| *** haomaiwang has quit IRC | 00:01 | |
| *** adurbin_ has quit IRC | 00:01 | |
| *** haomaiwang has joined #openstack-cinder | 00:02 | |
| tbarron | thingee: pong (sorry, just got back from commute home) | 00:03 |
|---|---|---|
| jgriffith | mkoderer: BTW... just for an example: https://gist.github.com/j-griffith/ad409d3beb67053b77cc | 00:05 |
| jgriffith | The decorators do in fact work, but like I said in the review; it's a mute point | 00:06 |
| jgriffith | and on that note.. I'm getting dinner :) | 00:06 |
| thingee | tbarron: it's fine, figured out the answer to my question in looking at the POC to your scaling backup spec | 00:06 |
| thingee | tbarron: scaling is going to be a pain for FC | 00:06 |
| tbarron | thingee: for FibreChannel? | 00:06 |
| thingee | yea | 00:06 |
| thingee | assuming they're not running on the same node as c-vol | 00:07 |
| *** hemna is now known as hemnafk | 00:07 | |
| tbarron | hmm, I hadn't thought of that problem. | 00:07 |
| thingee | tbarron: just all the additional hba | 00:07 |
| tbarron | remote attach is hard? | 00:07 |
| *** annashen has joined #openstack-cinder | 00:08 | |
| tbarron | thingee: who wants FC anyways? | 00:08 |
| * tbarron ducks | 00:08 | |
| thingee | tbarron: TRUTH | 00:08 |
| patrickeast | o/ | 00:09 |
| patrickeast | i’ve been told i need to like FC | 00:09 |
| tbarron | patrickeast: I've been told it covers a good part of my paycheck, but I consider that dirty money :-) | 00:09 |
| patrickeast | lol | 00:09 |
| *** rmcall has quit IRC | 00:10 | |
| *** raunak has joined #openstack-cinder | 00:10 | |
| tbarron | thingee: I hadn't thought of that problem but will add it to the list of unresolved issues. | 00:10 |
| *** garthb has quit IRC | 00:12 | |
| * thingee is getting angry with the usps, being on hold for 20 mins | 00:12 | |
| *** annashen has quit IRC | 00:12 | |
| tbarron | that sucks | 00:12 |
| thingee | at this rate, the office will be closed before they get to my call | 00:15 |
| *** mtanino has quit IRC | 00:15 | |
| *** tsekiyama has quit IRC | 00:16 | |
| thingee | we should've called brick cement | 00:16 |
| tbarron | thingee: are they holding hostage goods due to you but you were travelling when they arrived? | 00:16 |
| openstackgerrit | Vilobh Meshram proposed openstack/cinder: Cinder Nested Quota Driver https://review.openstack.org/205369 | 00:17 |
| tbarron | because it's glue | 00:17 |
| thingee | tbarron: no, I'm making a dispute with the usps because they're sorter didn't register a forward I had setup | 00:17 |
| tbarron | thingee: i see, and wish you luck | 00:17 |
| thingee | tbarron: their support has no idea where my package is, nor does the facility of where it was last at according to the tracking number. | 00:18 |
| tbarron | thingee: could a node with local FC attach and c-vol to FC connected backend offer an attach to a backup service node via tgtd/iscsi? | 00:18 |
| *** patrickeast has quit IRC | 00:18 | |
| thingee | tbarron: you want c-vol to attach, offer that device as a target to c-bak? | 00:19 |
| thingee | via iscsi | 00:19 |
| *** vokt has quit IRC | 00:19 | |
| tbarron | thingee: yes, but I may be deranged. | 00:20 |
| thingee | tbarron: that as interesting when I suggested to use a message queue as a DLM | 00:20 |
| tbarron | thingee: my general idea is that c-bak is accessing volumes more or less as nova does | 00:20 |
| * thingee is still sad people don't think he's a genius | 00:20 | |
| tbarron | thingee: I followed that conversation closely as we have an init problem for c-bak (decoupled version) as well as for c-vol (geguileo's problem) | 00:21 |
| *** _cjones_ has quit IRC | 00:21 | |
| tbarron | thingee: w/o bringing HA A/A in, there are similar cleanup issues if more than one service is running | 00:21 |
| thingee | tbarron: my problem isn't something to be alarmed about. This problem exists with c-vol today. | 00:21 |
| thingee | if you want c-vol to do anything that involves attaching a volume, you need an HBA on the node running c-vol | 00:22 |
| tbarron | and the ways I've brainstormed to deal with were: (1) store in the DB which service instance "owns" a bakup op, or (2) have the various backup | 00:22 |
| vilobhmm | jgriffith : ping | 00:22 |
| tbarron | service instances talk to one another about which ops they own. | 00:23 |
| *** jasondotstar has quit IRC | 00:23 | |
| tbarron | #2 is kinda like your idea | 00:23 |
| vilobhmm | thingee : https://review.openstack.org/#/c/206170/4 has lots of +1 and +2…if you find this ok will save few more rebase cycles.. | 00:24 |
| tbarron | thingee: but w.r.t. the FC problem you raised, how does a remote nova compute instance or a remote ironic node attach to an FC-connected volume today? | 00:25 |
| *** marcusvrn has quit IRC | 00:25 | |
| tbarron | thingee: I would think the problem for decoupled backup service would be similar | 00:25 |
| *** lixiaoy1 has joined #openstack-cinder | 00:25 | |
| *** annashen has joined #openstack-cinder | 00:29 | |
| *** salv-orlando has quit IRC | 00:31 | |
| *** laughterwym has joined #openstack-cinder | 00:32 | |
| *** tellesnobrega has joined #openstack-cinder | 00:33 | |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function to manage/unmanage snapshots https://review.openstack.org/144590 | 00:34 |
| *** edmondsw has quit IRC | 00:34 | |
| *** annashen has quit IRC | 00:34 | |
| *** tellesnobrega has quit IRC | 00:34 | |
| thingee | tbarron: that's what I'm saying | 00:36 |
| thingee | tbarron: I don't think it's anything to be alarmed about | 00:36 |
| *** dannywilson has quit IRC | 00:36 | |
| tbarron | thingee: got it, not alarmed, just thinkin' a little bit | 00:36 |
| thingee | I think it's fine if you have a node, and just bring up vms that represent c-bak instances. not sure if that accomplishes anything | 00:37 |
| openstackgerrit | wanghao proposed openstack/cinder: Add pagination to backups https://review.openstack.org/204493 | 00:37 |
| tbarron | there is smoke coming out of tbarron's ears righ tnow | 00:37 |
| *** heyun has joined #openstack-cinder | 00:39 | |
| tbarron | thingee: well, it might be a start towards elastic data motion services (waving hands). The backup, copy, migrate, etc. operations have scaling | 00:39 |
| tbarron | challenges very difft than most of the rest of cinder. | 00:39 |
| tbarron | A lot of the rest could run on a rasberry pi | 00:39 |
| * tbarron ducks | 00:39 | |
| tbarron | if the rasberry pi had massive DB access :-) | 00:40 |
| tbarron | thingee: that is to say, most is control plane scaling, where the danger is DOS b/c of bad DB access. | 00:40 |
| tbarron | (I'm sure there's a lot more there I don't know, but I think in broad strokes what I'm saying may be correct.) | 00:41 |
| tbarron | But the data motion operations, which tend to run a long time, use lots of network, lots of bandwitdh in data path to the actual backend. | 00:42 |
| tbarron | have a different set of scaling challenges. | 00:42 |
| tbarron | I see backup, which is already a separate service, as an opportunity to begin thinking about elastic data motion services for cinder. | 00:43 |
| tbarron | thingee: that could all be a crock, but it's my self-motivational speech for being interested in this backup problem right now. | 00:43 |
| tbarron | these data motion operations are in principle very loosely coupled to the rest of cinder, or even to one another as long as they operate on separate volumes. | 00:45 |
| tbarron | but there are corner cases where some of the interesting distributed computing consistency issues like those you were discussing with Gorka, John, and Eric come up. | 00:46 |
| *** daneyon has quit IRC | 00:48 | |
| *** daneyon has joined #openstack-cinder | 00:48 | |
| *** jwcroppe has joined #openstack-cinder | 00:49 | |
| *** daneyon has quit IRC | 00:49 | |
| thingee | tbarron: so this purely wip, not ready for testing? | 00:49 |
| *** daneyon has joined #openstack-cinder | 00:49 | |
| tbarron | thingee: I had it running before xyang's non-disruptive backup merged. I've rebased but need to do a bit more work to ensure it still works. | 00:50 |
| *** daneyon has quit IRC | 00:51 | |
| tbarron | By "works" I mean it worked for LVM c-vol and multi-node devstack with c-back running on three nodes. | 00:51 |
| *** daneyon has joined #openstack-cinder | 00:52 | |
| tbarron | It won't work for c-vol backends that are doing special weird stuff in their backup and restore methods like vmware. | 00:52 |
| thingee | tbarron: mind pasting me your local.conf's ... I haven't done a multinode devstack yet. | 00:52 |
| jdurgin | lixiaoy1: re: full restore when backup['volume_id'] == volume['id'], I think that's a short circuit to tell that the volume already contains data | 00:52 |
| thingee | jdurgin: hi | 00:53 |
| jdurgin | thingee: hey | 00:53 |
| thingee | jdurgin: haven't seen you in a while :) | 00:53 |
| tbarron | For those we'd probably need to make some kind of pre-attach (or overide the attach I amd using) so that they can do whatever they do besides just make a file handle available. | 00:53 |
| jdurgin | thingee: I'm still around :) | 00:53 |
| tbarron | thingee: yeah, I'll send them to you. | 00:53 |
| thingee | tbarron: anything to make getting things tested easier for me is well appreciated. | 00:53 |
| tbarron | thingee: yeah, I know you are real busy, especially right now. | 00:54 |
| thingee | tbarron: unfortunately this is my list, but you're in there: image caching, volume migration, nested quotas, scaling backups | 00:54 |
| thingee | tbarron: just in the order people have bugged me :P | 00:54 |
| thingee | or me being interested in testing something | 00:54 |
| tbarron | thingee: so since we are talking about it, the drop dead date for continuing to work on this backup scaliing in Liberty requires that this backup bp-spce be approved Friday? | 00:55 |
| thingee | tbarron: cool, I should be fine with the spec. I have dinner and dance lessons soon, but should be back on later. | 00:55 |
| thingee | DuncanT: ^ | 00:55 |
| *** vilobhmm has quit IRC | 00:55 | |
| thingee | tbarron: yes | 00:55 |
| tbarron | thingee: I think there is productive investigation we can do iin this time frame, but can't promise that we'll get all the problems solved. | 00:56 |
| thingee | tbarron: I haven't figured out how I'm going to approve my own specs. Maybe have jgriffith do them | 00:56 |
| *** daneyon has quit IRC | 00:56 | |
| tbarron | thingee: I'll +1 :-) | 00:56 |
| tbarron | thingee: ballroom dancing? | 00:56 |
| *** zhonghua has joined #openstack-cinder | 00:56 | |
| thingee | tbarron: swing and rumba right now | 00:56 |
| tbarron | thingee: I'll let you go. I gotta call my Dad in a minute. | 00:57 |
| thingee | seeya | 00:57 |
| tbarron | I took swing and rumba but that's another story. | 00:57 |
| *** vilobhmm has joined #openstack-cinder | 00:57 | |
| openstackgerrit | Vilobh Meshram proposed openstack/cinder: Cinder Nested Quota Driver https://review.openstack.org/205369 | 00:58 |
| *** jasondotstar has joined #openstack-cinder | 00:58 | |
| *** h00327910__ has joined #openstack-cinder | 00:59 | |
| *** mriedem has joined #openstack-cinder | 00:59 | |
| *** vilobhmm1 has joined #openstack-cinder | 00:59 | |
| *** haomaiwang has quit IRC | 01:01 | |
| *** vilobhmm has quit IRC | 01:02 | |
| *** haomaiwang has joined #openstack-cinder | 01:02 | |
| *** Apoorva has quit IRC | 01:03 | |
| *** vilobhmm1 has quit IRC | 01:04 | |
| *** leeantho has quit IRC | 01:06 | |
| *** Strony has quit IRC | 01:08 | |
| *** zhonghua is now known as zhonghua-lee | 01:12 | |
| openstackgerrit | wanghao proposed openstack/cinder: Enhance deletion efficiency when backup init host https://review.openstack.org/206359 | 01:15 |
| thingee | scottda__: did we get the coffee dispenser thing figured out? | 01:21 |
| thingee | scottda__: trying to figure out if I need to buy some beans before I head out. | 01:21 |
| *** ganso_ has quit IRC | 01:25 | |
| openstackgerrit | Li Yingjun proposed openstack/cinder: Validate 'is_public' when creating volume type https://review.openstack.org/206784 | 01:25 |
| openstackgerrit | Li Yingjun proposed openstack/cinder: Add the ability to update type public status https://review.openstack.org/206788 | 01:25 |
| *** mylu has joined #openstack-cinder | 01:28 | |
| *** chlong has joined #openstack-cinder | 01:29 | |
| *** annashen has joined #openstack-cinder | 01:30 | |
| lixiaoy1 | :jdurgin Thanks | 01:33 |
| *** annashen has quit IRC | 01:35 | |
| lixiaoy1 | Hi guys. Anyone who know encryption feature? I noticed that I created a volume in ceph with encrption type, but it failed to attach to VM as no device_path in connection_info. Does ceph volume with encryption not been supported? | 01:35 |
| *** jasondotstar has quit IRC | 01:35 | |
| lixiaoy1 | I am not sure whether this is by design. if yes, I think we should prevent created ceph volume with encryption type. | 01:36 |
| *** akerr has quit IRC | 01:42 | |
| *** vilobhmm has joined #openstack-cinder | 01:47 | |
| *** vilobhmm1 has joined #openstack-cinder | 01:51 | |
| *** vilobhmm has quit IRC | 01:51 | |
| *** mylu has quit IRC | 02:00 | |
| *** haomaiwang has quit IRC | 02:01 | |
| *** mylu has joined #openstack-cinder | 02:01 | |
| *** haomaiwang has joined #openstack-cinder | 02:02 | |
| *** annashen has joined #openstack-cinder | 02:05 | |
| *** davechen has joined #openstack-cinder | 02:06 | |
| *** davechen is now known as davehcne | 02:06 | |
| *** davehcne is now known as davechen | 02:06 | |
| *** annegentle has quit IRC | 02:06 | |
| *** mylu has quit IRC | 02:07 | |
| openstackgerrit | WenPing Zhang proposed openstack/cinder: New cinder driver to support Daowoo YeeSan https://review.openstack.org/202440 | 02:14 |
| *** davechen1 has joined #openstack-cinder | 02:15 | |
| *** davechen has quit IRC | 02:17 | |
| *** davechen1 is now known as davechen | 02:17 | |
| *** dims has quit IRC | 02:19 | |
| *** dims has joined #openstack-cinder | 02:19 | |
| *** mylu has joined #openstack-cinder | 02:22 | |
| *** akshai has quit IRC | 02:23 | |
| *** akshai has joined #openstack-cinder | 02:23 | |
| *** akshai has quit IRC | 02:24 | |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function to manage/unmanage snapshots https://review.openstack.org/144590 | 02:25 |
| *** vilobhmm1 has quit IRC | 02:26 | |
| *** BharatK has quit IRC | 02:26 | |
| *** raunak has quit IRC | 02:27 | |
| *** mriedem1 has joined #openstack-cinder | 02:29 | |
| *** bill_az has quit IRC | 02:30 | |
| *** jasondotstar has joined #openstack-cinder | 02:31 | |
| *** annashen_ has joined #openstack-cinder | 02:31 | |
| *** mriedem has quit IRC | 02:31 | |
| *** annashen_ has quit IRC | 02:36 | |
| *** earlephilhower_ has joined #openstack-cinder | 02:39 | |
| *** earlephilhower has quit IRC | 02:41 | |
| *** jasondotstar has quit IRC | 02:41 | |
| *** BharatK has joined #openstack-cinder | 02:43 | |
| *** Apoorva has joined #openstack-cinder | 02:46 | |
| *** changbl has quit IRC | 02:47 | |
| *** annegentle has joined #openstack-cinder | 02:50 | |
| *** hakimo_ has joined #openstack-cinder | 02:52 | |
| *** hakimo has quit IRC | 02:54 | |
| *** breitz has quit IRC | 02:55 | |
| *** dims has quit IRC | 02:57 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-cinder | 03:01 | |
| *** earlephilhower_ is now known as earlephilhower | 03:04 | |
| *** changbl has joined #openstack-cinder | 03:07 | |
| *** breitz has joined #openstack-cinder | 03:07 | |
| *** jungleboyj has joined #openstack-cinder | 03:14 | |
| *** annegentle has quit IRC | 03:20 | |
| *** markvoelker_ has quit IRC | 03:20 | |
| openstackgerrit | LisaLi proposed openstack/cinder: Fix error message in cinder/api/v2/volumes.py https://review.openstack.org/207280 | 03:21 |
| *** markvoelker_ has joined #openstack-cinder | 03:22 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Adds the migration progress support for migration https://review.openstack.org/204953 | 03:24 |
| *** annegentle has joined #openstack-cinder | 03:26 | |
| *** annegentle has quit IRC | 03:26 | |
| *** kvidvans has quit IRC | 03:28 | |
| *** annashen_ has joined #openstack-cinder | 03:32 | |
| *** julim has quit IRC | 03:32 | |
| *** jwcroppe has quit IRC | 03:33 | |
| *** jwcroppe has joined #openstack-cinder | 03:34 | |
| *** laughterwym_ has joined #openstack-cinder | 03:36 | |
| *** laughterwym has quit IRC | 03:36 | |
| *** annashen_ has quit IRC | 03:37 | |
| *** jwcroppe has quit IRC | 03:38 | |
| *** annegentle has joined #openstack-cinder | 03:40 | |
| *** jamielennox|away is now known as jamielennox | 03:47 | |
| *** shyama has joined #openstack-cinder | 03:53 | |
| *** esker has joined #openstack-cinder | 03:53 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: Role based properties protection https://review.openstack.org/160090 | 03:58 |
| openstackgerrit | Dave Chen proposed openstack/cinder: Policies based properties protection https://review.openstack.org/160145 | 03:58 |
| *** coolsvap|away is now known as coolsvap | 03:59 | |
| *** mylu has quit IRC | 04:00 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** avishay has joined #openstack-cinder | 04:01 | |
| openstackgerrit | Teruaki Ishizaki proposed openstack/cinder: Add missing backup/restore methods to Sheepdog driver https://review.openstack.org/162927 | 04:01 |
| *** 16WAAEKLZ has joined #openstack-cinder | 04:02 | |
| *** annashen has quit IRC | 04:03 | |
| *** harlowja_at_home has joined #openstack-cinder | 04:13 | |
| *** lpetrut has joined #openstack-cinder | 04:14 | |
| *** salv-orlando has joined #openstack-cinder | 04:21 | |
| *** salv-orlando has quit IRC | 04:25 | |
| *** annashen has joined #openstack-cinder | 04:33 | |
| *** jamielennox is now known as jamielennox|away | 04:33 | |
| *** zhonghua-lee has quit IRC | 04:34 | |
| *** zhonghua-lee has joined #openstack-cinder | 04:36 | |
| *** annashen has quit IRC | 04:38 | |
| *** sakaYK has joined #openstack-cinder | 04:42 | |
| *** annegentle has quit IRC | 04:44 | |
| *** deepakcs has joined #openstack-cinder | 04:48 | |
| *** annashen has joined #openstack-cinder | 04:50 | |
| *** shausy has joined #openstack-cinder | 04:57 | |
| *** 16WAAEKLZ has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-cinder | 05:02 | |
| *** ub2 has quit IRC | 05:05 | |
| *** yrabl_afk has quit IRC | 05:06 | |
| *** shyama has quit IRC | 05:06 | |
| *** yrabl has joined #openstack-cinder | 05:06 | |
| *** harlowja_at_home has quit IRC | 05:10 | |
| *** shyama has joined #openstack-cinder | 05:11 | |
| *** sakaYK has quit IRC | 05:14 | |
| *** annashen has quit IRC | 05:18 | |
| *** amickus has joined #openstack-cinder | 05:21 | |
| *** laughterwym__ has joined #openstack-cinder | 05:32 | |
| *** annashen has joined #openstack-cinder | 05:34 | |
| *** laughterwym_ has quit IRC | 05:35 | |
| *** vilobhmm has joined #openstack-cinder | 05:36 | |
| *** shausy has quit IRC | 05:37 | |
| *** shausy has joined #openstack-cinder | 05:38 | |
| *** zhonghua-lee has quit IRC | 05:38 | |
| *** zhonghua-lee has joined #openstack-cinder | 05:39 | |
| *** annashen has quit IRC | 05:39 | |
| *** lpetrut has quit IRC | 05:41 | |
| *** Apoorva has quit IRC | 05:43 | |
| *** annashen has joined #openstack-cinder | 05:45 | |
| *** dannywilson has joined #openstack-cinder | 05:49 | |
| *** shyama has quit IRC | 05:49 | |
| *** lpetrut has joined #openstack-cinder | 05:52 | |
| *** salv-orlando has joined #openstack-cinder | 05:54 | |
| *** salv-orlando has quit IRC | 05:55 | |
| *** salv-orlando has joined #openstack-cinder | 05:56 | |
| *** agarciam has joined #openstack-cinder | 05:57 | |
| *** lpetrut has quit IRC | 05:59 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-cinder | 06:02 | |
| *** BharatK has quit IRC | 06:04 | |
| *** nkrinner has joined #openstack-cinder | 06:06 | |
| *** dannywilson has quit IRC | 06:11 | |
| *** dannywilson has joined #openstack-cinder | 06:12 | |
| *** esker has quit IRC | 06:13 | |
| *** agarciam has quit IRC | 06:14 | |
| *** agarciam has joined #openstack-cinder | 06:14 | |
| *** shausy has quit IRC | 06:15 | |
| *** shausy has joined #openstack-cinder | 06:15 | |
| *** zhonghua-lee has quit IRC | 06:20 | |
| *** smyz709 has joined #openstack-cinder | 06:23 | |
| *** abehl has joined #openstack-cinder | 06:24 | |
| *** sakaYK has joined #openstack-cinder | 06:25 | |
| *** smyz709 has quit IRC | 06:29 | |
| *** ankit_ag has joined #openstack-cinder | 06:29 | |
| *** rmcall has joined #openstack-cinder | 06:30 | |
| *** rmcall has quit IRC | 06:33 | |
| *** thingee has quit IRC | 06:33 | |
| *** thingee_ is now known as thingee | 06:33 | |
| *** annashen_ has joined #openstack-cinder | 06:35 | |
| openstackgerrit | XinXiaohui proposed openstack/cinder: capacity-headroom https://review.openstack.org/206923 | 06:36 |
| *** annashen_ has quit IRC | 06:39 | |
| *** vlaza has joined #openstack-cinder | 06:43 | |
| *** david-lyle has quit IRC | 06:43 | |
| *** Apoorva has joined #openstack-cinder | 06:43 | |
| *** agarciam has quit IRC | 06:45 | |
| *** agarciam has joined #openstack-cinder | 06:45 | |
| *** annashen has quit IRC | 06:46 | |
| *** shyama has joined #openstack-cinder | 06:46 | |
| *** BharatK has joined #openstack-cinder | 06:48 | |
| *** Apoorva has quit IRC | 06:48 | |
| *** vilobhmm has quit IRC | 06:49 | |
| openstackgerrit | Vincent Hou proposed openstack/python-cinderclient: Change cinderclient according to volume migration improvement https://review.openstack.org/189547 | 06:49 |
| *** david-lyle has joined #openstack-cinder | 06:49 | |
| *** BharatK has quit IRC | 06:51 | |
| *** BharatK has joined #openstack-cinder | 06:51 | |
| kjelly | Hi, guys, Does anyonoe know why “invisible_key” disappear in expected result? | 06:51 |
| kjelly | https://github.com/openstack/cinder/blob/master/cinder/tests/unit/api/v2/test_volumes.py#L1525 | 06:51 |
| *** bkopilov has quit IRC | 06:52 | |
| *** sgotliv has quit IRC | 06:55 | |
| *** bkopilov has joined #openstack-cinder | 06:55 | |
| *** vilobhmm has joined #openstack-cinder | 06:56 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Volume status management during migration https://review.openstack.org/186312 | 06:59 |
| *** haomaiwang has quit IRC | 07:01 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Adds the migration progress support for migration https://review.openstack.org/204953 | 07:02 |
| *** haomaiwang has joined #openstack-cinder | 07:02 | |
| *** rakesh_mishra has joined #openstack-cinder | 07:05 | |
| rakesh_mishra | Hello, | 07:05 |
| *** abhishekk has joined #openstack-cinder | 07:06 | |
| rakesh_mishra | can anyone tell me how i can upgrade from juno to kilo | 07:06 |
| *** belmoreira has joined #openstack-cinder | 07:10 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Adds the migration progress support for migration https://review.openstack.org/204953 | 07:12 |
| *** ronis has joined #openstack-cinder | 07:14 | |
| *** salv-orlando has quit IRC | 07:15 | |
| *** markus_z has joined #openstack-cinder | 07:16 | |
| *** Lee1092 has joined #openstack-cinder | 07:17 | |
| *** dannywilson has quit IRC | 07:18 | |
| *** lihit has joined #openstack-cinder | 07:19 | |
| *** ub has joined #openstack-cinder | 07:19 | |
| *** lihit has quit IRC | 07:19 | |
| *** kodoku has joined #openstack-cinder | 07:24 | |
| *** vilobhmm1 has joined #openstack-cinder | 07:24 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Deprecate vCenter version less than 5.1 https://review.openstack.org/207321 | 07:24 |
| *** chlong has quit IRC | 07:25 | |
| *** vilobhmm has quit IRC | 07:26 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Deprecate vCenter version less than 5.1 https://review.openstack.org/207321 | 07:28 |
| *** jwcroppe has joined #openstack-cinder | 07:29 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Deprecate vCenter version less than 5.1 https://review.openstack.org/207321 | 07:30 |
| *** vilobhmm1 has quit IRC | 07:31 | |
| *** BharatK_ has joined #openstack-cinder | 07:33 | |
| *** shyama has quit IRC | 07:34 | |
| *** BharatK has quit IRC | 07:35 | |
| *** annashen has joined #openstack-cinder | 07:35 | |
| -openstackstatus- NOTICE: Our CI system is broken again today, jobs are not getting processed at all. | 07:39 | |
| *** ChanServ changes topic to "Our CI system is broken again today, jobs are not getting processed at all." | 07:39 | |
| *** annashen has quit IRC | 07:40 | |
| *** rmcall has joined #openstack-cinder | 07:45 | |
| *** amickus has quit IRC | 07:46 | |
| *** ndipanov has quit IRC | 07:46 | |
| -openstackstatus- NOTICE: CI system is broken and very far behind. Please do not approve any changes for a while. | 07:49 | |
| *** ChanServ changes topic to "CI system is broken and very far behind. Please do not approve any changes for a while." | 07:49 | |
| *** sgotliv has joined #openstack-cinder | 07:53 | |
| *** jordanP has joined #openstack-cinder | 07:55 | |
| *** jamielennox|away is now known as jamielennox | 07:57 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-cinder | 08:02 | |
| *** shyama has joined #openstack-cinder | 08:09 | |
| *** markus_z has quit IRC | 08:09 | |
| *** rmcall has quit IRC | 08:12 | |
| *** kodoku has quit IRC | 08:17 | |
| *** zhonghua-lee has joined #openstack-cinder | 08:19 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add Cinder API WSGI app to run under Apache/Nginx https://review.openstack.org/200551 | 08:20 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add instructions on how deploy API under Apache https://review.openstack.org/207020 | 08:21 |
| openstackgerrit | Pranali Deore proposed openstack/cinder: Validate name and description string https://review.openstack.org/182994 | 08:24 |
| openstackgerrit | Pranali Deore proposed openstack/cinder: Validate string, integer limit for input parameter https://review.openstack.org/194968 | 08:24 |
| *** jistr has joined #openstack-cinder | 08:25 | |
| *** kodoku has joined #openstack-cinder | 08:25 | |
| *** laughterwym__ has quit IRC | 08:28 | |
| *** zhonghua-li has joined #openstack-cinder | 08:29 | |
| DuncanT | aarefiev: Any idea who is working on CI for the WSGI deploy? | 08:29 |
| openstackgerrit | Abhishek Shrivastava proposed openstack/cinder: Adding delete-wait-loop for CloudByte Volumes https://review.openstack.org/201059 | 08:30 |
| DuncanT | aarefiev: It was something discussed during the spec phase as being a requirement | 08:30 |
| aarefiev | DuncanT: maybe I will or e0ne | 08:31 |
| DuncanT | aarefiev: Thanks. I just wanted to ensure it was on somebody's road map. At a glance the patches look fine, I'll take a proper look later | 08:31 |
| *** zhonghua-lee has quit IRC | 08:33 | |
| aarefiev | DuncanT: ok, thanks, I've made patch for devstack, so I think CI shouldn't be a problem | 08:33 |
| DuncanT | aarefiev: Perfect | 08:33 |
| *** afazekas has joined #openstack-cinder | 08:33 | |
| *** annashen has joined #openstack-cinder | 08:36 | |
| *** jwcroppe has quit IRC | 08:37 | |
| *** salv-orlando has joined #openstack-cinder | 08:37 | |
| *** jwcroppe has joined #openstack-cinder | 08:38 | |
| aarefiev | DuncanT: about https://review.openstack.org/#/c/143645/ | 08:39 |
| aarefiev | I have some doubts about tenant id validation after speaking with you | 08:39 |
| *** annashen has quit IRC | 08:41 | |
| DuncanT | aarefiev: It's not a trivial problem, certainly | 08:41 |
| DuncanT | aarefiev: On admin-only paths it is a little less worrying, but there's no guarantee that the tenant with cinder admin rights has enough keystone rights to validate the tenant | 08:42 |
| *** jwcroppe has quit IRC | 08:42 | |
| DuncanT | aarefiev: We /could/ use a magic service tenant to do it, but I don't want to see it in none-admin paths since blocking calls in the API lead to DoS conditions easily | 08:43 |
| aarefiev | DuncanT: yea, also we should think about more generally approach here, possibly nova has the same problem | 08:43 |
| johnthetubaguy | DuncanT: that has come up, we should co-ordinate around that, ideally | 08:44 |
| johnthetubaguy | swift and glance have a concept of swift resources needing both the tenant token and a service token to access them | 08:44 |
| johnthetubaguy | Nova has the concept of a "locked" resource, i.e. you can see it but not alter it in any way | 08:45 |
| johnthetubaguy | so we were thinking of combining the two, do you "lock" a Nova resources with a service token, or something like that | 08:45 |
| johnthetubaguy | so to boot/snapshot/delete the instance you would need to have both the service token and the users token | 08:46 |
| DuncanT | johnthetubaguy: The service token stuff we can use for swift access during backup too - I'd hoped to find time to implement it this cycle - I had some some input on the service token design on the swift end to make them suitable for our usecase | 08:46 |
| johnthetubaguy | although honestly, we are a long way from an agreed approach at this point | 08:46 |
| johnthetubaguy | DuncanT: ah, interesting, sounds like we are going in similar directions, although we are quite a long way from agreement at this point, in my head I am targeting getting a spec merged/proposed before the Mitaka summit | 08:47 |
| *** lpetrut has joined #openstack-cinder | 08:47 | |
| *** aix has joined #openstack-cinder | 08:47 | |
| DuncanT | johnthetubaguy: I'm sure it can be used to improve our nova interactions too - some calls to cinder should only ever come via nova (reserve, init_connection, etc) on a normal setup, never direct from the tenant, and nova has some api that cinder calls (assisted snap among them) that again should only ever come via cinder | 08:48 |
| johnthetubaguy | DuncanT: +100 | 08:48 |
| aarefiev | DuncanT: +1 | 08:48 |
| johnthetubaguy | DuncanT: I would love for the service token to be used, so it doesn't matter if a users token is expired, its still accepted as previously valid (its more of an issue with glance, honestly) | 08:48 |
| *** salv-orlando has quit IRC | 08:49 | |
| DuncanT | johnthetubaguy: The swift work should be far easier since it is 90% 'copy what glance did' | 08:49 |
| johnthetubaguy | yes, that makes sense | 08:49 |
| DuncanT | johnthetubaguy: I don't *think* you get that behaviour currently, you need to add a trust into the mix for that, I believe | 08:49 |
| johnthetubaguy | I feel a cross project session coming on, I think there is an etherpad for those now | 08:49 |
| johnthetubaguy | DuncanT: yeah, you are right, I think you do currently | 08:50 |
| johnthetubaguy | DuncanT: I got my dreaming hat on, where security issues seems to melt away, oops | 08:50 |
| DuncanT | johnthetubaguy: Trusts and backup are something else people are working on - there's a draft patch around for it somewhere | 08:50 |
| DuncanT | johnthetubaguy: We should probably use them for glance too at some point | 08:50 |
| johnthetubaguy | DuncanT: yeah, I need to get my nose into that, the policy stuff is similar | 08:50 |
| johnthetubaguy | DuncanT: we are talking about a new lib to talk to glance that could deal with all this, to stop the repeated code between cinder and glance, but I have to run in a bit, be I should try write some of this up on the ML next week (I hope I remember to do that) | 08:51 |
| johnthetubaguy | (or someone else does it :) ) | 08:51 |
| DuncanT | johnthetubaguy: I've been ignoring policy for a while | 08:52 |
| DuncanT | johnthetubaguy: I'll keep an eye out for the post. Please tag it as cinder, since I'll see it far faster that way | 08:52 |
| johnthetubaguy | DuncanT: yeah, good point | 08:53 |
| openstackgerrit | Kuo-tung Kao proposed openstack/cinder: Don't use context.elevated to get volume. https://review.openstack.org/206450 | 08:53 |
| *** zhonghua-li has quit IRC | 08:53 | |
| *** lixiaoy1 has quit IRC | 08:56 | |
| openstackgerrit | wanghao proposed openstack/python-cinderclient: Add to_dict method to Resource class https://review.openstack.org/207340 | 08:57 |
| aarefiev | DuncanT: could you take a look on https://review.openstack.org/#/c/165403/ once more time, please | 08:58 |
| *** ChanServ changes topic to "The OpenStack Block Storage Project Cinder | The New Kids On the Block | https://wiki.openstack.org/wiki/Cinder" | 08:58 | |
| -openstackstatus- NOTICE: CI is back online but has a huge backlog. Please be patient and if possible delay approving changes until it has caught up. | 08:58 | |
| kjelly | hi, guys. In the return value of “self.db.volume_get”, why the type of volumes[“volume_metadata”] is list of “cinder.db.sqlalchemy.models.VolumeMetadata” instead of dict ? The api expose the implement to caller. | 08:59 |
| kjelly | https://github.com/openstack/cinder/blob/master/cinder/volume/api.py#L407 | 08:59 |
| *** rakesh_mishra has quit IRC | 08:59 | |
| kjelly | the volume looks like {'migration_status': None, 'provider_id': None, 'availability_zone': u'nova', 'terminated_at': None, 'updated_at': datetime.datetime(2015, 7, 30, 2, 15, 25), 'provider_geometry': None, 'replication_extended_status': None, 'replication_status': u'disabled', 'snapshot_id': None, 'ec2_id': None, 'deleted_at': None, 'id': u'aa71155a-d17d-481a-bdb9-af7b97654ff2', 'size': 1L, 'user_id': u'd91955965f7a47ddbcc8f7c6a98ee488', 'volume_ | 09:00 |
| kjelly | attachment': [], 'display_description': None, 'volume_admin_metadata': [<cinder.db.sqlalchemy.models.VolumeAdminMetadata object at 0x7f57b3389350>], 'encryption_key_id': None, 'project_id': u'da3bbc6e175541b0ac1952135838bf30', 'launched_at': datetime.datetime(2015, 7, 30, 2, 15, 25), 'scheduled_at': datetime.datetime(2015, 7, 30, 2, 15, 24), 'status': u'available', 'volume_type_id': u'03357ed8-567e-4755-8cc4-8cd1ba725e7f', 'multiattach': False, ' | 09:00 |
| kjelly | deleted': False, 'provider_location': None, 'host': u'user-GA-6PXSV3@lvm#lvm', 'consistencygroup_id': None, 'source_volid': None, 'provider_auth': None, 'display_name': u'a', 'bootable': False, 'created_at': datetime.datetime(2015, 7, 30, 2, 15, 24), 'attach_status': u'detached', 'volume_type': <cinder.db.sqlalchemy.models.VolumeTypes object at 0x7f57b33dc990>, 'consistencygroup': None, '_name_id': None, 'volume_metadata': [<cinder.db.sqlalchemy. | 09:00 |
| kjelly | models.VolumeMetadata object at 0x7f57b33fd490>, <cinder.db.sqlalchemy.models.VolumeMetadata object at 0x7f57b3389910>], 'replication_driver_data': None} | 09:00 |
| *** haomaiwang has quit IRC | 09:01 | |
| *** ndipanov has joined #openstack-cinder | 09:01 | |
| *** haomaiwang has joined #openstack-cinder | 09:02 | |
| *** e0ne has joined #openstack-cinder | 09:03 | |
| *** chlong has joined #openstack-cinder | 09:07 | |
| *** bluex-pl has joined #openstack-cinder | 09:09 | |
| openstackgerrit | wanghao proposed openstack/cinder: Add ability to filter by volume_glance_metadata https://review.openstack.org/147738 | 09:11 |
| openstackgerrit | wanghao proposed openstack/python-cinderclient: Add to_dict method to Resource class https://review.openstack.org/207340 | 09:12 |
| *** kodoku has quit IRC | 09:20 | |
| *** jwcroppe has joined #openstack-cinder | 09:20 | |
| *** jwcroppe has quit IRC | 09:21 | |
| *** jwcroppe has joined #openstack-cinder | 09:21 | |
| *** jwcroppe_ has joined #openstack-cinder | 09:23 | |
| *** jwcroppe_ has quit IRC | 09:23 | |
| *** jwcroppe_ has joined #openstack-cinder | 09:24 | |
| *** jamielennox is now known as jamielennox|away | 09:25 | |
| *** sasi has joined #openstack-cinder | 09:26 | |
| *** jwcroppe has quit IRC | 09:27 | |
| *** jwcroppe_ has quit IRC | 09:28 | |
| *** sasi has quit IRC | 09:30 | |
| *** ig0r_ has joined #openstack-cinder | 09:32 | |
| *** sakaYK has quit IRC | 09:34 | |
| *** ishant has joined #openstack-cinder | 09:35 | |
| *** annashen has joined #openstack-cinder | 09:37 | |
| *** salv-orlando has joined #openstack-cinder | 09:38 | |
| *** jwcroppe has joined #openstack-cinder | 09:38 | |
| kjelly | hi, guys. In volume_get, the return value contains some object whose type is cinder.db.sqlalchemy.models.VolumeMetadata. Should I report bug for it? | 09:39 |
| *** jwcroppe has quit IRC | 09:40 | |
| *** annashen has quit IRC | 09:42 | |
| kjelly | https://github.com/openstack/cinder/blob/master/cinder/db/sqlalchemy/api.py#L1336 | 09:46 |
| *** davechen has left #openstack-cinder | 09:46 | |
| *** Apoorva has joined #openstack-cinder | 09:46 | |
| *** kjelly has quit IRC | 09:49 | |
| *** dims has joined #openstack-cinder | 09:50 | |
| *** Apoorva has quit IRC | 09:51 | |
| e0ne | kjelly: it's strange. this methid shoud return Volume object | 09:51 |
| *** coolsvap is now known as coolsvap|away | 09:55 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** alexpilotti has joined #openstack-cinder | 10:01 | |
| *** haomaiwang has joined #openstack-cinder | 10:01 | |
| deepakcs | e0ne: when u have some time, have implemented ur commetns. https://review.openstack.org/#/q/status:open+project:openstack/cinder+branch:master+topic:bp/assisted-snapshot-improvements,n,z | 10:06 |
| *** zhonghua-li has joined #openstack-cinder | 10:06 | |
| *** shyama has quit IRC | 10:10 | |
| e0ne | deepakcs: thanks. i added 'recheck' comment to run ci again | 10:10 |
| deepakcs | e0ne: given the state of CI, I didn't feel we should do a recheck :) | 10:11 |
| e0ne | deepakcs: :) | 10:11 |
| deepakcs | e0ne: anyways, can u review the patches, until then hopefully CI gets back to normak | 10:11 |
| deepakcs | *normal | 10:11 |
| e0ne | deepakcs: sure. i'll review your patches later today | 10:12 |
| *** shausy has quit IRC | 10:12 | |
| *** shausy has joined #openstack-cinder | 10:13 | |
| *** jasondotstar has joined #openstack-cinder | 10:15 | |
| geguileo | e0ne: Have a minute to discuss the backup import issue? | 10:19 |
| e0ne | geguileo: yep | 10:19 |
| geguileo | e0ne: I have done tests and it works r:-?? | 10:19 |
| e0ne | geguileo: did you change anything in your patch? | 10:20 |
| geguileo | e0ne: No | 10:20 |
| geguileo | e0ne: What were your steps? | 10:20 |
| e0ne | geguileo: o_0 | 10:20 |
| geguileo | e0ne: I think you are actually deleting the backup | 10:20 |
| geguileo | e0ne: cinder backup-delete BACKUP-ID | 10:20 |
| e0ne | geguileo: 2minutes, plz | 10:21 |
| geguileo | e0ne: And then you are trying to restore it | 10:21 |
| e0ne | geguileo: i'' delete volumes and backupd and try again | 10:21 |
| geguileo | e0ne: If you delete the backup then there's nothing to restore from | 10:21 |
| *** kbyrne has quit IRC | 10:21 | |
| geguileo | e0ne: You have deleted the backup | 10:21 |
| geguileo | e0ne: it's like deleting a volume and then trying to do something with it | 10:22 |
| e0ne | geguileo: oops | 10:22 |
| e0ne | geguileo: that was i did | 10:22 |
| geguileo | e0ne: You need to delete the row in the DB | 10:22 |
| geguileo | e0ne: Manually | 10:22 |
| e0ne | geguileo: you're right | 10:22 |
| e0ne | geguileo: please, give me few minutes to test it | 10:23 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder-specs: Rootwrap daemon mode https://review.openstack.org/137318 | 10:23 |
| geguileo | e0ne: Oh, as long as you need, there's no hurry | 10:23 |
| e0ne | geguileo: and i'll be happy to remove -1 from your patch | 10:23 |
| *** kbyrne has joined #openstack-cinder | 10:23 | |
| geguileo | e0ne: I just wanted to confirm that that's what you had done :-) | 10:23 |
| openstackgerrit | Xiangfei Zhu proposed openstack/cinder: VMware: Fix re-attach volume error for VC 5.1 https://review.openstack.org/207380 | 10:24 |
| e0ne | geguileo: you're absolutly right! | 10:27 |
| e0ne | geguileo: i failed | 10:27 |
| e0ne | geguileo: thank you for pointing me where i was wrong. removed -1 | 10:29 |
| e0ne | geguileo: i'll review your patch once again a bit later | 10:29 |
| geguileo | e0ne: Thanks | 10:30 |
| e0ne | geguileo: thanks you for clarification and fixing it | 10:30 |
| geguileo | e0ne: np | 10:30 |
| *** jwcroppe has joined #openstack-cinder | 10:32 | |
| *** shausy has quit IRC | 10:32 | |
| *** sgotliv has quit IRC | 10:37 | |
| *** annashen has joined #openstack-cinder | 10:38 | |
| *** shausy has joined #openstack-cinder | 10:38 | |
| *** shyama has joined #openstack-cinder | 10:39 | |
| *** aix has quit IRC | 10:39 | |
| openstackgerrit | Deepak C Shetty (deepakcs) proposed openstack/cinder: Cinder part of volume snapshot improvements https://review.openstack.org/172808 | 10:41 |
| openstackgerrit | Deepak C Shetty (deepakcs) proposed openstack/cinder: db: Add support for snapshot_admin_metadata https://review.openstack.org/203979 | 10:41 |
| *** annashen has quit IRC | 10:42 | |
| openstackgerrit | Vipin Balachandran proposed openstack/cinder: VMware: Fix protocol in backend stats https://review.openstack.org/207388 | 10:44 |
| *** msno has joined #openstack-cinder | 10:45 | |
| msno | did anyone had any issue configuring netapp NFS with Juno release of openstack | 10:46 |
| msno | Juno + Cinder + NetAPP + NFS | 10:46 |
| msno | the vol creation is not working for me when suing NetAPP with NFS | 10:46 |
| *** heyun has quit IRC | 10:46 | |
| msno | where as manual mounting of the NFS share to the controller node is working fine | 10:46 |
| openstackgerrit | Sergey Vilgelm proposed openstack/cinder: Switch to the oslo_utils.fileutils https://review.openstack.org/199537 | 10:47 |
| *** zhonghua-li has quit IRC | 10:51 | |
| *** zhonghua-li has joined #openstack-cinder | 10:51 | |
| *** sgotliv has joined #openstack-cinder | 10:51 | |
| *** timcl has joined #openstack-cinder | 10:55 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Adds the migration progress support for migration https://review.openstack.org/204953 | 11:00 |
| *** haomaiwang has quit IRC | 11:01 | |
| *** dims_ has joined #openstack-cinder | 11:02 | |
| *** svilgelm has joined #openstack-cinder | 11:02 | |
| *** jwcroppe has quit IRC | 11:02 | |
| *** dims has quit IRC | 11:02 | |
| *** 7GHAAT9JB has joined #openstack-cinder | 11:02 | |
| *** zhenguo has quit IRC | 11:04 | |
| tbarron | msno: does c-vol log show that the mount is failing? | 11:07 |
| tbarron | msno: there is not some general issue with cinder + netapp + juno + nfs, so something needs to be re-configured. | 11:08 |
| tbarron | msno: maybe take the discussion to #openstack-netapp ... | 11:09 |
| msno | tbarron, i alsready asked there.. let me try again | 11:09 |
| tbarron | msno: a lot of us are in the US, just waking up :-) | 11:10 |
| msno | tbarron, yes .. i know :) thats why i am waiting | 11:10 |
| *** sgotliv has quit IRC | 11:12 | |
| *** aix has joined #openstack-cinder | 11:13 | |
| *** shyama has quit IRC | 11:13 | |
| tbarron | msno: well, I'm getting a cup of coffee and heading over to that channel | 11:13 |
| msno | tbarron, sure.. meanwhile .. let me reproduce the issue | 11:14 |
| msno | if not today.. i will get in touch with you later | 11:14 |
| msno | in netapp channel | 11:14 |
| *** svilgelm has left #openstack-cinder | 11:15 | |
| tbarron | msno: cool. you should be able to get the mount to work manually then use the same nfs options in your cinder.conf | 11:16 |
| tbarron | msno: assuming that the problem is with cinder doing the mount successfully | 11:16 |
| msno | tbarron, sure .. let me check .. | 11:16 |
| msno | btw: does netapp NFS support both NFS V 4 and V3? | 11:16 |
| msno | tbarron, as per my debbugging .. the vol creation is getting failed when its doen via cinder.. Without cinder .. its working fine | 11:17 |
| tbarron | msno: both v4 and v3 are supported | 11:18 |
| msno | tbarron, ok | 11:18 |
| tbarron | msno: the filers have config options though and some admins turn off one or the other | 11:18 |
| msno | tbarron, ok.. i will check those | 11:19 |
| *** ganso_ has joined #openstack-cinder | 11:19 | |
| *** jasondotstar has quit IRC | 11:19 | |
| tbarron | the c-vol logs will show if the mount succeeds or fails from cinder. That is likely the first thing to check. | 11:20 |
| msno | tbarron, ok | 11:20 |
| *** sgotliv has joined #openstack-cinder | 11:24 | |
| *** zhonghua-li has quit IRC | 11:25 | |
| *** deepakcs has quit IRC | 11:25 | |
| *** salv-orl_ has joined #openstack-cinder | 11:25 | |
| *** salv-orlando has quit IRC | 11:26 | |
| *** cdelatte has quit IRC | 11:35 | |
| *** annashen has joined #openstack-cinder | 11:39 | |
| openstackgerrit | Daniel Tadrzak proposed openstack/cinder: ConsistencyGroup Object https://review.openstack.org/163641 | 11:41 |
| *** annashen has quit IRC | 11:43 | |
| *** aagrawal has joined #openstack-cinder | 11:45 | |
| *** jordanP has quit IRC | 11:46 | |
| *** marcusvrn has joined #openstack-cinder | 11:47 | |
| *** e0ne has quit IRC | 11:47 | |
| *** jordanP has joined #openstack-cinder | 11:47 | |
| openstackgerrit | Daniel Tadrzak proposed openstack/cinder: CGSnapshot Object https://review.openstack.org/195027 | 11:50 |
| *** cdelatte has joined #openstack-cinder | 11:51 | |
| *** delattec has joined #openstack-cinder | 11:51 | |
| openstackgerrit | Vincent Hou proposed openstack/cinder: Adds the migration progress support for migration https://review.openstack.org/204953 | 11:53 |
| *** bswartz has quit IRC | 11:57 | |
| *** aagrawal has quit IRC | 11:58 | |
| *** aagrawal has joined #openstack-cinder | 11:58 | |
| *** 7GHAAT9JB has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-cinder | 12:02 | |
| *** cdelatte has quit IRC | 12:03 | |
| *** delattec has quit IRC | 12:03 | |
| *** msno has quit IRC | 12:04 | |
| *** lpabon has joined #openstack-cinder | 12:07 | |
| *** shyama has joined #openstack-cinder | 12:07 | |
| *** akerr has joined #openstack-cinder | 12:09 | |
| *** agarciam has quit IRC | 12:10 | |
| *** ishant has quit IRC | 12:11 | |
| openstackgerrit | Li Yingjun proposed openstack/cinder: Validate 'is_public' when creating volume type https://review.openstack.org/206784 | 12:12 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Move import and export backup metadata to object https://review.openstack.org/200927 | 12:15 |
| openstackgerrit | Gorka Eguileor proposed openstack/cinder: Fix backup metadata import missing fields https://review.openstack.org/183222 | 12:15 |
| *** aagrawal has quit IRC | 12:18 | |
| *** aagrawal has joined #openstack-cinder | 12:18 | |
| *** edmondsw has joined #openstack-cinder | 12:19 | |
| *** eharney has quit IRC | 12:21 | |
| *** jasondotstar has joined #openstack-cinder | 12:22 | |
| *** jwcroppe has joined #openstack-cinder | 12:23 | |
| *** ericksonsantos has joined #openstack-cinder | 12:25 | |
| *** aagrawal has quit IRC | 12:30 | |
| *** aagrawal has joined #openstack-cinder | 12:30 | |
| *** jamielennox|away is now known as jamielennox | 12:35 | |
| *** bswartz has joined #openstack-cinder | 12:39 | |
| *** annashen has joined #openstack-cinder | 12:40 | |
| *** aagrawal has quit IRC | 12:41 | |
| *** aagrawal has joined #openstack-cinder | 12:41 | |
| *** annashen has quit IRC | 12:44 | |
| *** markstur_ has joined #openstack-cinder | 12:48 | |
| *** dustins has joined #openstack-cinder | 12:48 | |
| *** Apoorva has joined #openstack-cinder | 12:48 | |
| *** aagrawal has quit IRC | 12:50 | |
| *** aagrawal has joined #openstack-cinder | 12:50 | |
| *** Apoorva has quit IRC | 12:53 | |
| *** e0ne has joined #openstack-cinder | 12:54 | |
| *** salv-orl_ has quit IRC | 12:54 | |
| *** nkrinner has quit IRC | 12:55 | |
| *** aagrawal has quit IRC | 12:55 | |
| *** aagrawal has joined #openstack-cinder | 12:55 | |
| *** aagrawal has quit IRC | 12:59 | |
| *** aagrawal has joined #openstack-cinder | 13:00 | |
| *** haomaiwang has quit IRC | 13:01 | |
| *** haomaiwang has joined #openstack-cinder | 13:02 | |
| *** markstur_ has left #openstack-cinder | 13:03 | |
| *** Yogi1 has joined #openstack-cinder | 13:03 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Add support '--all-tenants' argument for admin for cinder backup-list https://review.openstack.org/207451 | 13:04 |
| *** akshai has joined #openstack-cinder | 13:04 | |
| *** markus_z has joined #openstack-cinder | 13:05 | |
| *** jamielennox is now known as jamielennox|away | 13:06 | |
| *** eharney has joined #openstack-cinder | 13:06 | |
| *** julim has joined #openstack-cinder | 13:10 | |
| *** ig0r_ has quit IRC | 13:10 | |
| *** markus_z has quit IRC | 13:12 | |
| *** lpabon has quit IRC | 13:13 | |
| *** ig0r_ has joined #openstack-cinder | 13:16 | |
| *** rushiagr_away is now known as rushiagr | 13:17 | |
| *** aagrawal has quit IRC | 13:19 | |
| *** dims_ has quit IRC | 13:20 | |
| *** dims has joined #openstack-cinder | 13:20 | |
| *** e0ne has quit IRC | 13:22 | |
| *** dustins_ has joined #openstack-cinder | 13:22 | |
| *** dustins has quit IRC | 13:22 | |
| *** dustins_ is now known as dustins | 13:22 | |
| *** lpabon has joined #openstack-cinder | 13:22 | |
| *** Guest10__ has joined #openstack-cinder | 13:23 | |
| *** bill_az has joined #openstack-cinder | 13:25 | |
| *** kbyrne has quit IRC | 13:26 | |
| *** kvidvans has joined #openstack-cinder | 13:26 | |
| *** jwcroppe has quit IRC | 13:26 | |
| *** jwcroppe has joined #openstack-cinder | 13:27 | |
| *** kbyrne has joined #openstack-cinder | 13:27 | |
| *** mylu has joined #openstack-cinder | 13:27 | |
| *** jasondotstar has quit IRC | 13:30 | |
| *** ig0r_ has quit IRC | 13:31 | |
| *** e0ne has joined #openstack-cinder | 13:31 | |
| *** jwcroppe has quit IRC | 13:32 | |
| *** xyang1 has joined #openstack-cinder | 13:35 | |
| *** annashen has joined #openstack-cinder | 13:40 | |
| *** annegentle has joined #openstack-cinder | 13:43 | |
| *** annashen has quit IRC | 13:45 | |
| *** abhishekk has quit IRC | 13:46 | |
| *** salv-orlando has joined #openstack-cinder | 13:49 | |
| *** jaypipes has joined #openstack-cinder | 13:53 | |
| *** cbader has joined #openstack-cinder | 13:54 | |
| *** krtaylor has quit IRC | 13:55 | |
| *** annegentle has quit IRC | 13:55 | |
| *** jungleboyj has quit IRC | 13:56 | |
| *** haomaiwang has quit IRC | 14:01 | |
| *** haomaiwang has joined #openstack-cinder | 14:02 | |
| *** ankit_ag has quit IRC | 14:04 | |
| *** markvoelker_ has quit IRC | 14:06 | |
| *** thangp has joined #openstack-cinder | 14:07 | |
| *** krtaylor has joined #openstack-cinder | 14:08 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/python-cinderclient: Add support '--all-tenants' argument for admin for cinder backup-list https://review.openstack.org/207469 | 14:09 |
| openstackgerrit | abhiram moturi proposed openstack/cinder: Oracle ZFSSA iSCSI driver to return 'free' instead of 'available' space https://review.openstack.org/207106 | 14:12 |
| openstackgerrit | Yuriy Nesenenko proposed openstack/python-cinderclient: Add support '--all-tenants' argument for admin for cinder backup-list https://review.openstack.org/207469 | 14:12 |
| *** liuxinguo has joined #openstack-cinder | 14:16 | |
| *** breitz has quit IRC | 14:22 | |
| *** annegentle has joined #openstack-cinder | 14:24 | |
| *** markvoelker has joined #openstack-cinder | 14:27 | |
| *** markvoelker_ has joined #openstack-cinder | 14:28 | |
| *** eharney has quit IRC | 14:29 | |
| *** jungleboyj has joined #openstack-cinder | 14:30 | |
| *** ig0r_ has joined #openstack-cinder | 14:32 | |
| *** markvoelker has quit IRC | 14:32 | |
| *** abehl has quit IRC | 14:32 | |
| *** garthb has joined #openstack-cinder | 14:34 | |
| *** BharatK_ has quit IRC | 14:35 | |
| openstackgerrit | Matthew Edmonds proposed openstack/cinder: mark oslo.vmware as optional requirement https://review.openstack.org/203237 | 14:38 |
| *** annashen has joined #openstack-cinder | 14:41 | |
| *** salv-orlando has quit IRC | 14:42 | |
| *** bapalm_ has joined #openstack-cinder | 14:43 | |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX Create CG from CG Snapshot https://review.openstack.org/195280 | 14:43 |
| *** agarciam has joined #openstack-cinder | 14:43 | |
| *** vlaza has quit IRC | 14:44 | |
| *** eharney has joined #openstack-cinder | 14:44 | |
| *** annashen has quit IRC | 14:46 | |
| *** BharatK_ has joined #openstack-cinder | 14:46 | |
| *** kjelly_ has joined #openstack-cinder | 14:48 | |
| *** kjelly_ is now known as kjelly | 14:48 | |
| *** timcl1 has joined #openstack-cinder | 14:49 | |
| *** mtanino has joined #openstack-cinder | 14:50 | |
| *** lcurtis has joined #openstack-cinder | 14:52 | |
| *** timcl has quit IRC | 14:52 | |
| *** afazekas has quit IRC | 14:55 | |
| *** breitz has joined #openstack-cinder | 14:55 | |
| *** avishay has quit IRC | 14:57 | |
| *** avishay has joined #openstack-cinder | 14:57 | |
| *** BharatK_ has quit IRC | 14:59 | |
| *** rmcall has joined #openstack-cinder | 15:00 | |
| *** haomaiwang has quit IRC | 15:01 | |
| *** krtaylor has quit IRC | 15:01 | |
| *** haomaiwang has joined #openstack-cinder | 15:02 | |
| *** edtubill has joined #openstack-cinder | 15:02 | |
| *** ig0r_ has quit IRC | 15:03 | |
| *** primechuck has joined #openstack-cinder | 15:03 | |
| *** rmcall has quit IRC | 15:03 | |
| *** ig0r__ has joined #openstack-cinder | 15:04 | |
| *** logan2 has quit IRC | 15:08 | |
| *** ig0r__ has quit IRC | 15:09 | |
| *** ig0r_ has joined #openstack-cinder | 15:09 | |
| *** crose has joined #openstack-cinder | 15:09 | |
| mriedem1 | hemnafk: thingee: fyi, this is going to be a release blocker for nova https://bugs.launchpad.net/devstack/+bug/1479842 | 15:10 |
| openstack | Launchpad bug 1479842 in OpenStack Compute (nova) "os-brick needs to provide it's own rootwrap filters file" [High,Confirmed] | 15:10 |
| mriedem1 | we're letting some things into nova now under the assertion that we need to resolve that rootwrap filters bug before liberty release | 15:11 |
| *** ig0r__ has joined #openstack-cinder | 15:13 | |
| *** krtaylor has joined #openstack-cinder | 15:14 | |
| *** ig0r__ has quit IRC | 15:15 | |
| *** ig0r_ has quit IRC | 15:15 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Add support '--all-tenants' argument for cinder backup-list https://review.openstack.org/207451 | 15:16 |
| Swanson | I need to set up a CI test bed. Anyone got a favorite link or blog post or something on how to do this? I don't want it to report so I don't need any of that. It just needs to fire off as patches come in. | 15:17 |
| *** diogogmt has joined #openstack-cinder | 15:17 | |
| *** dannywilson has joined #openstack-cinder | 15:17 | |
| *** timcl1 has quit IRC | 15:19 | |
| e0ne | smcginnis: hi. around? | 15:19 |
| openstackgerrit | Yuriy Nesenenko proposed openstack/python-cinderclient: Add support '--all-tenants' argument for cinder backup-list https://review.openstack.org/207469 | 15:21 |
| *** dannywil_ has joined #openstack-cinder | 15:22 | |
| *** dannywil_ has quit IRC | 15:22 | |
| *** timcl has joined #openstack-cinder | 15:22 | |
| *** dannywil_ has joined #openstack-cinder | 15:22 | |
| *** dannywi__ has joined #openstack-cinder | 15:23 | |
| *** ronis has quit IRC | 15:23 | |
| *** dannywi__ has quit IRC | 15:24 | |
| *** dannywilson has quit IRC | 15:24 | |
| *** IlyaG has joined #openstack-cinder | 15:24 | |
| *** garthb_ has joined #openstack-cinder | 15:25 | |
| *** garthb has quit IRC | 15:25 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/python-cinderclient: Set default OS_VOLUME_API_VERSION to '2' https://review.openstack.org/146541 | 15:25 |
| *** dannywi__ has joined #openstack-cinder | 15:26 | |
| *** dannywil_ has quit IRC | 15:27 | |
| *** cppforlife_ has quit IRC | 15:29 | |
| cebruns | Swanson: Have a look at: http://specs.openstack.org/openstack-infra/infra-specs/specs/openstackci.html and http://docs.openstack.org/infra/system-config/third_party.html | 15:29 |
| *** cebruns_ has quit IRC | 15:29 | |
| *** Lee1092 has quit IRC | 15:29 | |
| *** crose has quit IRC | 15:30 | |
| *** crose has joined #openstack-cinder | 15:30 | |
| *** agarciam has quit IRC | 15:30 | |
| xyang | hemnafk: I can help you with those bugs | 15:30 |
| *** scottda__ has left #openstack-cinder | 15:31 | |
| *** zul has quit IRC | 15:31 | |
| *** dannywi__ has quit IRC | 15:32 | |
| scottda | thingee: I haven't gotten the coffee thermos sorted. I won't be in the office until Monday, so that'd be the earliest I can get it figured out. | 15:33 |
| xyang1 | e0ne: hi, can you take a look of my reply here: https://review.openstack.org/#/c/203473/8/cinder/consistencygroup/api.py | 15:33 |
| *** dannywilson has joined #openstack-cinder | 15:34 | |
| Swanson | cebruns: Thanks! | 15:34 |
| *** dannywilson has quit IRC | 15:35 | |
| *** dannywilson has joined #openstack-cinder | 15:36 | |
| *** timcl1 has joined #openstack-cinder | 15:39 | |
| *** timcl has quit IRC | 15:39 | |
| *** liuxinguo has quit IRC | 15:40 | |
| *** Apoorva has joined #openstack-cinder | 15:41 | |
| ub | hi! why is jenkins not merging? https://review.openstack.org/#/c/204702/ Did i forget something? | 15:41 |
| *** jasondotstar has joined #openstack-cinder | 15:42 | |
| thangp | ub: zuul is backed up | 15:42 |
| *** annashen has joined #openstack-cinder | 15:42 | |
| thangp | ub: its going to take sometime for it to process the backlog | 15:42 |
| ub | thangp: thanks. so i all is ok? | 15:43 |
| thangp | ub: probably | 15:43 |
| Swanson | You're only in 3 1/2 hours. There are jobs in that have been in the queue for some 7+ hours. So, good luck. | 15:43 |
| *** hemnafk is now known as hemna | 15:45 | |
| hemna | mriedem1, morning | 15:45 |
| hemna | oh fun :) | 15:46 |
| *** annashen has quit IRC | 15:47 | |
| *** alexpilotti has quit IRC | 15:48 | |
| mriedem1 | yeah :( | 15:48 |
| hemna | mriedem1, I think thats the right way to go though. I had thought of doing that, but wasn't sure about the automated pieces | 15:48 |
| *** annashen has joined #openstack-cinder | 15:50 | |
| e0ne | xyang1: hi. sure. will do it right now | 15:50 |
| *** skraynev has quit IRC | 15:51 | |
| *** skraynev has joined #openstack-cinder | 15:51 | |
| xyang | e0ne: thanks | 15:52 |
| *** ig0r_ has joined #openstack-cinder | 15:53 | |
| xyang | hemna: I can work with you on that | 15:53 |
| hemna | xyang, I'm creating the filters file now. | 15:53 |
| xyang | hemna: cool | 15:53 |
| hemna | so an issue is that there are filters that are needed by os-brick that already exist in nova and cinder filters | 15:54 |
| hemna | such as tee | 15:54 |
| hemna | if the devstack script is simply going to do something like cat os-brick.filters >> /etc/nova/rootwrap.d/computer.filters, then there may be dupes | 15:55 |
| *** DuncanT has quit IRC | 15:55 | |
| *** dalgaaf has quit IRC | 15:55 | |
| *** h00327910__ has quit IRC | 15:55 | |
| *** marcusvrn has quit IRC | 15:55 | |
| hemna | so doesn't it make sense then to make os-brick have it's own rootwrap ? | 15:59 |
| *** amickus has joined #openstack-cinder | 15:59 | |
| hemna | including /etc/os-brick/rootwrap.conf, /etc/os-brick/rootwrap.d/os-brick.filters | 15:59 |
| hemna | dunno | 15:59 |
| e0ne | hemna: imo, because os-brick is a library, it shouldn't have own configs | 15:59 |
| hemna | e0ne, well, the problem is stated well in the bug and I tend to agree with it. | 16:00 |
| hemna | e0ne, https://bugs.launchpad.net/devstack/+bug/1479842 | 16:00 |
| openstack | Launchpad bug 1479842 in OpenStack Compute (nova) "os-brick needs to provide it's own rootwrap filters file" [High,Confirmed] | 16:00 |
| xyang | hemna: what about just copy it there, not cat | 16:00 |
| e0ne | hemna: it should be configured via cinder/nova conf/rootwrap | 16:00 |
| hemna | e0ne, well that's the problem | 16:00 |
| *** haomaiwang has quit IRC | 16:01 | |
| hemna | it is now, and it's a blocker | 16:01 |
| hemna | because any time os-brick changes and needs to change the filters, then we have the problem of touching nova and cinder's filters | 16:01 |
| hemna | and keep that in sync with the os-brick released version | 16:01 |
| e0ne | hemna: after reading the bug description I'm agree about rootwrap.conf | 16:01 |
| hemna | heh | 16:01 |
| *** timcl1 has quit IRC | 16:01 | |
| hemna | so, I think xyang might be right. | 16:01 |
| e0ne | very strong arguments are there:) | 16:01 |
| hemna | just create a self contained os-brick.filters | 16:01 |
| hemna | and dump that in /etc/nova/rootwrap.d, /etc/cinder/rootwrap.d | 16:02 |
| *** haomaiwa_ has joined #openstack-cinder | 16:02 | |
| hemna | so I'm going to create a completely self contained os-brick.filters | 16:02 |
| hemna | mount, tee, etc | 16:02 |
| hemna | which nova has as well | 16:02 |
| hemna | and just hope the dupes don't cause issues | 16:03 |
| hemna | ? | 16:03 |
| *** kjelly has quit IRC | 16:04 | |
| xyang | hemna: seems that should work | 16:05 |
| hemna | mriedem1, thoughts ? | 16:05 |
| hemna | I'll try and get this done shortly | 16:05 |
| *** salv-orl_ has joined #openstack-cinder | 16:06 | |
| mriedem1 | hemna: so we don't want configs in the library, i agree there | 16:07 |
| hemna | yah | 16:07 |
| mriedem1 | i think a totally self-contained os-brick.filters is what i had in mind | 16:07 |
| mriedem1 | we'll have dupes, but meh | 16:07 |
| hemna | ok that's what I'm working on, but we'll have dupes | 16:07 |
| mriedem1 | hemna: this also probably removes the need for nova/cinder to have to pass in a root_helper to os-brick when creating the intiator | 16:07 |
| hemna | the problem with the dupes long term is if nova changes it's definition of mkdir's filter vs. brick for example | 16:08 |
| mriedem1 | i'm not sure how rootwrap will treat that kind of case | 16:08 |
| hemna | so for now I'll just create a completely self contained filter for brick that includes everything it needs | 16:08 |
| mriedem1 | does it puke, pick the first found, other? | 16:08 |
| mriedem1 | we could ask ttx | 16:09 |
| hemna | if that turns into a problem, we could create a nova.filters and cinder.filters | 16:09 |
| mriedem1 | well, compute.filters/volume.filters | 16:09 |
| mriedem1 | hell, | 16:09 |
| mriedem1 | well, nvm | 16:09 |
| mriedem1 | i was going to say, you could just copy those now, but nvm | 16:09 |
| hemna | heh | 16:09 |
| hemna | yah, let me push this up as a WIP and get some feedback | 16:10 |
| mriedem1 | the depends-on with these changes is going to be interesting | 16:10 |
| hemna | we'll have to do something in devstack to copy it into /etc/nova, /etc/cinder | 16:10 |
| mriedem1 | yup | 16:10 |
| hemna | xyang, if you want to work on that part, that'd be great. | 16:10 |
| *** zul has joined #openstack-cinder | 16:10 | |
| hemna | assume I'll create an os-brick/etc/os-brick/rootwrap.d/os-brick.filters file. | 16:10 |
| mriedem1 | we'll also need an upgradeimpact in the os-brick change probably so that people writing deploy/config tools are aware of this in the liberty release notes | 16:11 |
| xyang | hemna: devstack? I'll see what needs to changed there | 16:11 |
| hemna | mriedem1, ok I'll make sure I add that to the commit message | 16:12 |
| mriedem1 | it would actually be really nice if pbr supported dropping files in setup.cfg into multiple locations | 16:12 |
| mriedem1 | actually, couldn't we just do that in the tree structure in os-brick? | 16:12 |
| mriedem1 | os-brick/etc/nova/os-brick.filters | 16:13 |
| mriedem1 | os-brick/etc/cinder/os-brick.filters | 16:13 |
| mriedem1 | they are the same | 16:13 |
| hemna | so, I'm planning on putting the file in os-brick/etc/os-brick/rootwrap.d/os-brick.filters | 16:13 |
| mriedem1 | then in setup.cfg both are listed, and pbr/setuptools should install both in the right place | 16:13 |
| hemna | so, 2 copies of the same file ? | 16:13 |
| *** Lee1092 has joined #openstack-cinder | 16:14 | |
| mriedem1 | yeah, well, the idea is, if setuptools can just install those in the right place when the library is installed, | 16:14 |
| mriedem1 | then there is no devstack change | 16:14 |
| mriedem1 | but... | 16:14 |
| mriedem1 | could be packaging issues, since nova/cinder require os-brick, os-brick would be installed first and create those directories, | 16:14 |
| hemna | yah | 16:14 |
| mriedem1 | i'm not sure if the nova/cinder users own /etc/nova and /etc/cinder in most packaging | 16:15 |
| mriedem1 | i'd have to look | 16:15 |
| mriedem1 | crap, furniture delivery guys are here, afk for awhile | 16:15 |
| openstackgerrit | Erickson Filipe Guedes dos Santos proposed openstack/cinder: Nested Quota Driver: Get Project Hierarchy https://review.openstack.org/206171 | 16:15 |
| hemna | on my vm it looks like /etc/nova is owned by the user that ran stack.sh, not root | 16:16 |
| *** zul has quit IRC | 16:16 | |
| *** leeantho has joined #openstack-cinder | 16:17 | |
| openstackgerrit | Scott DAngelo proposed openstack/cinder: sqlalchemy exception kills FixedIntervalLoopingCall thread https://review.openstack.org/207529 | 16:18 |
| mriedem1 | hemna: yeah, me too | 16:18 |
| mriedem1 | so....sounds crazy but maybe that would work? | 16:18 |
| *** logan2 has joined #openstack-cinder | 16:19 | |
| *** vokt has joined #openstack-cinder | 16:19 | |
| *** dalgaaf has joined #openstack-cinder | 16:21 | |
| *** timcl has joined #openstack-cinder | 16:21 | |
| mriedem1 | looking at some rpms that we create for nova, etc/nova/ contents are owned by the nova user rather than the user installing the package, but that's rhel based | 16:22 |
| mriedem1 | i'm not sure what rdo does | 16:22 |
| *** lpetrut has quit IRC | 16:22 | |
| *** jasondotstar has quit IRC | 16:23 | |
| *** _cjones_ has joined #openstack-cinder | 16:23 | |
| mriedem1 | rdo does the same http://pkgs.fedoraproject.org/cgit/openstack-nova.git/tree/openstack-nova.spec#n692 | 16:23 |
| mriedem1 | that package also adds the nova user http://pkgs.fedoraproject.org/cgit/openstack-nova.git/tree/openstack-nova.spec#n601 | 16:24 |
| *** shausy has quit IRC | 16:24 | |
| *** jistr has quit IRC | 16:24 | |
| mriedem1 | idk, i guess just letting os-brick lay it's own file down and leave it up to deployment/config tools sort it out works too - if you're not using libvirt you're not using these files anyway | 16:25 |
| *** marcusvrn has joined #openstack-cinder | 16:26 | |
| *** zul has joined #openstack-cinder | 16:28 | |
| mriedem1 | zigo: are you around? | 16:28 |
| *** apevec has joined #openstack-cinder | 16:28 | |
| apevec | mriedem1, here? | 16:28 |
| mriedem1 | apevec: yeah, we're talking about https://bugs.launchpad.net/devstack/+bug/1479842 | 16:28 |
| openstack | Launchpad bug 1479842 in os-brick "os-brick needs to provide it's own rootwrap filters file" [High,Confirmed] - Assigned to Walt Boring (walter-boring) | 16:28 |
| apevec | what's the packaging question? | 16:28 |
| mriedem1 | apevec: nova and cinder are now using os-brick (library), | 16:29 |
| apevec | right | 16:29 |
| mriedem1 | and right now, nova and cinder pass in their root helpers to os-brick to do rootwrap things | 16:29 |
| mriedem1 | which means the commands run as root in os-brick have to have rootwrap filters in nova and cinder, which is not good | 16:29 |
| *** e0ne has quit IRC | 16:29 | |
| mriedem1 | so we want to move the filters that os-brick needs into os-brick so nova/cinder don't need to be synced up with what os-brick needs, | 16:29 |
| mriedem1 | the question is, how to drop the os-brick filters into /etc/nova/ and /etc/cinder | 16:30 |
| mriedem1 | apevec: b/c i think you can specify that structure in os-brick and setup.cfg and setuptools will put them in /etc/nova and /etc/cinder at install time, | 16:30 |
| mriedem1 | but the os-brick package will be installed before nova/cinder packages, and it seems weird for the os-brick package to create the /etc/nova and /etc/cinder directories, | 16:31 |
| mriedem1 | especially since in rdo, the contents of /etc/nova are owned by the nova user, which is created in the openstack-nova package | 16:31 |
| *** aix has quit IRC | 16:31 | |
| apevec | right, in rpm it would required osbrick to depend on nova/cinder to be able to drop into their folders | 16:31 |
| mriedem1 | apevec: so i'm wondering if there is some packaging black magic that can be done, or do we just punt and let deploy/config tools sort it out | 16:31 |
| hemna | that creates a circular dep though | 16:32 |
| mriedem1 | yeah | 16:32 |
| apevec | yeah | 16:32 |
| mriedem1 | i guess the nova/cinder packages could just have a %post macro that copies the os-brick filters into /etc/nova/rootwrap.d | 16:32 |
| apevec | and looking at nova spec, it moves rootwrap files manually in %install | 16:32 |
| apevec | mriedem1, ugh | 16:33 |
| *** cppforlife_ has joined #openstack-cinder | 16:33 | |
| mriedem1 | as long as the nova package requires the os-brick package and knows it's a version with those files, it could do the copy (or just dev/null any failures if the files don't exist) | 16:33 |
| mriedem1 | apevec: yeah, it's hacky | 16:33 |
| *** dims has quit IRC | 16:33 | |
| *** zul has quit IRC | 16:33 | |
| openstackgerrit | Cindy Pallares proposed openstack/cinder: Fix multi-line docstrings to meet hacking rules https://review.openstack.org/184820 | 16:34 |
| *** jordanP has quit IRC | 16:36 | |
| apevec | mriedem1, discussing this in #rdo, symlinks might be better than copying | 16:38 |
| apevec | but it's tricky one, I'll think of it more | 16:38 |
| sdague | mriedem1: this makes it impossible to user from pip though, right? given that pip doesn't understand etc files | 16:39 |
| *** salv-orl_ has quit IRC | 16:39 | |
| jgriffith | Not to interrupt... but isn't this sort of library design problem? | 16:40 |
| mriedem1 | sdague: not sure, i thought that if setup.cfg had files pointing to etc/ that setuptools would lay those down on the file system.....but i guess it would probably throw them under /usr/lib/python...... | 16:40 |
| jgriffith | In other words, shouldn't the lib be unaware of policy like the rest of our libs? | 16:40 |
| jgriffith | Leave it to nova and cinder and .... specific policy usage and details completely independent of the library? | 16:41 |
| hemna | jgriffith, that's hinting at one of the alternatives I mentioned earlier, giving the library it's own rootwrap | 16:41 |
| mriedem1 | are there any other libraries that are running commands as root though? | 16:41 |
| hemna | which it would then use to execute commands | 16:41 |
| jgriffith | hemna: NO, I'm saying that's wrong | 16:42 |
| hemna | I'm not sure this has come up before though in other libs | 16:42 |
| jgriffith | hemna: I'm saying jus the opposite | 16:42 |
| sdague | jgriffith: doesn't that mean really tight coupling between cinder / nova / and osbrick then? | 16:42 |
| sdague | which is what this was trying to avoid | 16:42 |
| hemna | I think 'the rest of our libs' don't execute commands as root though. | 16:42 |
| *** zul has joined #openstack-cinder | 16:42 | |
| sdague | hemna: I think that's probably true | 16:43 |
| jgriffith | sdague: possibly. but I don't think that's a requirement | 16:43 |
| jgriffith | honestly I have to look at what's at issue here in the root-wrap to understand why this is an issue | 16:43 |
| *** annashen_ has joined #openstack-cinder | 16:43 | |
| *** DuncanT has joined #openstack-cinder | 16:43 | |
| hemna | jgriffith, so the problem is when new connectors are added to brick, they may require executing new commands. | 16:43 |
| jgriffith | sdague: We do things with the process module for example and use Nova or Cinder root wrap for that | 16:43 |
| jgriffith | this should be designed similarly IMHO | 16:44 |
| jgriffith | and sorry... I think I said "policy" earlier but meant rootwrap | 16:44 |
| hemna | which would then have to be added to nova's computer.filters and cinder's filters and tie that with the specific release of brick that contains that new connector | 16:44 |
| hemna | I think that's what we are trying to avoid | 16:44 |
| jgriffith | sdague: mriedem1 is that correct.... you don't want any of this info in the nova rootwrap for some reason? | 16:45 |
| *** alexpilotti has joined #openstack-cinder | 16:46 | |
| jgriffith | hemna: do you have an example of the entries? | 16:46 |
| jgriffith | NM... found the review | 16:46 |
| hemna | jgriffith, https://review.openstack.org/#/c/194454/16/etc/nova/rootwrap.d/compute.filters | 16:46 |
| mriedem1 | jgriffith: well we wanted to avoid the tight coupling | 16:46 |
| jgriffith | hemna: yeah.. I found it | 16:46 |
| mriedem1 | and duplication between nova and cinder filters to use os-brick | 16:46 |
| jgriffith | and IMHO that' s BS | 16:47 |
| jgriffith | The patch.. not the objections to it | 16:47 |
| *** annashen_ has quit IRC | 16:47 | |
| mriedem1 | the precedent is the multipathd entry to compute.filters in the os-brick integration change https://review.openstack.org/#/c/175569/32/etc/nova/rootwrap.d/compute.filters | 16:48 |
| *** ronis has joined #openstack-cinder | 16:48 | |
| mriedem1 | which has landed | 16:48 |
| jgriffith | mriedem1: yeah, reading that now | 16:48 |
| jgriffith | mriedem1: hemna so here's my take; if you're configuring a 3'rd party device like ScaleIO, then update the root wrap your self and install/config | 16:48 |
| hemna | well that's only one aspect | 16:49 |
| jgriffith | mriedem1: hemna leave that patch out of the mix; document the steps as part of install requirements | 16:49 |
| jgriffith | hemna: yes | 16:49 |
| jgriffith | hemna: it is | 16:49 |
| jgriffith | hemna: let me finish please | 16:49 |
| hemna | there are some changes that may happen that aren't 3rd party | 16:49 |
| jgriffith | mriedem1: hemna so rather then continue to compound the issue based on precedence, recognize it's probably not the right direction going forward. | 16:49 |
| jgriffith | mriedem1: hemna and note that we're going to change it | 16:50 |
| hemna | I think that's exactly what we are trying to do :) | 16:50 |
| jgriffith | mriedem1: hemna if somebody wants to crank out a patch to the include brick-filters today... so be it | 16:50 |
| *** lpetrut has joined #openstack-cinder | 16:50 | |
| mriedem1 | yeah, so my thinking was we landed the os-brick change and don't want to revert it, | 16:50 |
| mriedem1 | and we had the scaleio and hgst changes lined up which add 2 new entries to compute.filters, | 16:51 |
| mriedem1 | and today is nova feature freeze | 16:51 |
| jgriffith | mriedem1: +1 | 16:51 |
| mriedem1 | so we thought, we're already busted, this goes from 1->3 weird filters, let them in and fix this as a bug before liberty release | 16:51 |
| jgriffith | mriedem1: I'm also wondering if there's another option to this besides a brick specific rootwrap | 16:51 |
| jgriffith | mriedem1: yeah... there's some logic there that makes sense I suppose. I kinda wish I had noticed this before :( | 16:52 |
| openstackgerrit | Daniel Tadrzak proposed openstack/cinder: Cleanup for cinder tests with ConsistencyGroups. https://review.openstack.org/207544 | 16:52 |
| jgriffith | mriedem1: I wonder if there's a way to do a generic "brick" filter in each projects rootwrap | 16:52 |
| mriedem1 | jgriffith: well that would organize it better but doesn't decouple it | 16:53 |
| hemna | so I still think having an os-brick.filters is a way to decouple things and drop that file into nova and cinder. | 16:53 |
| jgriffith | mriedem1: it doesn't? | 16:53 |
| *** vokt has quit IRC | 16:53 | |
| hemna | brick knows what it needs to run, nova shouldn't have to keep track of that. | 16:53 |
| jgriffith | mriedem1: I'm tinking like an "enable_brick_methods" | 16:53 |
| jgriffith | mriedem1: yeah... true. | 16:53 |
| jgriffith | mriedem1: hemna yeah... it may in fact be the best answer :( | 16:54 |
| mriedem1 | hemna: well, drop it in on the part of deploy/config tools right? | 16:54 |
| jgriffith | hemna: mriedem1 what the heck... what's one more thing to make deployment more difficult | 16:54 |
| mriedem1 | puppet/chef/ansible/devstack | 16:54 |
| hemna | mriedem1, yah | 16:54 |
| jgriffith | We've managed to create an entire industry around that at this point | 16:54 |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Tooz locks https://review.openstack.org/183537 | 16:54 |
| *** jordanP has joined #openstack-cinder | 16:55 | |
| mriedem1 | yeah so i'm all for os-brick owning it's own rootwrap filter in it's own tree | 16:55 |
| jgriffith | docs.openstack.org install | 16:55 |
| mriedem1 | sounds like dhellmann has some ideas too | 16:55 |
| jgriffith | mriedem1: cool | 16:56 |
| mriedem1 | however, at this point, it's time for me to get lunch | 16:56 |
| dhellmann | we have oslo.privsep in development to replace rootwrap, I wonder if we can live with a little pain until gus finishes that and we can migrate over. https://launchpad.net/oslo.privsep | 16:56 |
| jgriffith | mriedem1: hemna so it does sound like you guys have the best short term answer for sure. But I'm wondering long-term | 16:56 |
| openstackgerrit | Szymon Wróblewski proposed openstack/cinder: Tooz locks https://review.openstack.org/183537 | 16:56 |
| hemna | jgriffith, +1 | 16:57 |
| jgriffith | dhellmann: oh.. that sounds interesting | 16:57 |
| *** dims has joined #openstack-cinder | 16:57 | |
| dhellmann | jgriffith: I don't want to promise too much, since it's early, but we should track gus down and get his input | 16:57 |
| *** ub has quit IRC | 16:57 | |
| dhellmann | I can't seem to find the spec | 16:57 |
| jgriffith | dhellmann: he's got a patch in neutron :) | 16:57 |
| dhellmann | https://blueprints.launchpad.net/oslo-incubator/+spec/privsep | 16:58 |
| hemna | dhellmann, would be nice that privsep could help solve this a little bit long term. | 16:58 |
| *** belmoreira has quit IRC | 16:58 | |
| jgriffith | https://review.openstack.org/#/c/155631/14/etc/neutron.conf | 16:58 |
| jgriffith | interesting | 16:58 |
| dhellmann | https://etherpad.openstack.org/p/YVR-oslo-rootwrap-plans | 16:58 |
| *** lpetrut has quit IRC | 16:59 | |
| *** timcl1 has joined #openstack-cinder | 16:59 | |
| dhellmann | jgriffith: I think that was POC for the code going into the library | 16:59 |
| jgriffith | haha... just run brick as Root :) | 16:59 |
| dhellmann | jgriffith: dims would know better about the current status | 16:59 |
| hemna | heh | 17:00 |
| jgriffith | dhellmann: well, I think it's def worth holding out for IMHO | 17:00 |
| dhellmann | it might make more sense than looking for ways to ship config files with the library | 17:00 |
| jgriffith | or waiting till we have more info on it before doing anything drastic :) | 17:00 |
| dhellmann | otoh, it might not solve this problem directly, so please do talk to dims & gus to make sure it's taken into account | 17:00 |
| *** haomaiwa_ has quit IRC | 17:01 | |
| jgriffith | dhellmann: indeed, I'm not certain why that idea bothers me so much | 17:01 |
| jgriffith | dhellmann: understood... it may not be a fit; but warrants discussion/investigation for sure | 17:01 |
| *** timcl has quit IRC | 17:02 | |
| *** haomaiwang has joined #openstack-cinder | 17:02 | |
| jgriffith | hemna: You want to follow up with dims and gus at some point? | 17:02 |
| jgriffith | or mriedem1 | 17:02 |
| jgriffith | or any lurkers about :) | 17:02 |
| hemna | jgriffith, yah | 17:03 |
| *** avishay has quit IRC | 17:03 | |
| *** cebruns_ has joined #openstack-cinder | 17:08 | |
| openstackgerrit | Daniel Tadrzak proposed openstack/cinder: ConsistencyGroup Object https://review.openstack.org/163641 | 17:09 |
| *** h00327910__ has joined #openstack-cinder | 17:09 | |
| *** liewegas has quit IRC | 17:11 | |
| openstackgerrit | Erlon R. Cruz proposed openstack/cinder: Fix HNAS iSCSI 32 targets limitation error https://review.openstack.org/206651 | 17:12 |
| *** bluex-pl has quit IRC | 17:14 | |
| openstackgerrit | Daniel Tadrzak proposed openstack/cinder: CGSnapshot Object https://review.openstack.org/195027 | 17:16 |
| *** vilobhmm has joined #openstack-cinder | 17:16 | |
| *** lpetrut has joined #openstack-cinder | 17:18 | |
| *** liewegas has joined #openstack-cinder | 17:19 | |
| *** ub has joined #openstack-cinder | 17:20 | |
| DTadrzak | Could someone forward me link to priority list :)? | 17:21 |
| *** e0ne has joined #openstack-cinder | 17:22 | |
| *** e0ne has quit IRC | 17:23 | |
| *** e0ne has joined #openstack-cinder | 17:23 | |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/os-brick: Add rootwrap filters https://review.openstack.org/207553 | 17:24 |
| *** bapalm_ has quit IRC | 17:25 | |
| hemna | bah, trailing whitespace. pita. | 17:25 |
| *** dims has quit IRC | 17:26 | |
| openstackgerrit | Erlon R. Cruz proposed openstack/cinder: Fix HNAS iSCSI 32 targets limitation error https://review.openstack.org/206651 | 17:26 |
| *** IlyaG has quit IRC | 17:26 | |
| *** annashen has quit IRC | 17:28 | |
| *** IlyaG has joined #openstack-cinder | 17:29 | |
| *** lpetrut has quit IRC | 17:30 | |
| openstackgerrit | Duncan Thomas proposed openstack/cinder: Add ability to get endpoints from catalog https://review.openstack.org/207555 | 17:31 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/os-brick: Add rootwrap filters https://review.openstack.org/207553 | 17:31 |
| *** IlyaG has quit IRC | 17:33 | |
| *** erlon has joined #openstack-cinder | 17:35 | |
| *** pv_ has joined #openstack-cinder | 17:36 | |
| jgriffith | hemna: put this in your .vimrc | 17:36 |
| jgriffith | autocmd BufWritePre * :%s/\s\+$//e | 17:36 |
| *** boris-42 has joined #openstack-cinder | 17:37 | |
| pv_ | i had a quick question on the configuration attribute in the volumedriver class, if the driver code i'm writing inherits from the FibreChannelDriver, and that in turn inherits from VolumeDriver which says it inherits from BaseVD where the line | 17:39 |
| pv_ | self.configuration = kwargs.get('configuration', None) occurs | 17:39 |
| pv_ | my devstack stack.sh isn't completing because the self.configuration attribute is a Nonetype, which is the default value for it | 17:39 |
| pv_ | but how can I check why kwargs.get(configuration) isn't setting the config variable to local.conf or a similar conf file? | 17:40 |
| jgriffith | pv_: Your local.conf file? | 17:41 |
| jgriffith | pv_: as in can we see it :) | 17:41 |
| * jgriffith should've saved your links last week :) | 17:41 | |
| pv_ | yup gimme a sec | 17:42 |
| *** ub has quit IRC | 17:42 | |
| *** daneyon has joined #openstack-cinder | 17:43 | |
| *** jungleboyj has quit IRC | 17:43 | |
| *** annashen has joined #openstack-cinder | 17:44 | |
| *** daneyon_ has joined #openstack-cinder | 17:44 | |
| pv_ | http://pastebin.com/30DXeD4D | 17:45 |
| *** ub has joined #openstack-cinder | 17:46 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Use prefix for SolidFire template account https://review.openstack.org/207561 | 17:47 |
| *** daneyon has quit IRC | 17:47 | |
| *** jungleboyj has joined #openstack-cinder | 17:48 | |
| *** xyang1 has left #openstack-cinder | 17:48 | |
| jgriffith | pv_: that should work... and without seeing your driver hard to know if you're missing something in the init | 17:48 |
| jgriffith | pv_: since you have to keep your stuff "secret" afraid I'm not going to be able to help much | 17:48 |
| hemna | jgriffith, thanks :) | 17:49 |
| *** annashen has quit IRC | 17:49 | |
| pv_ | hm alright, i know that the traceback is showing me the error is with this line self.configuration.append_config_values(volume_opts) in my __init__ method for my driver | 17:49 |
| *** sgotliv has quit IRC | 17:50 | |
| *** xyang1 has joined #openstack-cinder | 17:50 | |
| pv_ | line 117, in __init__ 2015-07-30 10:27:56.981 1329 ERROR cinder self.configuration.append_config_values(volume_opts) 2015-07-30 10:27:56.981 1329 ERROR cinder AttributeError: 'NoneType' object has no attribute 'append_config_values' 2015-07-30 10:27:56.981 1329 ERROR cinder c-vol failed to start | 17:50 |
| jgriffith | pv_: double check this: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L68 | 17:50 |
| *** apevec has quit IRC | 17:50 | |
| jgriffith | pv_: and this: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L73 | 17:50 |
| jgriffith | pv_: also make surrey have the opts setup properly in your cfg section | 17:51 |
| jgriffith | pv_: DONT use volume_opts | 17:51 |
| jgriffith | pv_: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L44 | 17:51 |
| xyang1 | hemna: if we just add "configure_rootwrap os-brick" in devstack/lib/nova, does that work | 17:51 |
| jgriffith | pv_: change that... s/volume_opts/ourdriver_opts/ | 17:52 |
| jgriffith | pv_: in your driver | 17:52 |
| hemna | xyang1, I don't know. | 17:52 |
| xyang1 | hemna: I'm not familiar with devstack:( | 17:53 |
| pv_ | alright i changed it from volume_opts to ourdriver_opts | 17:53 |
| pv_ | thats what you mean right? | 17:53 |
| pv_ | and i looked at the LVM driver's init method, im doing the same thing | 17:53 |
| hemna | xyang1, so, that function seems to do lots of stuffs | 17:54 |
| *** Tross1 has quit IRC | 17:54 | |
| *** Tross has joined #openstack-cinder | 17:54 | |
| hemna | xyang1, and I don't think it's what we want. all we want to do is copy the file into the project's rootwrap.d folder. | 17:54 |
| jgriffith | pv_: yeah | 17:54 |
| xyang1 | hemna: /etc/os-brick/rootwrap.d/os-brick.filters ? | 17:55 |
| hemna | xyang1, no | 17:55 |
| *** annashen has joined #openstack-cinder | 17:56 | |
| *** timcl has joined #openstack-cinder | 17:56 | |
| xyang1 | hemna: /etc/nova/rootwrap.d/os-brick.filters? | 17:56 |
| *** timcl1 has quit IRC | 17:56 | |
| *** alexpilotti has quit IRC | 17:56 | |
| hemna | yah | 17:57 |
| hemna | and cinder as well | 17:57 |
| xyang1 | hemna: then we can't use that configure_rootwrap project directly . I think that copy it directly under the "project" | 17:58 |
| hemna | yup | 17:58 |
| *** angela-s has joined #openstack-cinder | 17:58 | |
| *** coolsvap|away is now known as coolsvap | 17:59 | |
| *** ronis has quit IRC | 18:00 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-cinder | 18:02 | |
| *** ub has quit IRC | 18:02 | |
| *** rlucio has quit IRC | 18:03 | |
| *** salv-orlando has joined #openstack-cinder | 18:05 | |
| *** ub has joined #openstack-cinder | 18:05 | |
| smcginnis | e0ne: Sorry, just saw your ping. You still around? | 18:15 |
| jungleboyj | *insert snorting sound here* | 18:16 |
| jungleboyj | *snoring* | 18:16 |
| thingee | scottda: yeah that sounds good to sign up for nova liaison. thanks hemna | 18:16 |
| smcginnis | jungleboyj: Wondered why you were snorting. :) | 18:16 |
| e0ne | smcginnis: np. i just wanted to confirm that i understood your comment for https://review.openstack.org/#/ right | 18:17 |
| thingee | jgriffith: reminder on replication v2 spec https://review.openstack.org/#/c/155644/ if we can get an update on that for approval | 18:18 |
| smcginnis | e0ne: Looking | 18:18 |
| e0ne | smcginnis: thanks | 18:18 |
| smcginnis | e0ne: Oh, wait. Missed something there. | 18:18 |
| smcginnis | e0ne: # | 18:18 |
| e0ne | smcginnis: what do you mean? | 18:19 |
| smcginnis | e0ne: The link you pasted isn't complete. Which review number are you referring to? | 18:19 |
| thingee | scottda: no problem on the thermos. I have the coffee and gear in my bag in case. | 18:20 |
| e0ne | smcginnis: oh, i'm sorry. here it is: https://review.openstack.org/#/c/146541/ | 18:20 |
| smcginnis | e0ne: Thanks. OK, now I'm looking. :) | 18:20 |
| thingee | mriedem1: looking at release blocker | 18:20 |
| e0ne | :) | 18:20 |
| mriedem1 | thingee: we kind of beat that horse to death about an hour ago | 18:21 |
| mriedem1 | thingee: feel free to add thoughts though | 18:21 |
| * thingee catches up on backlog | 18:21 | |
| smcginnis | e0ne: OK, I see. So you're saying it's OK that those tests still validate v1 is being called? | 18:21 |
| thingee | looks like a fix is in os-brick and we're waiting on some change change with devstack | 18:21 |
| hemna | thingee, mriedem1 xyang1 is working on a devstack patch to plop the os-brick.filters file into rootwrap.d dirs for cinder and nova. | 18:22 |
| e0ne | smcginnis: tbh, i didn't check coverage | 18:22 |
| thingee | hemna: gotcha | 18:22 |
| e0ne | smcginnis: but we've got tests in v1 directory | 18:22 |
| thingee | mriedem1, hemna : so we're ok? | 18:22 |
| smcginnis | e0ne: I'm trying to find the specific lines I was concerned about. | 18:22 |
| hemna | thingee, we have a plan....we just need to test it out, once the devstack piece is in place | 18:22 |
| thingee | mriedem1, hemna, xyang1: whew thanks everyone. | 18:22 |
| * thingee goes back to specs | 18:22 | |
| hemna | thingee, it's my top priority right now. | 18:23 |
| *** ub has quit IRC | 18:23 | |
| smcginnis | e0ne: Well, I can't find it. Not sure what I was looking at for that one. | 18:24 |
| smcginnis | e0ne: I don't see an issue with any of your changes. I'll wait for jenkins and make one more pass. | 18:24 |
| smcginnis | e0ne: In the mean time I'll keep looking and if I find anything elsewhere of concern I'll make sure to be more explicit. | 18:24 |
| e0ne | smcginnis: thank you! | 18:25 |
| *** ig0r__ has joined #openstack-cinder | 18:25 | |
| smcginnis | e0ne: NP. Sorry for the confusion. | 18:25 |
| *** ndipanov has quit IRC | 18:25 | |
| *** ndipanov has joined #openstack-cinder | 18:26 | |
| *** patrickeast has joined #openstack-cinder | 18:26 | |
| e0ne | smcginnis: don't worry. imo, it's better to -1 and ask question than merge something broken | 18:26 |
| *** dims has joined #openstack-cinder | 18:26 | |
| *** ronis has joined #openstack-cinder | 18:27 | |
| *** jasondotstar has joined #openstack-cinder | 18:30 | |
| *** wanghao_ has joined #openstack-cinder | 18:30 | |
| vilobhmm | jgriffith : ping | 18:31 |
| *** briancurtin_ has joined #openstack-cinder | 18:32 | |
| *** sweston_ has joined #openstack-cinder | 18:32 | |
| *** thingee_ has joined #openstack-cinder | 18:32 | |
| *** coolsvap is now known as coolsvap|away | 18:32 | |
| thingee_ | hemna: thank you :) | 18:32 |
| *** bill_az_ has joined #openstack-cinder | 18:33 | |
| *** amickus has quit IRC | 18:33 | |
| *** marcusvrn has quit IRC | 18:33 | |
| *** briancurtin has quit IRC | 18:34 | |
| *** dalgaaf has quit IRC | 18:34 | |
| *** wanghao has quit IRC | 18:34 | |
| *** thingee has quit IRC | 18:34 | |
| *** sweston has quit IRC | 18:34 | |
| *** rushiagr has quit IRC | 18:34 | |
| *** xyang1 has quit IRC | 18:34 | |
| *** ericksonsantos has quit IRC | 18:34 | |
| *** kbyrne has quit IRC | 18:34 | |
| *** bill_az has quit IRC | 18:34 | |
| *** ericksonfgds has joined #openstack-cinder | 18:34 | |
| *** briancurtin_ is now known as briancurtin | 18:34 | |
| *** wanghao_ is now known as wanghao | 18:34 | |
| *** thingee_ is now known as thingee | 18:35 | |
| *** kbyrne has joined #openstack-cinder | 18:35 | |
| geguileo | thingee: Are you in the mood for some specs review pestering? | 18:35 |
| geguileo | :-) | 18:35 |
| *** sweston_ is now known as sweston | 18:35 | |
| vilobhmm | smcginnis : ping | 18:36 |
| vilobhmm | any specific reason you can think of why the root parameter is added in this method https://github.com/openstack/cinder/blob/master/cinder/tests/unit/api/contrib/test_quotas.py#L31 | 18:37 |
| *** salv-orlando has quit IRC | 18:40 | |
| smcginnis | vilobhmm: Not sure. Looks like it is only overridden in the last test. | 18:42 |
| *** rushiagr has joined #openstack-cinder | 18:42 | |
| openstackgerrit | Jordan Pittier proposed openstack/cinder: Rework Scality SOFS driver to use RemoteFS class https://review.openstack.org/200610 | 18:42 |
| *** dalgaaf has joined #openstack-cinder | 18:42 | |
| smcginnis | vilobhmm: So it's just setting the quota_set. I think that's normal. Not a quota expert by any means though. | 18:43 |
| *** changbl has quit IRC | 18:43 | |
| jgriffith | vilobhmm: pong | 18:44 |
| Swanson | hemna: did your create_export patch land? | 18:45 |
| hemna | not yet | 18:45 |
| hemna | the spec did | 18:45 |
| Swanson | thanks | 18:45 |
| openstackgerrit | Jon Bernard proposed openstack/cinder: Add support for file I/O volume migration https://review.openstack.org/187270 | 18:47 |
| *** ig0r_ has quit IRC | 18:50 | |
| *** xyang1 has joined #openstack-cinder | 18:52 | |
| *** changbl has joined #openstack-cinder | 18:56 | |
| *** bapalm_ has joined #openstack-cinder | 19:00 | |
| *** haomaiwang has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-cinder | 19:02 | |
| *** ig0r_ has joined #openstack-cinder | 19:04 | |
| *** bapalm_ has quit IRC | 19:05 | |
| *** annashen has quit IRC | 19:08 | |
| *** annashen has joined #openstack-cinder | 19:09 | |
| *** lpabon has quit IRC | 19:10 | |
| *** annashen has quit IRC | 19:10 | |
| *** annashen has joined #openstack-cinder | 19:10 | |
| *** kvidvans has quit IRC | 19:11 | |
| *** uberjay has quit IRC | 19:11 | |
| *** uberjay has joined #openstack-cinder | 19:12 | |
| *** marcusvrn has joined #openstack-cinder | 19:14 | |
| thingee | hemna, mriedem1 looks like https://review.openstack.org/#/c/194454 is already resolved. Sorry just now getting back to it due to my backlog | 19:15 |
| hemna | yah, so at this point we are waiting for the devstack patch that will use my WIP for brick patch, to install the filters. | 19:16 |
| hemna | https://review.openstack.org/#/c/207553/ | 19:16 |
| thingee | geguileo: not sure if you saw my earlier message, but I'm already looking at specs. Not sure what good it's going to do. | 19:16 |
| thingee | hemna: got it, it's related to that | 19:16 |
| geguileo | thingee: Oh, I didn't see it, thanks for pointing that out | 19:16 |
| thingee | does anyone besides tbarron care about scalable backup service spec? https://review.openstack.org/#/c/203215/4 | 19:17 |
| geguileo | thingee: I care, but I haven't had time to review it | 19:18 |
| thingee | winston-d_zZZ agrees it's a good approach according to the WIP patch https://review.openstack.org/#/c/203291/ | 19:18 |
| thingee | just haven't see any movement on the spec | 19:19 |
| *** ronis has quit IRC | 19:20 | |
| tbarron | and I think when winston said that he was awake :) | 19:20 |
| thingee | aarefiev: rootwrap daemon spec is approved | 19:23 |
| hemna | xyang, heading to lunch...bbiab | 19:25 |
| *** ub has joined #openstack-cinder | 19:25 | |
| *** e0ne has quit IRC | 19:25 | |
| *** annashen has quit IRC | 19:26 | |
| *** annashen has joined #openstack-cinder | 19:26 | |
| *** julim has quit IRC | 19:27 | |
| *** lpetrut has joined #openstack-cinder | 19:28 | |
| *** jasondotstar has quit IRC | 19:29 | |
| *** salv-orlando has joined #openstack-cinder | 19:29 | |
| *** annashen_ has joined #openstack-cinder | 19:29 | |
| *** Apoorva has quit IRC | 19:30 | |
| *** kvidvans has joined #openstack-cinder | 19:30 | |
| *** lpetrut has quit IRC | 19:30 | |
| thingee | geguileo: did you give some of the other solutions thoughts that were discussed with jgriffith https://review.openstack.org/#/c/202615/6 | 19:31 |
| thingee | and I | 19:31 |
| geguileo | thingee: The thing is I think we'll need locking for any initial approach | 19:32 |
| geguileo | thingee: And some drivers will need locks anyway | 19:32 |
| thingee | geguileo: you must've not understood what I was explaining yesterday. | 19:32 |
| geguileo | thingee: Even when we finally remove all locks from the manager | 19:32 |
| geguileo | thingee: Oh, the UUID broadcasting | 19:32 |
| *** timcl has quit IRC | 19:32 | |
| thingee | If I can avoid the deployer having to learn about zookeeper, etcd, whatever, that's a huge huge win | 19:33 |
| geguileo | thingee: I thought a little bit (I must admit not as much as I should have) | 19:33 |
| tbarron | vs storing the uuid owner for each operation | 19:33 |
| thingee | tbarron: that's already too complex from what I was explaining | 19:34 |
| thingee | tbarron: storing the operation is not needed. we've already agreed we don't need allow parallel tasks for a resource. | 19:34 |
| *** annashen_ has quit IRC | 19:34 | |
| geguileo | thingee: I think your solution will add more complexity and after all we are going to remove most locks eventually | 19:34 |
| thingee | how so? | 19:34 |
| thingee | unlike every solution explained here, mine can be explained in one sentence. | 19:35 |
| thingee | not a blog post or spec. | 19:35 |
| geguileo | thingee: You need to distribute the UUIDs and keep track of the timeouts to remove them if we haven't received updates in a while | 19:35 |
| geguileo | thingee: And maybe we aren't receiving notifications because we have lost connection to the broker | 19:36 |
| thingee | uuids are broadcast to each c-api, c-vol, do a time out if needed. | 19:36 |
| tbarron | how to distinguish between broken link and node down ... | 19:36 |
| thingee | geguileo: if you lose connection you have more problems than that | 19:36 |
| thingee | that argument is a dead horse at this point | 19:36 |
| geguileo | thingee: The effort I believe is smaller for locks | 19:37 |
| geguileo | thingee: And we are going to remove it later, so... | 19:38 |
| thingee | what is "it" in this context? | 19:38 |
| geguileo | thingee: And we would need to provide some kind of methods for the drivers as well | 19:38 |
| geguileo | thingee: The locking (using locks or the UUID broadcasting) | 19:38 |
| geguileo | thingee: except in some drivers | 19:38 |
| thingee | geguileo: if you're relying on drivers to enforce things, that's the wrong approach | 19:38 |
| thingee | geguileo: the manager is enforcing | 19:38 |
| geguileo | thingee: No, what I mean is that some drivers | 19:39 |
| geguileo | thingee: need locks | 19:39 |
| thingee | geguileo: removing the broadcast approach was never something I said | 19:39 |
| geguileo | thingee: For their own purposes | 19:39 |
| geguileo | thingee: Not for resource locking | 19:39 |
| geguileo | thingee: Like you said, that will be taken care by Cinder (API and Manager) | 19:39 |
| geguileo | thingee: But some drivers still need locking mechanism for other stuff | 19:39 |
| geguileo | thingee: I think glusterFS was one such drivers | 19:40 |
| thingee | geguileo: they can keep their local locks. Only one c-vol can work with a uuid anyways. | 19:40 |
| thingee | geguileo: with the broadcast approach | 19:40 |
| *** annashen has quit IRC | 19:40 | |
| *** annashen has joined #openstack-cinder | 19:41 | |
| geguileo | thingee: But sometimes it's not a matter of the UUID of the resource they need to lock on | 19:41 |
| thingee | geguileo: can you give me an example? I've been noticing <method>-<resource-uuid> | 19:41 |
| geguileo | thingee: And that's the problem, because the UUID exclusion will be handled by Cinder | 19:41 |
| geguileo | thingee: When you say <metho>-<resource-uuid> you are talking about the locks? | 19:42 |
| thingee | geguileo: why is that a problem if it's a matter of keeping track of uuids in a global pool? | 19:42 |
| thingee | geguileo: yes, I want an example of what you're talking about where drivers it's not a matter of the uuid | 19:42 |
| geguileo | thingee: Locks are weird right now, and they would be changed to remove the <method> part | 19:42 |
| thingee | geguileo: I want an example | 19:42 |
| geguileo | thingee: They would be <resource-type>-<uuid> | 19:42 |
| thingee | geguileo> thingee: But sometimes it's not a matter of the UUID of the resource they need to lock on | 19:43 |
| geguileo | thingee: Dulek was the one that saw that | 19:43 |
| geguileo | thingee: But he's on vacation until next week | 19:43 |
| geguileo | thingee: I'll need a couple of minutes to dig into that | 19:43 |
| thingee | geguileo: mute point then | 19:43 |
| *** lpetrut has joined #openstack-cinder | 19:43 | |
| thingee | geguileo: point is, I'm not completely sold on putting the deployer through what developers are excited about. | 19:43 |
| geguileo | thingee: Let me get back to you in a couple of minutes | 19:43 |
| thingee | geguileo: I appreciate your work on this, but I just feel like we're assuming DLM and tooz is just going to solve all of our problems. | 19:45 |
| geguileo | thingee: Nooooooo, they are not | 19:45 |
| geguileo | thingee: I wish!!! | 19:45 |
| thingee | :) | 19:45 |
| geguileo | thingee: That's just a little piece of the puzzle, and one that will mostly go away after Walter's work on Nova is completed | 19:45 |
| *** openstackgerrit has quit IRC | 19:46 | |
| *** openstackgerrit has joined #openstack-cinder | 19:47 | |
| *** lpetrut has quit IRC | 19:48 | |
| *** jasondotstar has joined #openstack-cinder | 19:50 | |
| geguileo | thingee: As far as I can see the example dulek mentioned should be solved if Cinder did the resource exclusion properly | 19:52 |
| geguileo | thingee: I'll ask bluex tomorrow | 19:52 |
| *** lpetrut has joined #openstack-cinder | 19:54 | |
| *** lpetrut has quit IRC | 19:55 | |
| geguileo | thingee: Sorry I couldn't give more info :-( | 19:56 |
| *** annashen has quit IRC | 19:56 | |
| *** ig0r_ has quit IRC | 19:56 | |
| openstackgerrit | James Carey proposed openstack/cinder: Sync scheduler module from oslo-incubator https://review.openstack.org/158534 | 19:56 |
| *** julim has joined #openstack-cinder | 19:57 | |
| *** annashen has joined #openstack-cinder | 19:57 | |
| *** belmoreira has joined #openstack-cinder | 19:57 | |
| *** haomaiwang has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-cinder | 20:02 | |
| *** e0ne has joined #openstack-cinder | 20:11 | |
| e0ne | thingee: thanks for approving rootwrap spec | 20:11 |
| *** mriedem1 is now known as mriedem_away | 20:12 | |
| hemna | xyang, back. any updates? need any help ? | 20:13 |
| *** annegentle has quit IRC | 20:13 | |
| *** annegentle has joined #openstack-cinder | 20:13 | |
| e0ne | jgriffith, hemna: is any chance that we'll implemtn https://blueprints.launchpad.net/cinder/+spec/use-cinder-without-nova for liberty? | 20:14 |
| hemna | e0ne, I think it's too late for that guy for Liberty | 20:14 |
| hemna | I'm definitely still interested in it. | 20:15 |
| e0ne | me too | 20:15 |
| e0ne | we still need to have a spec and/or api poc | 20:15 |
| hemna | yah, I think jgriffith had mentioned he had an idea for a new API that could help | 20:16 |
| hemna | and then the discussion was where to put the tool that did all of the coordination and discovery | 20:16 |
| e0ne | yep. that's why i'm asking about it | 20:16 |
| hemna | I still like the cinderclient change | 20:16 |
| hemna | but others didn't | 20:16 |
| *** openstackgerrit has quit IRC | 20:16 | |
| e0ne | it would be confusing | 20:16 |
| hemna | yah it would be | 20:17 |
| e0ne | now, we've got attach in cinderclient | 20:17 |
| *** openstackgerrit has joined #openstack-cinder | 20:17 | |
| e0ne | and we know what it does:) | 20:17 |
| hemna | yah, which is a call to update the db for a volume. | 20:17 |
| hemna | bleh | 20:17 |
| e0ne | :) | 20:17 |
| e0ne | so, i would like ho have some kind of brick- or cinder-baremetal-client | 20:18 |
| *** Apoorva_ has joined #openstack-cinder | 20:18 | |
| xyang1 | hemna: hi | 20:23 |
| hemna | xyang1, hey | 20:23 |
| vilobhmm | e0ne, xyang1 : can you please review this | 20:23 |
| vilobhmm | https://review.openstack.org/#/c/206170/4 | 20:23 |
| thingee | hemna: thoughts? https://review.openstack.org/#/c/207101/1 | 20:23 |
| vilobhmm | it has already lots of +1 and + 2 | 20:23 |
| vilobhmm | dont want to end up in the rebase cycles | 20:23 |
| hemna | thingee, I'll check it out | 20:24 |
| xyang1 | hemna: so I'm thinking about adding the following to devstack/inc/rootwrap/configure_rootwrap | 20:25 |
| xyang1 | hemna: if [ "${project}" == "nova" ]; then | 20:25 |
| xyang1 | sudo install -o root -g root -m 644 $rootwrap_brick_dir/rootwrap.d/*.filters /etc/${project}/rootwrap.d | 20:25 |
| xyang1 | fi | 20:25 |
| hemna | xyang1, hrmm. so why not add a generic function in inc/rootwrap for installing a lib's rootwrap file. Then in lib/nova make a call to it to install the os-brick.filters file, then you can reuse that in lib/cinder as well. | 20:27 |
| hemna | and then inc/rootwrap doesn't have checks for nova, and cinder | 20:27 |
| hemna | xyang1, does that make sense ? | 20:29 |
| xyang1 | hemna: yes, but I don't know how to tell it is a lib there | 20:30 |
| hemna | xyang1, so the call should include 2 params, where the file is, and where you want to put it. basically just a copy operation at that point. | 20:31 |
| hemna | xyang1, I think we should avoid putting project specific checks into the inc/rootwrap functions. | 20:32 |
| hemna | e0ne, heh yah. we have lots of things to change. https://etherpad.openstack.org/p/CinderNovaAPI | 20:32 |
| xyang1 | hemna: makes sense. I just need to figure out how to specify that is os-brick | 20:32 |
| e0ne | hemna: it'd great plan. did you talk about cinde-nova liaisons? | 20:33 |
| hemna | e0ne, I think scottda got assigned to be the nova liason. not me. | 20:34 |
| e0ne | hemna: oh, great! so that's idea not a new:) | 20:35 |
| *** vokt has joined #openstack-cinder | 20:35 | |
| thingee | jaypipes: would it be possible for scottda to do the update on https://review.openstack.org/#/c/84048/ ? | 20:38 |
| thingee | jaypipes: I'm also unsure if johnthetubaguy would grant it at this point. | 20:38 |
| xyang1 | hemna: # configure_rootwrap lib | 20:39 |
| xyang1 | function configure_rootwrap_lib { | 20:39 |
| xyang1 | local lib=$1 | 20:39 |
| xyang1 | local project=$2 | 20:39 |
| xyang1 | # Fix me | 20:39 |
| xyang1 | local lib_dir="" | 20:39 |
| xyang1 | local rootwrap_conf_src_dir="${lib_dir}/etc/${lib}" | 20:39 |
| xyang1 | sudo install -o root -g root -m 644 $rootwrap_conf_src_dir/rootwrap.d/*.filters /etc/${proj | 20:39 |
| xyang1 | ect}/rootwrap.d | 20:39 |
| xyang1 | } | 20:39 |
| *** annegentle has quit IRC | 20:40 | |
| hemna | heh | 20:40 |
| xyang1 | hemna: should have used paste bin:) | 20:41 |
| hemna | so I just did an install of os-brick with my os-brick.filters patch, and the etc dir doesn't get installed | 20:41 |
| xyang1 | where is it then? | 20:41 |
| xyang1 | should still be part of os-brick, right | 20:42 |
| hemna | http://paste.openstack.org/show/406503/ | 20:42 |
| *** Yogi1 has quit IRC | 20:42 | |
| hemna | have to tell setup.py to install the etc dir along with it I guess. | 20:42 |
| hemna | needs some pbr magic | 20:43 |
| *** thangp has quit IRC | 20:43 | |
| *** akshai has quit IRC | 20:44 | |
| *** akerr has quit IRC | 20:44 | |
| xyang1 | oh, ok | 20:44 |
| hemna | http://docs.openstack.org/developer/pbr/#files | 20:45 |
| *** Guest10__ has quit IRC | 20:46 | |
| *** bswartz has quit IRC | 20:46 | |
| *** openstackgerrit has quit IRC | 20:46 | |
| *** openstackgerrit has joined #openstack-cinder | 20:47 | |
| xyang1 | hemna: so those need to be added to setup.cfg of os-brick? | 20:48 |
| hemna | yah I think so | 20:48 |
| hemna | or the filters won't get installed | 20:48 |
| hemna | xyang1, http://paste.openstack.org/show/406504/ | 20:48 |
| hemna | so that installs them in /usr/local/etc/os_brick | 20:48 |
| xyang1 | ok | 20:49 |
| openstackgerrit | Kurt Martin proposed openstack/cinder: Over subscription for HP LeftHand iSCSI driver https://review.openstack.org/207248 | 20:49 |
| thingee | tbarron: there are comments in ps3 from Lisa https://review.openstack.org/#/c/203215/ | 20:49 |
| hemna | and pip show os-brick doesn't show the location of the files. so i guess it's ok to assume /usr/local/etc/os_brick ? | 20:50 |
| xyang1 | hemna: for me it is different | 20:50 |
| hemna | xyang1, you can see the installed files in the output from that paste | 20:50 |
| *** lpetrut has joined #openstack-cinder | 20:51 | |
| hemna | xyang1, the data files are installed in a different location for you ? | 20:51 |
| hemna | with that setup.cfg patch included in the paste ? | 20:51 |
| hemna | xyang1, https://gist.github.com/WaltHP/e2b07a660b421f5ebfed | 20:52 |
| xyang1 | hemna: I mean the /usr/local part is different. /usr/local/lib/python2.7/dist-packages | 20:53 |
| hemna | xyang1, right that should be the lib itself and it's .py files | 20:54 |
| hemna | but not the os-brick.filters file | 20:54 |
| xyang1 | hemna: oh, you are saying etc is installed at /usr/local/etc always | 20:55 |
| hemna | yah, hold on I have another patch gist for you | 20:55 |
| *** jasondotstar has quit IRC | 20:56 | |
| hemna | xyang1, https://gist.github.com/d760d736861a972b6805 | 20:56 |
| hemna | xyang1, http://paste.openstack.org/show/406505/ | 20:57 |
| *** jamielennox|away is now known as jamielennox | 20:57 | |
| hemna | you can see in that paste it installs the os-brick.filters file into /usr/local/etc/os-brick/rootwrap.d | 20:57 |
| xyang1 | yes, I see | 20:57 |
| thingee | hemna: nevermind on that spec. I mean feel free to review it...I'm just undecided with tooz atm | 20:58 |
| hemna | yah | 20:59 |
| hemna | thingee, so the tooz thing | 20:59 |
| hemna | I'm 'ok' with it as long as it's a short term approach. | 20:59 |
| thingee | hemna: I would like to speak at the midcycle about my thoughts there. jgriffith I'm sure is going to share as well. | 20:59 |
| hemna | presuming that tooz works really well and doesn't have it's own set of problems that now Cinder will have in inherit. | 21:00 |
| openstackgerrit | Walter A. Boring IV (hemna) proposed openstack/os-brick: Add rootwrap filters https://review.openstack.org/207553 | 21:00 |
| hemna | xyang1, ok that update patch has the setup.cfg changes to install the filters file. | 21:00 |
| thingee | hemna: so I have no knowledge of how any of this works so ever, but another idea is using swift's ring solution. akanda is doing work in bringing that in | 21:00 |
| *** haomaiwang has quit IRC | 21:01 | |
| *** lpetrut has quit IRC | 21:01 | |
| *** openstack has joined #openstack-cinder | 21:09 | |
| thingee | hemna: yeah lets get that update. leeantho still have spare cycles with this in l-3? | 21:09 |
| hemna | hehe | 21:10 |
| hemna | I can work with leeantho on the spec and see if we can get it updated. | 21:10 |
| thingee | hemna: I also mean cycles to do the actual change too :) | 21:11 |
| leeantho | I have some time after I finish a few other things. I can work on some of it after that | 21:11 |
| thingee | tbarron: ok, I'll hold off until tomorrow. Please try to finish up discussions :) | 21:12 |
| thingee | leeantho: thanks | 21:12 |
| thingee | jaypipes: also don't be ashamed. People are busy. I get it. Trust me, I understand. | 21:12 |
| hemna | thingee, yah. I'm probably going to help out with that effort as well. | 21:13 |
| scottda | OK, but we cannot return VolumeIsBusy to nova until they are able to catch the exception. That should just be a bugfix (no API change) so should be able to get in L | 21:13 |
| thingee | scottda: +1 | 21:13 |
| thingee | We have a bug | 21:13 |
| thingee | I kept bugging winston-d_zZZ about it, but he has been busy | 21:14 |
| hemna | so that atomic status change spec is basically a bug fix | 21:14 |
| hemna | just for db updates for status | 21:14 |
| * thingee looks for it | 21:14 | |
| scottda | And if we do this, where does that leave a DLM/tooz solution? Is it even necessary? | 21:14 |
| hemna | scottda, yah the nova side needs to be updated before we can change the Cinder API to report VolumeIsBusy | 21:14 |
| hemna | scottda, I think if we get that in place, then we don't need locks in the volume manager. | 21:15 |
| *** annashen has quit IRC | 21:15 | |
| *** mylu has quit IRC | 21:16 | |
| scottda | hemna: yeah, and that seems to be the long-term solution we want (this is pointed out in Add Cinder Volume A/A HA lock spec) | 21:16 |
| hemna | yup | 21:16 |
| *** annegentle has joined #openstack-cinder | 21:16 | |
| * thingee would really like to solve this problem in liberty versus introducing a DLM at this point | 21:16 | |
| thingee | I'd like to just see what problems we have after introducing this change | 21:16 |
| scottda | so, it seems that is could be short-term (fixed in L) not long-term to report VolumeIsBusy | 21:16 |
| hemna | so the real catch here is Nova is FF after today. | 21:16 |
| thingee | and *see* what problems by actually testing | 21:16 |
| scottda | thingee: +1 | 21:17 |
| thingee | hemna: it's a bug not a feature | 21:17 |
| hemna | so we'd have to file a bug and see if we can get buyoff on it for adding VolumeIsBusy checks. | 21:17 |
| thingee | hemna: we have one, I'm digging it up | 21:17 |
| hemna | I think leeantho might have already filed a bug on this for me a while ack | 21:17 |
| hemna | back | 21:17 |
| thingee | vilobhmm filed one for me :P | 21:17 |
| * thingee continues to look for it | 21:17 | |
| *** mylu has joined #openstack-cinder | 21:18 | |
| *** annashen has joined #openstack-cinder | 21:18 | |
| leeantho | hemna, yes I have an old WIP patch: https://review.openstack.org/#/c/153748/ | 21:18 |
| scottda | https://bugs.launchpad.net/nova/+bug/1458958 | 21:18 |
| openstack | Launchpad bug 1460044 in OpenStack Compute (nova) "duplicate for #1458958 Data loss can occur if cinder attach fails" [High,In progress] - Assigned to Vipin Balachandran (vbala) | 21:18 |
| hemna | leeantho, did we have a bug for the nova side though ? | 21:18 |
| jaypipes | thingee: cheers and thanks for your understanding | 21:19 |
| *** zul has joined #openstack-cinder | 21:19 | |
| hemna | scottda, :) thanks | 21:19 |
| thingee | jaypipes: can you answer my question on this being an issue of getting liberty at this point? | 21:19 |
| thingee | :) | 21:19 |
| *** mylu has quit IRC | 21:19 | |
| thingee | jaypipes: if you want me to just figure that out with johnthetubaguy, I can do that as well. | 21:20 |
| tbarron | thingee: thanks. I have responded in the review as well, as maybe I do understand what she's suggesting after all. | 21:20 |
| jaypipes | thingee: I will likely need to talk with mriedem and johnthetubaguy about it. it will likely be a stretch, to be entirely up-front with you. I will see what I can do. | 21:20 |
| thingee | jaypipes, scottda: understood, thanks | 21:21 |
| *** crose has quit IRC | 21:22 | |
| thingee | breitz: ping | 21:23 |
| thingee | breitz: how are we doing with getting started? | 21:23 |
| thingee | scottda, hemna ah thanks, yeah that's the bug | 21:24 |
| scottda | I'll get with winston-d_zZZ on that to see if he needs help | 21:24 |
| hemna | hrmm that bug is marked a dupe of 1460044, which I don't think is related. | 21:25 |
| thingee | winston-d_zZZ: I think he's just busy. I remember in the cinder meeting he said he couldn't do it, then changed his mind...but still nothing | 21:25 |
| thingee | could probably just take it | 21:25 |
| thingee | scottda: whoops ^ | 21:26 |
| scottda | also note that jaypipes mentioned in the bug a change to cinderclient to return proper HTTP error ... | 21:26 |
| hemna | scottda, yah I noticed that. | 21:26 |
| hemna | scottda, so if we do that, is a Nova change needed? | 21:26 |
| *** bswartz has joined #openstack-cinder | 21:27 | |
| hemna | I think nova would just get an exception from cinderclient method call at that point and puke, which it's not currently expecting. | 21:27 |
| scottda | I would think so. I'll look at the code and/or test to see... | 21:27 |
| hemna | I'm pretty sure that's the case from what I recall. | 21:28 |
| scottda | yes, I think Nova will still fail, he's just saying that's the proper HTTP return. | 21:28 |
| hemna | sure | 21:28 |
| scottda | so, the cinderclient change will probably be required to make reviewers happy, that's all I'm saying on this. | 21:28 |
| hemna | scottda, +1 | 21:29 |
| breitz | thingee, pong | 21:29 |
| breitz | doing ok - got devstack up and trying to play with image cache - reading code - trying to set breakpoints - the usual thrashing | 21:30 |
| *** annashen_ has joined #openstack-cinder | 21:30 | |
| openstackgerrit | Erickson Filipe Guedes dos Santos proposed openstack/cinder: Nested Quota Driver: Get Project Hierarchy https://review.openstack.org/206171 | 21:31 |
| thingee | breitz: excellent. reach out to patrickeast about any questions :D | 21:31 |
| breitz | I have been - he has been very helpful | 21:31 |
| *** annashen_ has quit IRC | 21:35 | |
| *** ub2 has joined #openstack-cinder | 21:35 | |
| scottda | breitz: I see you are coming to the Cinder mid-cycle next week. | 21:36 |
| *** vilobhmm has left #openstack-cinder | 21:37 | |
| *** ub has quit IRC | 21:38 | |
| *** edtubill has left #openstack-cinder | 21:39 | |
| thingee | breitz: who are you with? | 21:39 |
| thingee | jwang: what's going on with the nimble CI? | 21:40 |
| * thingee is going to disappear soon.. plane landing soon | 21:40 | |
| scottda | thingee: I don't like this: "The nova-manage isn't properly documented, but it's going away in Folsom...." | 21:42 |
| scottda | https://wiki.openstack.org/wiki/NovaManage | 21:42 |
| thingee | lololo | 21:42 |
| thingee | jaypipes: ^ | 21:42 |
| scottda | I thought jaypipes said nova-manage was supported and here to stay? | 21:43 |
| thingee | Cinder just found a reason to keep it around! | 21:43 |
| scottda | hehe | 21:43 |
| thingee | it's a trap | 21:43 |
| scottda | It's probably OK...I see them talking about adding things to nova-manage in theiNOva meeng ATM | 21:45 |
| scottda | in the Nova meeing | 21:46 |
| * thingee is going to propose a spec to remove it | 21:46 | |
| thingee | I was there at folsom. I remember | 21:46 |
| *** markvoelker_ has quit IRC | 21:46 | |
| * thingee disappears | 21:47 | |
| *** jasondotstar has joined #openstack-cinder | 21:48 | |
| *** annegentle has quit IRC | 21:49 | |
| *** rushiagr is now known as rushiagr_away | 21:50 | |
| jaypipes | scottda: "it's going away in Folsom" LOL. | 21:51 |
| jaypipes | scottda: no, nova-manage isn't going anywhere. | 21:51 |
| jaypipes | http://datanetpacific.com/wp-content/uploads/2015/05/targeting_computer_in_blog_3.gif | 21:51 |
| *** ig0r__ has quit IRC | 21:51 | |
| *** markstur_ has joined #openstack-cinder | 21:51 | |
| scottda | jaypipes: thingee https://review.openstack.org/#/c/84048/44 | 21:53 |
| *** markstur_ has left #openstack-cinder | 21:53 | |
| scottda | Uploaded patch to use nova-manage | 21:54 |
| *** jungleboyj has joined #openstack-cinder | 21:59 | |
| *** annegentle has joined #openstack-cinder | 22:00 | |
| *** haomaiwang has quit IRC | 22:01 | |
| breitz | thingee: <sorry stepped away from my machine> I work for Oracle - if that is what you were asking | 22:01 |
| *** jwcroppe has joined #openstack-cinder | 22:02 | |
| *** annashen has quit IRC | 22:02 | |
| *** haomaiwang has joined #openstack-cinder | 22:02 | |
| *** e0ne has quit IRC | 22:02 | |
| scottda | breitz: Are you in Broomfield, CO? | 22:05 |
| breitz | No - I wish - I work from my home in Minneapolis, MN | 22:06 |
| breitz | I used to live in CO, I miss it | 22:06 |
| scottda | ahh..OK. | 22:06 |
| scottda | But you are coming to Fort Collins next week? | 22:07 |
| breitz | yes - that is my plan | 22:07 |
| breitz | tickets purchased and all that...... | 22:07 |
| *** markvoelker has joined #openstack-cinder | 22:08 | |
| scottda | cool See you there (here) | 22:09 |
| breitz | my flight is on Tuesday morning - so with any luck I'll be there before noon | 22:09 |
| *** jamielennox is now known as jamielennox|away | 22:15 | |
| zigo | mriedem: I am now. | 22:20 |
| *** jordanP has quit IRC | 22:22 | |
| openstackgerrit | Tomoki Sekiyama proposed openstack/cinder: Efficient image transfer for Glance cinder store https://review.openstack.org/201754 | 22:22 |
| *** markvoelker_ has joined #openstack-cinder | 22:32 | |
| *** Lee1092 has quit IRC | 22:33 | |
| *** markvoelker has quit IRC | 22:35 | |
| *** belmoreira has quit IRC | 22:37 | |
| *** eharney has quit IRC | 22:38 | |
| *** primechuck has quit IRC | 22:39 | |
| *** annegentle has quit IRC | 22:44 | |
| *** harlowja has quit IRC | 22:44 | |
| *** harlowja has joined #openstack-cinder | 22:44 | |
| *** dustins has quit IRC | 22:52 | |
| *** boris-42 has quit IRC | 23:00 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** haomaiwa_ has joined #openstack-cinder | 23:02 | |
| *** zhenguo has joined #openstack-cinder | 23:03 | |
| *** xyang1 has quit IRC | 23:04 | |
| hemna | xyang, I'm about to head out for a while | 23:06 |
| hemna | xyang, my son has soccer until 8pm locally | 23:06 |
| openstackgerrit | Patrick East proposed openstack/cinder: Generic image-volume cache https://review.openstack.org/195795 | 23:06 |
| *** krypton has joined #openstack-cinder | 23:09 | |
| *** markvoelker_ has quit IRC | 23:14 | |
| *** _cjones_ has quit IRC | 23:24 | |
| *** jungler has quit IRC | 23:27 | |
| *** jungler has joined #openstack-cinder | 23:28 | |
| *** diogogmt has quit IRC | 23:28 | |
| *** dims_ has joined #openstack-cinder | 23:31 | |
| *** dims_ has quit IRC | 23:31 | |
| *** annashen has joined #openstack-cinder | 23:31 | |
| *** dims has quit IRC | 23:35 | |
| *** annashen has quit IRC | 23:36 | |
| *** jaypipes has quit IRC | 23:39 | |
| *** daneyon_ has quit IRC | 23:41 | |
| *** vokt has quit IRC | 23:43 | |
| *** marcusvrn has quit IRC | 23:43 | |
| *** lcurtis has quit IRC | 23:46 | |
| *** patrickeast has quit IRC | 23:54 | |
| *** dannywilson has quit IRC | 23:55 | |
| *** salv-orlando has quit IRC | 23:57 | |
| openstackgerrit | John Griffith proposed openstack/cinder-specs: Cinder Volume replication V2 https://review.openstack.org/155644 | 23:57 |
| anteaya | thingee: the account is disabled: http://lists.openstack.org/pipermail/third-party-announce/2015-July/000259.html | 23:57 |
| jgriffith | thingee: ^^ | 23:58 |
| jgriffith | thingee: I stripped out the detail while keeping something close to what the patch is shaping up to be. Hopefully that can just be approved and we can focus on the patch. | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
