/swift/common/middleware/encrypter.py@203 - slo appends size to content-type but i can't see etag there? am I missing it? | 17:34 |
| patchbot | acoles: patch 312662 - swift (feature/crypto) - crypto - cleanup encrypter error handling | 17:34 |
| *** pgbridge has joined #openstack-swift | 17:36 | |
| *** mmcardle has quit IRC | 17:39 | |
| *** lakshmiS has joined #openstack-swift | 17:42 | |
| notmyname | acoles: in here you said patch 304806 but you added patch 313640 to the wiki | 17:42 |
| patchbot | notmyname: https://review.openstack.org/#/c/304806/ - swift (feature/crypto) - crypto - cleanup decrypter exception handling | 17:42 |
| patchbot | notmyname: https://review.openstack.org/#/c/313640/ - swift (feature/crypto) - crypto - wrap random body key | 17:42 |
| acoles | notmyname: yes. my thinking was that since timburke (and jrichli) is already on 304806 then it may not need to be on the priority list | 17:44 |
| notmyname | ok :-) | 17:44 |
| openstackgerrit | Nandini Tata proposed openstack/swift: add Tempest Plugin for functional tests https://review.openstack.org/313676 | 17:47 |
| *** garthb has joined #openstack-swift | 17:51 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - cleanup encrypter error handling https://review.openstack.org/312662 | 17:53 |
| *** acoles is now known as acoles_ | 17:57 | |
| *** ChubYann has joined #openstack-swift | 17:59 | |
| *** mvk has quit IRC | 18:12 | |
| *** CaioBrentano has joined #openstack-swift | 18:15 | |
| *** openstackgerrit has quit IRC | 18:33 | |
| *** openstackgerrit has joined #openstack-swift | 18:33 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift: Refactor server side copy as middleware https://review.openstack.org/156923 | 18:35 |
| pdardeau | notmyname: i vaguely remember seeing some kind of autodoc comment in the new contributor doc reviews. now i don't see it. was i dreaming? | 18:35 |
| *** jmccarthy1 has joined #openstack-swift | 18:37 | |
| *** ozialien10 has quit IRC | 18:43 | |
| *** ozialien10 has joined #openstack-swift | 18:44 | |
| *** ozialien10 has quit IRC | 18:44 | |
| *** tqtran has joined #openstack-swift | 18:45 | |
| *** ozialien10 has joined #openstack-swift | 18:45 | |
| *** portante has quit IRC | 18:46 | |
| *** dmellado has quit IRC | 18:47 | |
| *** ndk_ has quit IRC | 18:47 | |
| *** natarej has quit IRC | 18:55 | |
| *** natarej has joined #openstack-swift | 18:56 | |
| *** silor has quit IRC | 19:00 | |
| *** garthb has quit IRC | 19:07 | |
| *** SkyRocknRoll has quit IRC | 19:32 | |
| *** jmccarthy1 has quit IRC | 19:34 | |
| notmyname | pdardeau: that was probably on the "movie script" page that i took out. vSAIO has autodoc set up | 19:50 |
| *** kaleta has quit IRC | 19:51 | |
| *** onovy has quit IRC | 19:51 | |
| *** peterlisak has quit IRC | 19:51 | |
| *** onovy has joined #openstack-swift | 19:53 | |
| *** onovy has quit IRC | 19:53 | |
| *** onovy has joined #openstack-swift | 19:53 | |
| *** peterlisak has joined #openstack-swift | 19:56 | |
| *** kaleta has joined #openstack-swift | 19:56 | |
| sgundur- | notmyname: I checkedout from one commit previous to port changes i.e. https://github.com/openstack/swift/commit/ae2b86129e1606b8dbdae45a0074e5b9add13c1a | 20:11 |
| sgundur- | notmyname: the probetests still fail with - http://paste.openstack.org/show/496139/ | 20:11 |
| sgundur- | notmyname: I was wondering if the probetests for - container-sync supposed to work in SAIO setup ? | 20:11 |
| notmyname | sgundur-: supposed to work? yes. do they work? seems like maybe not? | 20:14 |
| sgundur- | notmyname: in /srv/1/node/sdb1 ; sync_containers is created and also the .db file is symlinked to /srv/..containers... .db | 20:15 |
| sgundur- | notmyname: I can trace till the object being put in the source containers, returns with 201 | 20:15 |
| *** rcernin has quit IRC | 20:15 | |
| sgundur- | notmyname: after the Manager(['container-sync']).once() ; get object from dest_container is failing with 404 | 20:16 |
| sgundur- | notmyname: I was trying to find if the error was in src/dest containers being created or being pointed to wrong container/object | 20:18 |
| sgundur- | notmyname: pls give me some pointers to debug or what to look for | 20:20 |
| *** klamath has quit IRC | 20:22 | |
| jrichli | sgundur- notmyname : I will give oshritf and eranrom a heads up - they might be able to help if we havent solved it by the time their next work day starts | 20:22 |
| jrichli | I can take a look now as well - but i am pretty new to the area | 20:22 |
| jrichli | notmyname: i probably missed when people discussed this before. but why were the probetest gates not in place? | 20:24 |
| sgundur- | jrichli: thanks :) , let me know what to look for | 20:25 |
| notmyname | jrichli: probetest gates have never been "gates" (ie voting) because they are set up to run against the SAIO. and nobody has figured out how to do that with the gate images. there are probetests running in the community qa cluster though, but they aren't voting | 20:31 |
| jrichli | notmyname: ah, ok thanks | 20:32 |
| *** jeblair has quit IRC | 20:33 | |
| *** jeblair has joined #openstack-swift | 20:33 | |
| *** suyash has joined #openstack-swift | 20:34 | |
| openstackgerrit | Paul Dardeau proposed openstack/swift: resurrect gholt blog posts on building consistent hashing ring https://review.openstack.org/314270 | 20:36 |
| notmyname | pdardeau: thanks! | 20:40 |
| pdardeau | notmyname: np, happy to do that | 20:43 |
| notmyname | I feel like I've unlocked an openstack merit badge or something. my email started quite a nice little bikeshed and will maybe grow into it's own flamewar? | 20:45 |
| notmyname | personal preference opinion poll time | 20:54 |
| notmyname | do you prefer "range(10)[:-6:-1]" or "list(reversed(range(10)[-5:]))" | 20:55 |
| pdardeau | notmyname: dang, i thought the poll was going to be on what should be the first pluggable backend :-) | 20:55 |
| *** mvk has joined #openstack-swift | 20:57 | |
| notmyname | list(reversed()) | 21:08 |
| notmyname | [1.3574769496917725, 1.336827039718628, 1.3806898593902588] | 21:08 |
| notmyname | slicing | 21:08 |
| notmyname | [0.2137138843536377, 0.21651601791381836, 0.2115950584411621] | 21:08 |
| notmyname | so there's the answer | 21:08 |
| *** mingdang1 has joined #openstack-swift | 21:14 | |
| *** sheel has quit IRC | 21:15 | |
| *** pauloewerton has quit IRC | 21:17 | |
| *** esker has joined #openstack-swift | 21:19 | |
| *** lakshmiS has quit IRC | 21:22 | |
| *** d0ugal has quit IRC | 21:26 | |
| *** esker has joined #openstack-swift | 21:27 | |
| *** gyee has quit IRC | 21:33 | |
| *** resker has joined #openstack-swift | 21:38 | |
| *** esker has quit IRC | 21:40 | |
| ahale | yeah that thread is awful. if you dd want to grow a flamewar then "perhaps Swift will start to look more like the rest of the projects in OpenStack " looks a tempting place to start | 21:40 |
| notmyname | oh, I'm on fire today. I just pasted a 15 paragraph review comment against a cross-project spec :-) | 21:41 |
| notmyname | patch 312718 | 21:41 |
| patchbot | notmyname: https://review.openstack.org/#/c/312718/ - governance - add resolution explaining which tests we think def... | 21:41 |
| *** esker has joined #openstack-swift | 21:41 | |
| briancline | oh, *that* spec | 21:42 |
| *** CaioBrentano has quit IRC | 21:43 | |
| *** resker has quit IRC | 21:43 | |
| zaitcev | I hope "defcore" does not mean "defective core". | 21:44 |
| notmyname | that's way too easy of a setup. I'd feel bad just for making the joke ;-) | 21:44 |
| notmyname | I like how I get to spend time on that instead of, you know, writing or reviewing code | 21:45 |
| zaitcev | And now I'm pondering if I should write my own Swift client in Python. | 21:46 |
| bkeller` | can we rewrite it in Piet? | 21:46 |
| notmyname | bkeller`: only after you fully define the testing, documentation, deployment, and cross-project library processes for all projects written in Piet | 21:47 |
| briancline | TIL a new esoteric language | 21:48 |
| notmyname | bkeller`: lol, I was actually thinking of a reply referencing "mondrian". I had mis-remembered the name :-) | 21:49 |
| *** _JZ_ has quit IRC | 21:50 | |
| briancline | what are the chances of patch 309524 being merged in the next few days? | 21:54 |
| patchbot | briancline: https://review.openstack.org/#/c/309524/ - swift - SwiftLogFormatter will log transaction IDs on INFO... | 21:54 |
| notmyname | briancline: IIRC torgomatic asked me about that last week. where are you seeing this cause problems for you? INFO level is the access logs, so that's already done | 21:57 |
| notmyname | I mean, it sure looks like we did that on purpose (test for it and everything) | 21:58 |
| notmyname | but torgomatic said this patch doesn't result in double-logging the txn_id, so I don't really know why it is the way it is | 21:58 |
| briancline | notmyname: on anything that middlewares log. the keystone middleware is the most painful of the lot. it logs plenty of useful info, but with lots of keystone requests, there's no helpful way of tying back its messages to the originating request | 21:59 |
| notmyname | that's true | 21:59 |
| briancline | custom middlewares is another. we have one or two middlewares that i think have done it in hackier ways, but it's not always effective, somehow | 22:00 |
| *** catintheroof has quit IRC | 22:01 | |
| torgomatic | yeah, it's a weird decision to skip txid for just INFO. I suspect it started out sane, then got twisted up in a couple different patch sets and now it's all wonky | 22:02 |
| torgomatic | but since it checks that txid isn't in the log message before adding it, removing the INFO check seems harmless at worst | 22:03 |
| *** NM has quit IRC | 22:09 | |
| notmyname | http://d.not.mn/python_vs_golang_gets.png | 22:13 |
| notmyname | http://d.not.mn/python_vs_golang_puts.png | 22:13 |
| notmyname | python vs golang on the object server, from numbers provided by dfg_ ^ | 22:13 |
| notmyname | also, what's not shown there is the dramatically smaller cpu time for the golang ones | 22:14 |
| mattoliverau | morning | 22:17 |
| briancline | I traced back a bit since I got curious. the INFO level check seems to have been originally introduced here: https://github.com/openstack/swift/commit/442e3c8a1a92eb3da140fc39f78b4ff86c50d11a#diff-67e43d32a534c1f6d2e1ca157dc8a1a0R345 | 22:19 |
| *** greghaynes has joined #openstack-swift | 22:22 | |
| torgomatic | briancline: good call... and you can see there that it doesn't have the check for duplicate txids, so this made some sense as a way to not crud up the access logs | 22:23 |
| briancline | yeah. probably a bit closer to constant time compared to how its done today as well, but i doubt anyone would ever notice the difference | 22:29 |
| briancline | notmyname: those are really nice numbers.. are there similar ones for avg time to first byte or avg request duration? | 22:30 |
| *** nadeem has quit IRC | 22:31 | |
| notmyname | briancline: no, but slowet 1% on golang was about 30% of the python number (for 1byte) and 62% for 1M object (for PUTs) | 22:32 |
| *** jamielennox|away is now known as jamielennox | 22:35 | |
| notmyname | (lower is better) | 22:35 |
| notmyname | IOW, the long tail also sees improvements | 22:35 |
| *** klamath has joined #openstack-swift | 22:45 | |
| *** cdelatte has quit IRC | 22:47 | |
| *** cdelatte has joined #openstack-swift | 22:48 | |
| *** cdelatte has quit IRC | 22:51 | |
| *** cdelatte has joined #openstack-swift | 22:51 | |
| *** ametts has quit IRC | 22:58 | |
| jrichli | I am about to be offline for a few hours, but I am gonna put a question out there: | 23:03 |
| jrichli | is it only sgundur- and I that see the container-sync probetest failures? how about somebody not using Ubuntu 14 ? I am testing out older versions and I keep getting these failures on my VM | 23:04 |
| notmyname | jrichli: after which patch? | 23:05 |
| jrichli | I keep going backwards trying to find one that affected container-sync where I DONT get the failures. but i have gone back all the way to patch 270961 and my VM shows the same failures there too | 23:06 |
| patchbot | jrichli: https://review.openstack.org/#/c/270961/ - swift - Container-Sync to perform HEAD before PUT object o... (MERGED) | 23:06 |
| jrichli | but the vm-saio-probe passed when that was merged | 23:06 |
| *** diogogmt has quit IRC | 23:06 | |
| jrichli | so I think my setup is wrong somehow | 23:07 |
| jrichli | I guess I would like to know if somebody has all container-sync tests passing for probetests against master right now | 23:07 |
| notmyname | testing with patch 311899 right now | 23:08 |
| patchbot | notmyname: https://review.openstack.org/#/c/311899/ - swift - Fix probe tests from commit cf48e75 | 23:08 |
| jrichli | thanks! ok, gotta run. bb in a few hours | 23:08 |
| *** mingdang1 has quit IRC | 23:09 | |
| notmyname | hmm...no new wiki accounts for anyone. I just learned that http://lists.openstack.org/pipermail/openstack-dev/2016-May/094408.html | 23:11 |
| notmyname | that will make our no-specs thing harder if you can't actually go update the ideas page | 23:12 |
| *** cdelatte has quit IRC | 23:17 | |
| mattoliverau | hmmm, bugger | 23:20 |
| sgundur- | mattoliverau: hi | 23:22 |
| *** kei_yama has joined #openstack-swift | 23:28 | |
| *** chlong has joined #openstack-swift | 23:33 | |
| *** km has joined #openstack-swift | 23:34 | |
| notmyname | jrichli: I can say that I do not see those errors in my probetests | 23:35 |
| notmyname | https://gist.github.com/notmyname/c7514864290cd29524b4e6058dc75516 | 23:39 |
| sgundur- | notmyname: curious to know if u get the above results with current code and torgomatic's patch or older version ? | 23:46 |
| notmyname | yeah, that was with master + patch 311899 | 23:46 |
| patchbot | notmyname: https://review.openstack.org/#/c/311899/ - swift - Fix probe tests from commit cf48e75 | 23:46 |
| sgundur- | notmyname: last time I had checked I had got different results, let me recreate again | 23:48 |
| notmyname | FWIW, I don't think any of the errors I'm seeing have to do with your port numbers patch | 23:48 |
| *** hosanai has joined #openstack-swift | 23:52 | |
| *** ChanServ sets mode: +v hosanai | 23:52 | |
| kota_ | morning | 23:53 |
| torgomatic | notmyname: I was also seeing some probetest failures that appeared to be related to busted info-cache stuff around container-sync (I think) | 23:54 |
| mattoliverau | kota_: morning | 23:54 |
| kota_ | mattoliverau: o/ | 23:55 |
| *** morgan has left #openstack-swift | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
