| *** diogogmt has quit IRC | 00:00 | |
| *** nadeem has quit IRC | 00:01 | |
| *** kei_yama has quit IRC | 00:06 | |
| *** kei_yama has joined #openstack-swift | 00:07 | |
| notmyname | tsg_: ping | 00:11 |
|---|---|---|
| notmyname | tsg_: given the client-visible length of an object (eg I uploaded a 100MB file), can pyeclib tell me how long the fragment archive will be? | 00:12 |
| notmyname | it's soemthing like this | 00:12 |
| notmyname | expected_fa_size = client_cl // segment_size * fragment_size + tail | 00:12 |
| notmyname | where tail = pyeclib.tell_me_size_of_this(client_cl % segments) | 00:13 |
| notmyname | ie the ec'd left over bits that don't fill up the last segment | 00:13 |
| *** haomaiwang has joined #openstack-swift | 00:18 | |
| *** haomaiwang has quit IRC | 00:24 | |
| *** openstackgerrit has quit IRC | 00:32 | |
| *** openstackgerrit has joined #openstack-swift | 00:33 | |
| redbo_ | I don't know, how do we feel about these? I don't mind them, but some people do. "port = match.group('port') or port" | 00:35 |
| openstackgerrit | Timur Alperovich proposed openstack/swift: Fix IPv6 handling in MemcacheConnPool. https://review.openstack.org/258704 | 00:39 |
| *** lcurtis has quit IRC | 00:39 | |
| *** m_kazuhiro has joined #openstack-swift | 00:40 | |
| torgomatic_ | notmyname: it's not even that generators have tp_del defined; it's that has_finalizer() in gcmodule.c has an explicit check for generators with a corresponding set of conditions | 00:41 |
| torgomatic_ | for everything else, it's whether or not tp_del is filled in | 00:41 |
| notmyname | torgomatic_: what could possibly go wrong! | 00:42 |
| *** zhill has joined #openstack-swift | 00:46 | |
| openstackgerrit | Catherine Northcott proposed openstack/swift: [WIP] Get Accounts by Cluster https://review.openstack.org/262087 | 00:50 |
| Zyric | I think I have the first version of the get accounts by cluster patch going - which is to say it's logging, filtering, and displaying okay in my local environment. | 00:52 |
| *** zhill has quit IRC | 00:53 | |
| *** garthb__ has quit IRC | 00:56 | |
| *** garthb__ has joined #openstack-swift | 00:59 | |
| notmyname | Zyric: cool | 01:12 |
| *** haomaiwang has joined #openstack-swift | 01:12 | |
| Zyric | notmyname: The account data (names/paths) doesn't have a real home yet. Should be stored in the recon cache, or someplace in info? I'll add some comments to the patch. | 01:14 |
| Zyric | notmyname: I could use some feedback at this point, should I take the WIP off to encourage review even though it's still very rough? | 01:16 |
| *** haomaiwang has quit IRC | 01:16 | |
| notmyname | Zyric: yeah, that's a good idea | 01:17 |
| notmyname | Zyric: I'm about to head home, so I won't get a chance to look at it today | 01:18 |
| Zyric | notmyname: No worries, I can spend the rest of the week doing patch review and fixing small bugs. | 01:19 |
| notmyname | cool, nice | 01:20 |
| *** tsg_ has quit IRC | 01:28 | |
| *** garthb__ has quit IRC | 01:34 | |
| *** changbl has joined #openstack-swift | 01:36 | |
| *** fthiagogv has quit IRC | 01:39 | |
| *** badari_ has quit IRC | 01:44 | |
| clayg | timur: I think maybe like latest memcache with trusty (1.4.14) does *not* support ipv6? | 01:48 |
| clayg | timur: I tried all the invocations of memcached -l ::1 that I could think of but it kept complaining about parsing it like a port | 01:49 |
| *** badari has joined #openstack-swift | 01:50 | |
| *** badari has quit IRC | 02:02 | |
| *** haomaiwang has joined #openstack-swift | 02:10 | |
| *** robefran has quit IRC | 02:14 | |
| openstackgerrit | Bing Hu proposed openstack/python-swiftclient: Support uploading to an object in swift from stdin update code and add unittest use mkstemp instead of NamedTemporaryFile, upadte unittest Closes-bug: #1521342 https://review.openstack.org/259410 | 02:16 |
| openstack | bug 1521342 in python-swiftclient "support uploading from a stdin stream" [Wishlist,In progress] https://launchpad.net/bugs/1521342 - Assigned to Bing Hu (hubingsh) | 02:16 |
| clayg | timur: so you have to have *the latest* memcache 1.4.25 - tagged Nov '15 | 02:17 |
| clayg | https://github.com/memcached/memcached/wiki/ReleaseNotes1425 | 02:17 |
| clayg | i would have swore we had memcache_server.conf reloading now? | 02:48 |
| clayg | GD, i bet we never merged it | 02:49 |
| clayg | patch 218490 | 02:50 |
| patchbot | clayg: https://review.openstack.org/#/c/218490/ - Automatic refresh of memcache config settings | 02:50 |
| clayg | i guess I can't complain if I never reviewed it - i thought i did | 02:51 |
| *** badari has joined #openstack-swift | 03:00 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-swift | 03:01 | |
| *** gyee has quit IRC | 03:04 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:06 | |
| clayg | how do you review a change in gerrit!? | 03:11 |
| clayg | i found reply at the top - but it has no scores - i am very confused | 03:11 |
| clayg | wow ok, i was old an old patch set | 03:12 |
| *** janonymous has quit IRC | 03:16 | |
| *** badari has quit IRC | 03:27 | |
| *** janonymous has joined #openstack-swift | 03:30 | |
| *** dmorita has quit IRC | 03:40 | |
| openstackgerrit | Catherine Northcott proposed openstack/swift: Get Accounts by Cluster https://review.openstack.org/262087 | 03:42 |
| *** Jeffrey4l has quit IRC | 03:43 | |
| mattoliverau | clayg: yeah, patchsets are also confusing, ie which one your in etc. | 03:43 |
| *** dmorita has joined #openstack-swift | 03:43 | |
| *** Jeffrey4l has joined #openstack-swift | 03:55 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwa_ has joined #openstack-swift | 04:01 | |
| *** diogogmt has joined #openstack-swift | 04:14 | |
| *** links has joined #openstack-swift | 04:27 | |
| *** diogogmt has quit IRC | 04:32 | |
| *** ppai has joined #openstack-swift | 04:47 | |
| openstackgerrit | Merged openstack/swift: Merge branch 'master' into feature/hummingbird https://review.openstack.org/264517 | 04:54 |
| *** ajiang has joined #openstack-swift | 04:58 | |
| *** bhakta has joined #openstack-swift | 04:59 | |
| *** haomaiwa_ has quit IRC | 05:01 | |
| *** haomaiwa_ has joined #openstack-swift | 05:01 | |
| *** doxavore has quit IRC | 05:02 | |
| *** suresh_ has joined #openstack-swift | 05:03 | |
| suresh_ | hi, using a separate partition for Swift data or using a loopback device instead of another partition Which is the best practice ? | 05:05 |
| suresh_ | please someone help | 05:05 |
| *** trifon has joined #openstack-swift | 05:10 | |
| *** klrmn1 has quit IRC | 05:12 | |
| *** blmartin has quit IRC | 05:13 | |
| *** yatin has joined #openstack-swift | 05:13 | |
| kota_ | hello | 05:17 |
| mattoliverau | suresh_: for a SAIO you mean? | 05:19 |
| mattoliverau | kota_: o/ | 05:19 |
| *** trifon has quit IRC | 05:19 | |
| kota_ | mattoliverau: o/ | 05:21 |
| *** yatin has quit IRC | 05:24 | |
| *** jamielennox is now known as jamielennox|away | 05:27 | |
| *** esker has joined #openstack-swift | 05:31 | |
| *** yatin has joined #openstack-swift | 05:36 | |
| *** yatin has quit IRC | 05:37 | |
| suresh_ | mattoliverau: In any case either SAIO or Cluster | 05:39 |
| *** trifon has joined #openstack-swift | 05:40 | |
| *** mragupat has joined #openstack-swift | 05:40 | |
| *** mragupat has quit IRC | 05:42 | |
| *** mragupat has joined #openstack-swift | 05:42 | |
| mattoliverau | suresh_: loop back is really just for SAIO, it's better to use a bunch of disks, mounted those mount points become the devices in the ring. | 05:43 |
| suresh_ | mattoliverau: For SAIO i added my strogae nodes | 05:45 |
| suresh_ | but in swift-recon --diskusage command shows only disks which are not loop backdevices | 05:46 |
| *** ppai has quit IRC | 05:49 | |
| * suresh_ slaps bsdkurt around a bit with a large fishbot | 05:58 | |
| *** haomaiwa_ has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-swift | 06:01 | |
| *** wuhg has joined #openstack-swift | 06:02 | |
| *** SkyRocknRoll has joined #openstack-swift | 06:02 | |
| *** dslevin has quit IRC | 06:06 | |
| *** dslevin has joined #openstack-swift | 06:07 | |
| *** SkyRocknRoll has quit IRC | 06:07 | |
| *** logan- has quit IRC | 06:08 | |
| *** logan- has joined #openstack-swift | 06:08 | |
| *** siva_krishnan has quit IRC | 06:09 | |
| *** patchbot has quit IRC | 06:09 | |
| *** notmyname has quit IRC | 06:09 | |
| *** sgundur has quit IRC | 06:09 | |
| *** patchbot has joined #openstack-swift | 06:09 | |
| *** notmyname has joined #openstack-swift | 06:10 | |
| *** ChanServ sets mode: +v notmyname | 06:10 | |
| *** asettle has joined #openstack-swift | 06:10 | |
| *** sgundur has joined #openstack-swift | 06:11 | |
| *** esker has quit IRC | 06:13 | |
| *** siva_krishnan has joined #openstack-swift | 06:15 | |
| *** bkmz_ has quit IRC | 06:16 | |
| *** bkmz has joined #openstack-swift | 06:16 | |
| *** ChubYann has quit IRC | 06:21 | |
| *** suresh_ has quit IRC | 06:22 | |
| *** yatin has joined #openstack-swift | 06:26 | |
| *** Jeffrey4l has quit IRC | 06:27 | |
| *** dmorita has quit IRC | 06:35 | |
| *** dmorita has joined #openstack-swift | 06:36 | |
| *** Lickitysplitted has joined #openstack-swift | 06:36 | |
| *** dmorita_ has joined #openstack-swift | 06:38 | |
| *** dmorita has quit IRC | 06:38 | |
| *** dmorita_ has quit IRC | 06:38 | |
| *** Jeffrey4l has joined #openstack-swift | 06:40 | |
| *** mragupat has quit IRC | 06:45 | |
| openstackgerrit | LiuNanke proposed openstack/python-swiftclient: Put py34 first in the env order of tox https://review.openstack.org/265106 | 06:48 |
| *** ppai has joined #openstack-swift | 06:49 | |
| *** itlinux has joined #openstack-swift | 06:58 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:01 | |
| *** jsuchome has joined #openstack-swift | 07:02 | |
| *** Jeffrey4l has quit IRC | 07:09 | |
| *** itlinux has quit IRC | 07:14 | |
| *** itlinux has joined #openstack-swift | 07:15 | |
| *** itlinux has quit IRC | 07:16 | |
| *** Jeffrey4l has joined #openstack-swift | 07:21 | |
| *** yarkot has quit IRC | 07:28 | |
| *** logan- has quit IRC | 07:44 | |
| *** arnox has joined #openstack-swift | 07:59 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-swift | 08:02 | |
| *** asettle has quit IRC | 08:06 | |
| *** asettle has joined #openstack-swift | 08:06 | |
| *** cschwede has joined #openstack-swift | 08:10 | |
| *** zaitcev has quit IRC | 08:10 | |
| *** asettle has quit IRC | 08:11 | |
| *** rledisez has joined #openstack-swift | 08:15 | |
| *** haomaiwang has quit IRC | 08:19 | |
| *** philipw_ has quit IRC | 08:22 | |
| *** philipw has joined #openstack-swift | 08:22 | |
| *** ChanServ sets mode: +v cschwede | 08:22 | |
| *** jistr has joined #openstack-swift | 08:24 | |
| *** haomaiwang has joined #openstack-swift | 08:24 | |
| *** Jeffrey4l has quit IRC | 08:33 | |
| *** SkyRocknRoll has joined #openstack-swift | 08:35 | |
| openstackgerrit | renminmin proposed openstack/swift: Byte-quotas should allow object re-upload https://review.openstack.org/263227 | 08:37 |
| *** yatin has quit IRC | 08:37 | |
| *** Jeffrey4l has joined #openstack-swift | 08:49 | |
| *** haomaiwang has quit IRC | 08:50 | |
| *** marekkaleta has joined #openstack-swift | 08:51 | |
| timur | clayg: right -- I did not notice when they packaged it up. When I was looking at the change initially, they still haven't released it and I actually tested with the master branch | 08:52 |
| timur | I'll update the review to reflect that | 08:52 |
| *** trifon has quit IRC | 08:52 | |
| timur | about the failure mode: I agree that it would be better to fail in the constructor (or if there is any other config validation that runs, we could fail ealier?). Although, I'd rather defer that into another patchset | 08:52 |
| timur | I'm happy to submit it for review, though -- will look into that tomorrow | 08:53 |
| timur | are there strong objections to defering that work? my rationale is to keep commits to the smallest changes possible and the IP address issue existed prior to IPv6 work | 08:54 |
| *** haomaiwa_ has joined #openstack-swift | 08:54 | |
| *** trifon has joined #openstack-swift | 08:56 | |
| *** kaleta has joined #openstack-swift | 08:59 | |
| *** haomaiwa_ has quit IRC | 09:01 | |
| *** haomaiwa_ has joined #openstack-swift | 09:05 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permissions https://review.openstack.org/265145 | 09:05 |
| *** marekkaleta has quit IRC | 09:06 | |
| *** marekkaleta has joined #openstack-swift | 09:06 | |
| *** wmn-tech has quit IRC | 09:22 | |
| *** wuhg has quit IRC | 09:28 | |
| *** suresh has joined #openstack-swift | 09:31 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add note COPY with conditional headers https://review.openstack.org/265154 | 09:32 |
| suresh | hi, I am getting dispersion while doing swift-ring-builder rebalance Is it really a problem | 09:33 |
| suresh | please someone help | 09:33 |
| openstackgerrit | ting wang proposed openstack/swift: Use the oslo.utils.reflection to extract the class name https://review.openstack.org/265155 | 09:35 |
| *** yatin has joined #openstack-swift | 09:39 | |
| *** jordanP has joined #openstack-swift | 09:39 | |
| *** suresh has quit IRC | 09:42 | |
| *** acoles_ is now known as acoles | 09:45 | |
| ahale | you are probably trying to make a ring change that needs to move more partitions around than is possible - i would deploy the unbalanced ring then wait until replication has moved that data around and/or your dispersion-report looks better, then do more rebalances until it can get balanced | 09:45 |
| *** suresh has joined #openstack-swift | 09:46 | |
| suresh | I am getting dispersion while doing rebalance | 09:46 |
| suresh | please someone help | 09:46 |
| *** haomaiwa_ has quit IRC | 09:47 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permissions from files which are not run https://review.openstack.org/265145 | 09:49 |
| *** haomaiwang has joined #openstack-swift | 09:50 | |
| *** joeljwright has joined #openstack-swift | 09:51 | |
| *** ChanServ sets mode: +v joeljwright | 09:51 | |
| *** Jeffrey4l has quit IRC | 10:00 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| *** Jeffrey4l has joined #openstack-swift | 10:06 | |
| *** janonymous has quit IRC | 10:06 | |
| *** yatin has quit IRC | 10:16 | |
| *** yatin has joined #openstack-swift | 10:17 | |
| *** hseipp has joined #openstack-swift | 10:17 | |
| acoles | cd ../swift | 10:42 |
| acoles | oops | 10:42 |
| onovy | jupiter:~ onovy$ passwd | 10:47 |
| onovy | Changing password for onovy. | 10:47 |
| onovy | Old Password: mySecretPassword | 10:47 |
| onovy | oops sry too | 10:47 |
| acoles | onovy: :) | 10:48 |
| acoles | mahatic: patch 211451 merged so I will rebase others now | 10:51 |
| patchbot | acoles: https://review.openstack.org/#/c/211451/ - Modify unit tests to include real crypto (MERGED) | 10:51 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Remove content-type encryption/decryption https://review.openstack.org/261292 | 10:54 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/173982 | 10:56 |
| *** kaleta has quit IRC | 10:57 | |
| *** kaleta has joined #openstack-swift | 10:57 | |
| *** daemontool has joined #openstack-swift | 10:58 | |
| *** kaleta has quit IRC | 10:59 | |
| *** kaleta has joined #openstack-swift | 10:59 | |
| acoles | mahatic: so to pick up from yesterday...you now need to pull latest version of patch 261292 from gerrit, check its ok, then rebase your other patch (multipart) on that and push it to gerrit | 11:00 |
| patchbot | acoles: https://review.openstack.org/#/c/261292/ - Remove content-type encryption/decryption | 11:00 |
| acoles | mahatic: i abandoned the patches that got pushed by mistake yesterday to avoid any confusion | 11:00 |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| *** acoles is now known as acoles_ | 11:02 | |
| *** kaleta has quit IRC | 11:05 | |
| *** kaleta has joined #openstack-swift | 11:05 | |
| *** kaleta has quit IRC | 11:09 | |
| *** kaleta has joined #openstack-swift | 11:09 | |
| *** marekkaleta has quit IRC | 11:10 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: Change schedule priority of daemon/server in config https://review.openstack.org/238799 | 11:10 |
| *** acoles_ is now known as acoles | 11:10 | |
| *** janonymous has joined #openstack-swift | 11:11 | |
| *** janonymous has quit IRC | 11:24 | |
| *** haomaiwang has quit IRC | 11:28 | |
| *** sanchitmalhotra has quit IRC | 11:34 | |
| mahatic | acoles: thanks much! I was busy with some paper work, just around to these, will pick it up | 11:35 |
| mahatic | just got around* | 11:36 |
| acoles | mahatic: ok. it would be worth checking 261292 to make sure it is still removing all the content-type stuff, because there may have been stuff added in 211451 since 261292 got authored | 11:36 |
| mahatic | correct | 11:37 |
| mahatic | acoles: ccould you let me know what is that you had to do 261292? Because since 211451 one is merged, I thought, you'd point 261292 to the latest feature/crypto? | 11:38 |
| acoles | mahatic: actually i took a risk and pushed rebase on gerrit! but checked that the parent it used was in fact the commit id of 211451 (gerrit will show you the parent commit id for a patch set) | 11:39 |
| suresh | I am getting dispersion while doing rebalance | 11:40 |
| suresh | please someone help | 11:40 |
| mahatic | acoles: oh, how come it didn't show conflicts? (there was the small change I pointed out and you corrected on 211451) | 11:40 |
| *** silor has joined #openstack-swift | 11:41 | |
| mahatic | acoles: I'll start with checking 261292 | 11:41 |
| acoles | suresh see ahale reply 2 hours ago. i'm not best person to answer your question. | 11:44 |
| acoles | suresh: but in general if you see large dispersion it means that some partitions' replicas cannot be spread across your devices ideally. | 11:44 |
| acoles | which may be due to device weights preventing that happening. the overload option can be used to allow weight to be 'overridden' somewhat. | 11:45 |
| acoles | suresh: but, i don't have a great deal of experience doing that in production, so maybe ask here again in US timezone | 11:46 |
| acoles | mahatic: it would only show conflict if it failed to merge changes | 11:46 |
| suresh | acoles: Thanks for your reply | 11:47 |
| mahatic | acoles: yes, come to think of it, you only removed stuff | 11:47 |
| acoles | mahatic: well, i did put in a couple of extra tests...for the bad iv and bad key | 11:49 |
| mahatic | acoles: oh no, I mean the very last patchset | 11:49 |
| mahatic | :) | 11:49 |
| mahatic | acoles: because of my comment on some bad iv that was not required in a test | 11:49 |
| mahatic | 261292 has your changes apart from the last minor one | 11:50 |
| acoles | mahatic: looking | 11:55 |
| suresh | Is having more Zones is a problem or Not | 11:56 |
| onovy | suresh: hi, are you adding new devices? | 11:56 |
| onovy | if yes, how much vs. how much do you have actually | 11:56 |
| onovy | +what is your replication factor | 11:56 |
| suresh | onovy: yes i am adding new devices and i have 3 replicas | 11:57 |
| mahatic | acoles: oh nevermind, I meant it before looking at the changes (and before jenkin's +1). I verified now, looks like it does | 11:57 |
| onovy | all questions and answers pls :) | 11:57 |
| onovy | swift will move only one replica of partion per rebalance, so all data is available. so if you added too much space to cluster, it's normal to have dispersion | 11:58 |
| onovy | wait at least min_part_hours and do rebalance again | 11:58 |
| onovy | min_part_hours should be bigger "convergate" time. so all replicas are on correct place after it | 11:59 |
| suresh | onovy: I am setting min_part_hours to zero Is it is a problem | 12:01 |
| suresh | onovy: vs means | 12:03 |
| onovy | it's production cluster? | 12:04 |
| onovy | with r/w traffic? | 12:04 |
| suresh | onovy: yeah it's for production | 12:05 |
| onovy | so don't set min_part_hours to 0 | 12:06 |
| acoles | mahatic: errr, https://review.openstack.org/#/c/261292/4/test/unit/common/middleware/test_decrypter.py lines 92 126 and 161 - that is what I was worried about. #willfix | 12:06 |
| onovy | you can have data unvailable | 12:06 |
| onovy | no data lost, but unavailable for read or for new put to existing cont./acc. | 12:06 |
| onovy | you should do rebalance, wait for rebalance finished and then do another rebalance | 12:07 |
| onovy | i could recommend one book for swift admins. mmnt | 12:07 |
| onovy | suresh: http://shop.oreilly.com/product/0636920033288.do this | 12:07 |
| mahatic | acoles: how did you even find them? :) when they were in "skipped common lines" | 12:07 |
| onovy | there is explained how to add new devices to small cluster | 12:07 |
| onovy | with small weight for example | 12:07 |
| *** yatin has quit IRC | 12:09 | |
| suresh | onovy: if i have 1 hour min_part_time means after adding new device i should wait one hour to rebalance | 12:09 |
| onovy | no :) | 12:09 |
| onovy | it means swift-ring-builder will rebalance it after 1 hour | 12:09 |
| onovy | at first | 12:10 |
| onovy | if you call it | 12:10 |
| *** km_ has quit IRC | 12:10 | |
| onovy | you should look into logs of replicator | 12:10 |
| onovy | min_part_time is just soft limit. you choose "how long will replication probably take" | 12:10 |
| onovy | you can put 24/48 hours here. after this time, cluster __should__ be balance | 12:11 |
| onovy | but you couldn't be sure | 12:11 |
| onovy | use swift-dispersion-populate/report to check. and logs of replicator | 12:11 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Remove content-type encryption/decryption https://review.openstack.org/261292 | 12:12 |
| acoles | mahatic: ^^ I'm not sure about the skipped tests in there but i can't review it all today | 12:12 |
| *** yatin has joined #openstack-swift | 12:12 | |
| mahatic | acoles: sure, no problem. They are mean to used to test etag | 12:13 |
| mahatic | meant* | 12:13 |
| acoles | mahatic: so grab that version ^^ and use it to rebase your other patch | 12:14 |
| suresh | onovy: when doing rebalance its giving some two lines warning like No partitions could be reassigned. Either none need to be or none can be due to min_part_hours [1]. | 12:14 |
| suresh | Is this is a problem | 12:14 |
| mahatic | acoles: yup, thanks | 12:14 |
| onovy | suresh: yes, because you are rebalancing before min_part_hours time | 12:14 |
| onovy | if you wait this time, rebalance will fix dispersion then | 12:14 |
| onovy | of course if it's possible to "fix it". sometimes it's not possible | 12:14 |
| suresh | onovy: when should i wait Before rebalancing or after rebalancing | 12:15 |
| onovy | rebalance, push it to cluster, wait, rebalance, push it to cluster, wait, ... | 12:16 |
| *** yatin has quit IRC | 12:17 | |
| onovy | swift cores: this review https://review.openstack.org/#/c/257577/ should help with problems like suresh have now | 12:17 |
| *** ppai has quit IRC | 12:19 | |
| *** kei_yama has quit IRC | 12:23 | |
| suresh | onovy: If i am using different Zones for different devices i am not getting dispersion Is it recommendable way | 12:24 |
| *** jmccarthy has quit IRC | 12:26 | |
| *** jmccarthy has joined #openstack-swift | 12:27 | |
| onovy | suresh: different zones is fine | 12:28 |
| *** suresh has quit IRC | 12:28 | |
| *** ppai has joined #openstack-swift | 12:32 | |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift: Decrypt multipart mime doc https://review.openstack.org/248377 | 12:53 |
| *** m_kazuhiro has quit IRC | 12:53 | |
| *** Jeffrey4l has quit IRC | 12:53 | |
| *** esker has joined #openstack-swift | 13:13 | |
| *** esker has quit IRC | 13:20 | |
| *** links has quit IRC | 13:25 | |
| *** miurahr has quit IRC | 13:32 | |
| *** ppai has quit IRC | 13:40 | |
| openstackgerrit | Stuart McLaren proposed openstack/python-swiftclient: Retry download of object body https://review.openstack.org/159208 | 13:42 |
| *** Jeffrey4l has joined #openstack-swift | 13:45 | |
| *** sileht has quit IRC | 13:46 | |
| *** sileht has joined #openstack-swift | 13:48 | |
| *** badari has joined #openstack-swift | 14:02 | |
| *** asettle has joined #openstack-swift | 14:07 | |
| *** dslevin has quit IRC | 14:10 | |
| *** asettle has quit IRC | 14:12 | |
| *** jordanP is now known as jordanP|away | 14:14 | |
| *** SkyRocknRoll has quit IRC | 14:16 | |
| *** esker has joined #openstack-swift | 14:17 | |
| *** esker has quit IRC | 14:22 | |
| *** esker has joined #openstack-swift | 14:22 | |
| *** jsuchome has quit IRC | 14:25 | |
| *** C1ShaunID has joined #openstack-swift | 14:27 | |
| *** petertr7_away is now known as petertr7 | 14:47 | |
| *** esker has quit IRC | 14:48 | |
| *** jordanP|away is now known as jordanP | 14:50 | |
| *** haomaiwa_ has joined #openstack-swift | 14:51 | |
| *** esker has joined #openstack-swift | 14:54 | |
| *** yarkot has joined #openstack-swift | 14:55 | |
| *** blmartin has joined #openstack-swift | 14:56 | |
| *** esker has quit IRC | 14:58 | |
| *** haomaiwa_ has quit IRC | 15:01 | |
| *** esker has joined #openstack-swift | 15:01 | |
| *** haomaiwang has joined #openstack-swift | 15:02 | |
| *** lpabon has joined #openstack-swift | 15:03 | |
| *** esker has quit IRC | 15:08 | |
| *** esker has joined #openstack-swift | 15:09 | |
| *** esker has quit IRC | 15:15 | |
| *** petertr7 is now known as petertr7_away | 15:19 | |
| *** esker has joined #openstack-swift | 15:20 | |
| *** petertr7_away is now known as petertr7 | 15:24 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:24 | |
| *** esker has quit IRC | 15:25 | |
| *** diogogmt has joined #openstack-swift | 15:25 | |
| *** esker has joined #openstack-swift | 15:30 | |
| *** C1ShaunID has left #openstack-swift | 15:30 | |
| *** breitz has quit IRC | 15:30 | |
| *** esker has quit IRC | 15:30 | |
| *** breitz has joined #openstack-swift | 15:31 | |
| *** mragupat has joined #openstack-swift | 15:38 | |
| *** bjkeller has joined #openstack-swift | 15:41 | |
| *** links has joined #openstack-swift | 15:47 | |
| *** diogogmt has quit IRC | 15:50 | |
| *** garthb__ has joined #openstack-swift | 15:51 | |
| *** lcurtis has joined #openstack-swift | 15:53 | |
| *** lcurtis has quit IRC | 15:58 | |
| *** esker has joined #openstack-swift | 16:00 | |
| *** esker has quit IRC | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-swift | 16:01 | |
| *** diogogmt has joined #openstack-swift | 16:03 | |
| *** klrmn has joined #openstack-swift | 16:06 | |
| *** diogogmt has quit IRC | 16:08 | |
| *** zigo has quit IRC | 16:09 | |
| *** zigo has joined #openstack-swift | 16:10 | |
| *** yarkot has quit IRC | 16:11 | |
| *** petertr7 is now known as petertr7_away | 16:11 | |
| *** diogogmt has joined #openstack-swift | 16:12 | |
| *** daemontool has quit IRC | 16:15 | |
| *** petertr7_away is now known as petertr7 | 16:17 | |
| *** links has quit IRC | 16:27 | |
| *** jistr has quit IRC | 16:33 | |
| *** lcurtis has joined #openstack-swift | 16:34 | |
| *** dslevin has joined #openstack-swift | 16:40 | |
| *** gyee has joined #openstack-swift | 16:40 | |
| *** dslevin has quit IRC | 16:42 | |
| *** dslevin__ has joined #openstack-swift | 16:42 | |
| *** dslevi___ has joined #openstack-swift | 16:45 | |
| *** dslevin__ has quit IRC | 16:47 | |
| *** blmartin has quit IRC | 16:48 | |
| *** nadeem has joined #openstack-swift | 16:50 | |
| *** blmartin has joined #openstack-swift | 16:50 | |
| *** nadeem has quit IRC | 16:50 | |
| *** nadeem has joined #openstack-swift | 16:51 | |
| *** dslevi___ has quit IRC | 16:59 | |
| *** mragupat has quit IRC | 16:59 | |
| *** mragupat has joined #openstack-swift | 17:00 | |
| *** haomaiwang has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-swift | 17:01 | |
| *** dslevin has joined #openstack-swift | 17:07 | |
| *** dslevin has quit IRC | 17:10 | |
| *** klrmn has quit IRC | 17:13 | |
| *** esker has joined #openstack-swift | 17:16 | |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Retry download of object body https://review.openstack.org/159208 | 17:20 |
| *** esker has quit IRC | 17:25 | |
| *** joeljwright has quit IRC | 17:30 | |
| *** dslevin__ has joined #openstack-swift | 17:39 | |
| *** dmorita has joined #openstack-swift | 17:41 | |
| *** dslevin__ has quit IRC | 17:42 | |
| *** jordanP has quit IRC | 17:45 | |
| notmyname | good morning | 17:47 |
| *** dslevin has joined #openstack-swift | 17:48 | |
| *** dslevin has quit IRC | 17:48 | |
| *** dslevin has joined #openstack-swift | 17:48 | |
| *** zhill has joined #openstack-swift | 17:58 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-swift | 18:01 | |
| *** dslevin has quit IRC | 18:02 | |
| *** dslevin has joined #openstack-swift | 18:02 | |
| *** mragupat_ has joined #openstack-swift | 18:02 | |
| *** tristanC_ has joined #openstack-swift | 18:04 | |
| *** wbhuber has quit IRC | 18:04 | |
| *** tristanC has quit IRC | 18:04 | |
| *** bsdkurt has quit IRC | 18:04 | |
| *** bsdkurt has joined #openstack-swift | 18:04 | |
| *** onovy has quit IRC | 18:05 | |
| *** wbhuber has joined #openstack-swift | 18:05 | |
| *** rledisez has quit IRC | 18:05 | |
| *** mragupat has quit IRC | 18:06 | |
| *** dslevin has quit IRC | 18:07 | |
| *** proteusguy_ has quit IRC | 18:11 | |
| *** onovy has joined #openstack-swift | 18:12 | |
| *** SkyRocknRoll has quit IRC | 18:13 | |
| *** klrmn has joined #openstack-swift | 18:16 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Autovivify X-Versions-Location container https://review.openstack.org/265015 | 18:21 |
| *** proteusguy_ has joined #openstack-swift | 18:23 | |
| *** klrmn has quit IRC | 18:26 | |
| *** klrmn has joined #openstack-swift | 18:28 | |
| clayg | timur: yeah I think follow up dependent patch for ipv6 memcache config validation is the right thing - good judgement - still need to get the doc update in. | 18:34 |
| *** arnox has quit IRC | 18:40 | |
| *** acoles is now known as acoles_ | 18:41 | |
| *** hseipp has quit IRC | 18:41 | |
| *** pgbridge has joined #openstack-swift | 18:41 | |
| *** petertr7 is now known as petertr7_away | 18:43 | |
| *** silor has quit IRC | 18:43 | |
| *** zaitcev has joined #openstack-swift | 18:43 | |
| *** ChanServ sets mode: +v zaitcev | 18:43 | |
| *** dmorita has quit IRC | 18:52 | |
| *** itlinux has joined #openstack-swift | 18:53 | |
| *** delattec has joined #openstack-swift | 18:53 | |
| *** geaaru has joined #openstack-swift | 18:54 | |
| *** dmorita has joined #openstack-swift | 18:54 | |
| *** delatte has quit IRC | 18:56 | |
| *** dmorita has quit IRC | 18:57 | |
| *** dmorita has joined #openstack-swift | 18:59 | |
| *** haomaiwang has quit IRC | 19:01 | |
| *** dmorita has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-swift | 19:01 | |
| *** dmorita has joined #openstack-swift | 19:03 | |
| *** asettle has joined #openstack-swift | 19:06 | |
| *** asettle has quit IRC | 19:11 | |
| *** dmorita has quit IRC | 19:13 | |
| *** dmorita has joined #openstack-swift | 19:13 | |
| *** dmorita has quit IRC | 19:16 | |
| *** dmorita has joined #openstack-swift | 19:17 | |
| *** dmorita has quit IRC | 19:20 | |
| *** itlinux has quit IRC | 19:20 | |
| *** dmorita has joined #openstack-swift | 19:20 | |
| *** dmorita has quit IRC | 19:22 | |
| *** dmorita has joined #openstack-swift | 19:24 | |
| notmyname | I cleaned up the starred patches. I unstarred a lot; the 3 that are starred now are ones that I'd like to see land today or monday so we can have a release asap (swift 2.6.0) | 19:27 |
| notmyname | after a release, I can go back through and re-star things that are prioritized | 19:28 |
| *** Jeffrey4l has quit IRC | 19:31 | |
| *** Jeffrey4l has joined #openstack-swift | 19:32 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: _RetryBody doesn't need to track etag/content-length https://review.openstack.org/265417 | 19:32 |
| *** dmorita has quit IRC | 19:37 | |
| *** chsc has joined #openstack-swift | 19:38 | |
| timur | clayg: yep -- I'll upload both in the next hour or so | 19:39 |
| *** petertr7_away is now known as petertr7 | 19:39 | |
| *** delatte has joined #openstack-swift | 19:40 | |
| *** delattec has quit IRC | 19:40 | |
| *** dmorita has joined #openstack-swift | 19:43 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: _RetryBody doesn't need to track etag/content-length https://review.openstack.org/265417 | 19:44 |
| *** chsc has quit IRC | 19:46 | |
| *** delatte has quit IRC | 19:55 | |
| *** dmorita has quit IRC | 19:57 | |
| *** dmorita has joined #openstack-swift | 19:59 | |
| *** haomaiwang has quit IRC | 20:01 | |
| *** zhill has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-swift | 20:01 | |
| Zyric | Good morning | 20:01 |
| *** dmorita has quit IRC | 20:02 | |
| *** dmorita has joined #openstack-swift | 20:03 | |
| *** dmorita has quit IRC | 20:03 | |
| *** ChubYann has joined #openstack-swift | 20:08 | |
| *** cdelatte has joined #openstack-swift | 20:13 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Fix versioned_writes functional test skipping https://review.openstack.org/265017 | 20:19 |
| *** dmorita has joined #openstack-swift | 20:32 | |
| *** blmartin has quit IRC | 20:38 | |
| *** dmorita has quit IRC | 20:55 | |
| *** garthb_ has joined #openstack-swift | 20:58 | |
| *** garthb__ has quit IRC | 20:59 | |
| *** daemontool has joined #openstack-swift | 20:59 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** haomaiwang has joined #openstack-swift | 21:01 | |
| clayg | timur: yay! | 21:08 |
| *** cdelatte has quit IRC | 21:09 | |
| clayg | acoles_: timburke: tell stuart congrats on the retry download of object body patch! nice work on the review. | 21:09 |
| *** dmorita has joined #openstack-swift | 21:13 | |
| *** blmartin has joined #openstack-swift | 21:14 | |
| *** minwoob has joined #openstack-swift | 21:15 | |
| *** dmorita has quit IRC | 21:18 | |
| *** daemontool has quit IRC | 21:27 | |
| *** zhill has joined #openstack-swift | 21:29 | |
| *** dmorita has joined #openstack-swift | 21:32 | |
| *** dmorita has quit IRC | 21:32 | |
| *** daemontool has joined #openstack-swift | 21:32 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Fix versioned_writes functional test skipping https://review.openstack.org/265017 | 21:33 |
| *** diogogmt has quit IRC | 21:40 | |
| *** diogogmt has joined #openstack-swift | 21:41 | |
| *** garthb_ has quit IRC | 21:44 | |
| *** garthb_ has joined #openstack-swift | 21:44 | |
| *** esker has joined #openstack-swift | 21:59 | |
| *** esker has quit IRC | 21:59 | |
| *** dmorita has joined #openstack-swift | 21:59 | |
| *** haomaiwang has quit IRC | 22:01 | |
| *** haomaiwang has joined #openstack-swift | 22:01 | |
| *** esker has joined #openstack-swift | 22:03 | |
| *** esker has quit IRC | 22:17 | |
| timur | clayg: I finished the follow on patch and looked at documenting the IPv6 behavior. I think it makes sense to either add it to the top level module paragraph "Why our own memcache client?.." or to the description of the MemcacheConnPool class. I'd rather not document the internal _get_addr() method. What do you think is best? | 22:22 |
| *** esker has joined #openstack-swift | 22:22 | |
| timur | I think the best way would be to document the parameters to the constructor, but it looks like the __init__ documentation is skipped/not desired by default and I'd rather not break away from that | 22:23 |
| clayg | timur: well i obviously thought it better to extract _get_addr to a pure function and paramertize server - documenting our parser as a toplevel function | 22:23 |
| clayg | timur: I think a class docstring is preferred to explicitly documenting __init__ | 22:24 |
| timur | clayg: right, I got that from the gist, but my hesitation is that its really internal to the pool class (hence not wanting to move it out) | 22:24 |
| clayg | timur: if there's not already a __doc__ on the MemcacheConnPool I think you should add one | 22:24 |
| timur | There is a very short blurb and I was thinking about expanding it to describe the _server_ parameter | 22:25 |
| clayg | timur: yup, probably best | 22:25 |
| timur | alright, let me do that and upload it and we can iterate on that if you'd prefer the top-level get_addr function | 22:25 |
| *** esker has quit IRC | 22:27 | |
| clayg | timur: i'm sure I'd prefer the MemcacheConnPool's parsing of ipv6 strings moved to a pure function - in fact I'd probably just rather have it in utils - but you're correct that's neither here nor their for this patch - get something in docs somewhere (doc-tree for config opts, class docstring, etc/sample.conf, man pages - SOMEWHERE) and then we can do follow up patches for all of that and the others | 22:27 |
| *** petertr7 is now known as petertr7_away | 22:34 | |
| timur | clayg: alright, sounds good. utils sounds like the right place for it | 22:36 |
| clayg | timur: ok, again, can be follow up - and now that you've mentioned it - probably want to update the class doc for the server param anyway (and if you extrat the parser then to be stay dry you'll want to figure out the sphinx syntax to reference the utils function!) | 22:44 |
| clayg | timur: sorry :\ | 22:44 |
| clayg | timur: I really just wanted words somewhere a deployer/operator might have a chance of using it - in case we don't get any of the other docs in at least having words in the sample configs would be pretty good | 22:45 |
| openstackgerrit | Paul Dardeau proposed openstack/swift: Look for device holes that can be reused when adding new device. https://review.openstack.org/265461 | 22:51 |
| clayg | ^ pdardeau whoot whoot! | 22:51 |
| pdardeau | clayg: :-) | 22:54 |
| *** mragupat_ has quit IRC | 22:55 | |
| clayg | pdardeau: at first glance I feel like try: devs.index(None) except ValueError: len(devs) + 1 might do the same thing? | 22:56 |
| clayg | ... i'm checking out the branch now | 22:56 |
| pdardeau | clayg: i figured there was probably a better way, but my python-fu is still low | 22:57 |
| clayg | pdardeau: what about if you try to create a new dev with device_id > 65K - seperate patch? | 22:57 |
| *** bjkeller has left #openstack-swift | 22:57 | |
| pdardeau | clayg: good question. let me have a look | 22:58 |
| *** haomaiwang has quit IRC | 23:01 | |
| *** haomaiwang has joined #openstack-swift | 23:01 | |
| openstackgerrit | Timur Alperovich proposed openstack/swift: Fix IPv6 handling in MemcacheConnPool. https://review.openstack.org/258704 | 23:01 |
| openstackgerrit | Timur Alperovich proposed openstack/swift: Fail early if the memcache address is invalid. https://review.openstack.org/265465 | 23:01 |
| pdardeau | clayg: OverflowError: unsigned short is greater than maximum | 23:02 |
| timur | clayg: no problem. Updated the original review to add documentation and the commit message. I also uploaded the follow-on patch to fail early. Will work on the next patch to move _get_addr() and its test into utils in a bit | 23:03 |
| clayg | timur: sure np, i'll take a look! | 23:04 |
| *** asettle has joined #openstack-swift | 23:06 | |
| clayg | pdardeau: well, unittests seem to think this would be good enough -> https://gist.github.com/clayg/409b9ed47fa550cfc8ca | 23:07 |
| pdardeau | clayg: more elegant, for sure | 23:11 |
| *** asettle has quit IRC | 23:12 | |
| pdardeau | clayg: agreed. cleaner and much more succinct | 23:14 |
| clayg | pdardeau: i think there might be some test cases missing (you're going to have to submit again 'cause pep8 anyway I think) | 23:14 |
| clayg | so... it'd be nice to have a assertRaises(Duplicate) for when you add a dev specifying id that exists (is there already such a test?) | 23:15 |
| clayg | pdardeau: k, scratch that - test_add_dev does exactly that - so already covered there | 23:15 |
| pdardeau | clayg: good point. i don't think there is a test for that | 23:15 |
| pdardeau | ok | 23:16 |
| clayg | pdardeau: I think also both tests use the common add/remove-first-n setup - so you never exercise this code on a ring like [dev, dev, None, dev, dev, None] | 23:16 |
| clayg | it'd be nice to like add three devs to that ring and verify new_dev_ids is [2, 5, 6] or something like that maybe? | 23:16 |
| clayg | ... so just that one more test I think - I'm glad it seems to be a fairly stright forward fix - thanks for filing the bug | 23:17 |
| pdardeau | clayg: thanks. i'll make those updates and make another patch | 23:18 |
| clayg | AWESOME! | 23:19 |
| clayg | timur: !! rfc3986 link - KUDOS! | 23:21 |
| timur | clayg: yea, I figured that's better than the wiki :) | 23:23 |
| clayg | timur: although on my machine https://tools.ietf.org/html/rfc3986#page-19 links into the relevant 3.2.2 Host section a little better : | 23:24 |
| clayg | timur: I guess using external links in docs is always risky like that :'( | 23:24 |
| timur | clayg: that's actually my fault, it should be: https://tools.ietf.org/html/rfc3986#section-3.2.2 | 23:25 |
| timur | because we want to refer to the _section_ and not the page | 23:25 |
| timur | let me fix that | 23:25 |
| clayg | timur: oh yeah, that's probably less brittle - good call | 23:25 |
| clayg | timur: I like the cut of your jib man | 23:26 |
| openstackgerrit | Timur Alperovich proposed openstack/swift: Fix IPv6 handling in MemcacheConnPool. https://review.openstack.org/258704 | 23:27 |
| timur | I'll update the follow-on commit after the first one merges (to avoid the unnecessary verifier runs) | 23:28 |
| clayg | timur: poor jenkins :'( | 23:29 |
| timur | more like, poor other people waiting for patches | 23:29 |
| timur | but jenkins too :) | 23:30 |
| *** minwoob has quit IRC | 23:32 | |
| *** itlinux has joined #openstack-swift | 23:47 | |
| *** pgbridge has quit IRC | 23:55 | |
| *** nadeem has quit IRC | 23:59 | |
| *** trifon has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!