| *** annashen has joined #openstack-cinder | 00:04 | |
| *** annegent_ has quit IRC | 00:07 | |
| vilobhmm | thingee : will make sure jenkins passes and will ping you guys again :) | 00:08 |
|---|---|---|
| *** EmilienM is now known as EmilienM|afk | 00:11 | |
| *** markvoelker has quit IRC | 00:12 | |
| *** markvoelker has joined #openstack-cinder | 00:12 | |
| *** IanGovett1 has joined #openstack-cinder | 00:15 | |
| *** markvoelker has quit IRC | 00:17 | |
| *** IanGovett has quit IRC | 00:17 | |
| *** jecarey has quit IRC | 00:25 | |
| *** Miouge has quit IRC | 00:26 | |
| *** annegent_ has joined #openstack-cinder | 00:30 | |
| *** Mandell has quit IRC | 00:36 | |
| *** Mandell has joined #openstack-cinder | 00:37 | |
| *** annegent_ has quit IRC | 00:38 | |
| *** tbarron has joined #openstack-cinder | 00:38 | |
| *** patrickeast has quit IRC | 00:39 | |
| *** chipmanc has joined #openstack-cinder | 00:41 | |
| *** gary-smith has quit IRC | 00:43 | |
| *** oomichi has joined #openstack-cinder | 00:49 | |
| *** markvoelker has joined #openstack-cinder | 00:52 | |
| *** garthb_ has joined #openstack-cinder | 00:55 | |
| *** nellysmitt has joined #openstack-cinder | 00:55 | |
| *** garthb has quit IRC | 00:57 | |
| *** annashen has quit IRC | 01:00 | |
| *** nellysmitt has quit IRC | 01:00 | |
| *** markvoelker has quit IRC | 01:00 | |
| *** junhongl__ has quit IRC | 01:00 | |
| *** wpf1 has quit IRC | 01:00 | |
| *** lan_ has quit IRC | 01:00 | |
| *** jungleboyj has quit IRC | 01:01 | |
| *** jwang has joined #openstack-cinder | 01:01 | |
| *** takedakn has joined #openstack-cinder | 01:02 | |
| *** bill_az_ has quit IRC | 01:03 | |
| openstackgerrit | Li Min Liu proposed openstack/cinder: Make storwize debug log more readable https://review.openstack.org/152358 | 01:04 |
| *** rmesta has quit IRC | 01:07 | |
| *** Mandell has quit IRC | 01:08 | |
| *** zhiyan has quit IRC | 01:10 | |
| *** briancurtin has quit IRC | 01:11 | |
| *** ameade has quit IRC | 01:11 | |
| *** tbarron has quit IRC | 01:13 | |
| *** tbarron has joined #openstack-cinder | 01:14 | |
| *** Yogi1 has joined #openstack-cinder | 01:17 | |
| *** wpf1 has joined #openstack-cinder | 01:17 | |
| *** junhongl__ has joined #openstack-cinder | 01:17 | |
| *** briancurtin has joined #openstack-cinder | 01:18 | |
| *** lan_ has joined #openstack-cinder | 01:18 | |
| *** zhiyan has joined #openstack-cinder | 01:21 | |
| *** ameade has joined #openstack-cinder | 01:21 | |
| *** tbarron has quit IRC | 01:22 | |
| *** xyang has quit IRC | 01:24 | |
| *** _cjones_ has quit IRC | 01:26 | |
| *** _cjones_ has joined #openstack-cinder | 01:26 | |
| *** jungleboyj has joined #openstack-cinder | 01:27 | |
| *** _cjones_ has quit IRC | 01:31 | |
| *** 21WAA8C7O has joined #openstack-cinder | 01:32 | |
| *** IanGovett1 has quit IRC | 01:32 | |
| *** primechuck has joined #openstack-cinder | 01:35 | |
| *** garthb_ has quit IRC | 01:44 | |
| *** Mandell has joined #openstack-cinder | 01:45 | |
| *** diegows has quit IRC | 01:47 | |
| *** Yogi1 has quit IRC | 01:49 | |
| *** Mandell has quit IRC | 01:49 | |
| *** Mandell has joined #openstack-cinder | 01:50 | |
| *** markvoelker has joined #openstack-cinder | 01:56 | |
| *** takedakn has quit IRC | 01:58 | |
| *** jecarey has joined #openstack-cinder | 01:58 | |
| *** tbarron has joined #openstack-cinder | 01:59 | |
| *** tbarron has joined #openstack-cinder | 02:00 | |
| *** Guest95188 has quit IRC | 02:00 | |
| anish | thingee: I'm not sure what baseline to follow here for the driver certs | 02:01 |
| *** primechuck has quit IRC | 02:01 | |
| anish | I see volume_extend/volume_transfer failing | 02:01 |
| *** markvoelker has quit IRC | 02:01 | |
| anish | but these are not handled by our iscsi implementation, we're simply an iscsi_helper | 02:02 |
| *** theanalyst has joined #openstack-cinder | 02:04 | |
| *** Mandell has quit IRC | 02:08 | |
| openstackgerrit | Mitsuhiro Tanino proposed openstack/cinder: Refactoring for export functions in Target object https://review.openstack.org/148038 | 02:08 |
| *** leeantho has quit IRC | 02:08 | |
| *** kaisers1 has joined #openstack-cinder | 02:11 | |
| *** hflai has quit IRC | 02:12 | |
| *** mberlin1 has quit IRC | 02:12 | |
| *** mberlin has joined #openstack-cinder | 02:12 | |
| *** Longgeek has joined #openstack-cinder | 02:12 | |
| openstackgerrit | Tina Tang proposed openstack/cinder: Pool-aware scheduler support in EMC VNX Cinder driver https://review.openstack.org/147408 | 02:13 |
| *** Longgeek has quit IRC | 02:13 | |
| *** kaisers has quit IRC | 02:13 | |
| *** Longgeek has joined #openstack-cinder | 02:15 | |
| thingee | anish: baseline for what? | 02:16 |
| anish | should all tests be passing with an lvm backend for cinder_driver_cert.sh | 02:16 |
| thingee | anish: yes | 02:17 |
| anish | thingee: but some of this stuff is completely out of our control. I have a failure in dmcrypt for eg | 02:21 |
| thingee | anish: would need to take a closer look. SCST, TGT, LIO have gone through the same test. | 02:24 |
| *** Yogi1 has joined #openstack-cinder | 02:24 | |
| anish | ok, let me run against them and make sure we pass everythign they pass at least | 02:24 |
| *** hflai has joined #openstack-cinder | 02:25 | |
| thingee | anish: TGT is gated in openstack on these tests for every cinder patch that comes in. LIO is gated on other Cinder volume driver CIs. SCST just posted their results passing here https://drive.google.com/file/d/0Bxs_qlih5bLYQ1JXb0xDV3huLU0/view?usp=sharing | 02:27 |
| *** Mandell has joined #openstack-cinder | 02:27 | |
| thingee | anish: I think it would be better for us to see your stacktraces of what you're failing on and help. | 02:28 |
| anish | thingee: I have all the logs, just finished a run. any file in particular ? | 02:28 |
| thingee | yes | 02:28 |
| thingee | although dinner is ready here and I already promised five mins to be over there ;) | 02:29 |
| anish | go ahead | 02:29 |
| anish | I'll make sure it's not a setup issue in the meantime | 02:29 |
| thingee | I'll be back, but paste me your logs and I'll take a look for ya | 02:29 |
| thingee | paste.openstack.org | 02:29 |
| anish | thingee: just c-vol ? | 02:29 |
| thingee | anish: If I don't see you when I get back, have a good night. | 02:29 |
| anish | thingee: sure. any files you would like besides c-vol ? | 02:30 |
| thingee | anish: do the stacktraces show up in the output of cinder_driver_cert.sh? | 02:31 |
| anish | yes, at the very end | 02:31 |
| thingee | ok, give me that and c-vol please. | 02:31 |
| thingee | I'll bb | 02:31 |
| *** primechuck has joined #openstack-cinder | 02:33 | |
| anish | thingee: c-vol http://paste.openstack.org/show/165900/ traces : http://paste.openstack.org/show/165901/ | 02:34 |
| *** vilobhmm has quit IRC | 02:36 | |
| *** vilobhmm has joined #openstack-cinder | 02:38 | |
| *** mwichmann has quit IRC | 02:38 | |
| *** Mandell has quit IRC | 02:41 | |
| anish | thingee: there is an actual failure in there as well, which I am addressing right now | 02:41 |
| *** scottda_ has joined #openstack-cinder | 02:44 | |
| *** scottda_ has quit IRC | 02:44 | |
| thingee | anish: your logs for c-vol don't really show everything. It just show setting the conf opts. | 02:45 |
| thingee | anish: copying the logs from a screen session won't work well. You can tell devstack to have the screen logs put to a file: http://docs.openstack.org/developer/devstack/configuration.html#logging-the-service-output | 02:45 |
| *** jungleboyj has quit IRC | 02:51 | |
| anish | thingee: I think they might have been truncated | 02:54 |
| anish | I'm copying /opt/stack/logs/c-vol completely | 02:54 |
| thingee | anish: oh bummer. well volume drivers used to open a bug and attach files https://bugs.launchpad.net/cinder | 02:55 |
| *** nellysmitt has joined #openstack-cinder | 02:56 | |
| *** rushiagr_away is now known as rushiagr | 02:57 | |
| anish | thingee: yeah, retried. it's getting truncated when I paste into paste.openstack.org | 02:57 |
| *** markvoelker has joined #openstack-cinder | 02:57 | |
| anish | thingee: https://www.dropbox.com/s/5a5grf4tuygnpnm/c-vol.log?dl=0 | 02:58 |
| winston-d | xyang1: you were looking for me? | 02:58 |
| anish | or I can open a bug if you prefer | 02:58 |
| thingee | winston-d: probably about the rpc stuff I mentioned to ya yesterday :) | 02:59 |
| thingee | anish: no worries, just trying to figure out a way for you to upload them...taking a look | 02:59 |
| winston-d | thingee: ah, ok. | 03:00 |
| thingee | anish: ah nice, even has color :) | 03:00 |
| *** nellysmitt has quit IRC | 03:01 | |
| *** markvoelker has quit IRC | 03:02 | |
| xyang1 | winston-d: Hi, can you take a look of the over subscription patch again? I merged the virtual capacity weigher and capacity weigher into one. Have a question on total capacity if it is infinite or unknown | 03:04 |
| thingee | xyang1: oh right, my bad | 03:04 |
| thingee | xyang1: wow you really do stay up late | 03:04 |
| *** chipmanc has quit IRC | 03:04 | |
| xyang1 | thingee: Just 10:00pm:) | 03:04 |
| thingee | xyang1: I'm ready for sleep | 03:05 |
| xyang1 | thingee: Good night:) | 03:05 |
| thingee | xyang1: no, no...not yet | 03:05 |
| xyang1 | thingee: :) | 03:06 |
| winston-d | xyang1: sure, i'll go through oversubscription patch now. | 03:06 |
| xyang1 | winston-d: Thanks! | 03:06 |
| *** vilobhmm has quit IRC | 03:07 | |
| *** vilobhmm has joined #openstack-cinder | 03:08 | |
| *** tbarron has quit IRC | 03:08 | |
| *** chipmanc has joined #openstack-cinder | 03:09 | |
| *** theanalyst has quit IRC | 03:10 | |
| *** theanalyst has joined #openstack-cinder | 03:12 | |
| winston-d | xyang1: i know it's odd that a backend report total as infinite but actual capacity for free, but if that happens, and unfortunately the reserved_percentage is >0, we'd not let it pass capacity filter. | 03:12 |
| winston-d | xyang1: what do you think? | 03:12 |
| winston-d | xyang1: your current implementation will break if that happens, btw. | 03:13 |
| *** rushiagr is now known as rushiagr_away | 03:13 | |
| winston-d | xyang1: and you missed my comment about removing self.provisioned_capacity_gb += volume_gb in host_manager.py consume_from_volume(). | 03:14 |
| xyang1 | winston-d: I did not miss it. Maybe i forgot to comment. So i removed it because it was already added in filter | 03:15 |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset snapshot image metadata https://review.openstack.org/149163 | 03:16 |
| xyang1 | Provisioned capacity is not calculated by the scheduler like the allocated cap | 03:16 |
| xyang1 | winston-d: So i just added volume size to provisioned cap temporarily in filter | 03:17 |
| *** vilobhmm has quit IRC | 03:18 | |
| xyang1 | winston-d: So you want me to always return true if total is infinite? | 03:19 |
| *** xyang has joined #openstack-cinder | 03:19 | |
| winston-d | xyang1: are you talking about line 85 in capacity filter? | 03:20 |
| xyang1 | Let me see | 03:20 |
| xyang1 | 88 | 03:21 |
| thingee | anish: are the cert test logs being truncated too? | 03:21 |
| xyang1 | winston-d: +volume_size | 03:21 |
| winston-d | xyang1: provisioned_capacity_gb of HostState() should be updated every time scheduler made a decision, it's done in consume_from_volume(). | 03:21 |
| *** vilobhmm has joined #openstack-cinder | 03:21 | |
| winston-d | xyang1: what you did was to calcualte a value for provsioned_ratio, these two are for totaly different purpose. | 03:22 |
| *** Yogi1 has quit IRC | 03:22 | |
| xyang1 | winston-d: Does it just add volume gb once? | 03:23 |
| xyang1 | winston-d: In consume from volime for one volume request? | 03:23 |
| winston-d | filter/weigher shouldn't touch HostState() at all | 03:23 |
| winston-d | it would mess up HostState() for a backend. | 03:24 |
| xyang1 | winston-d: Ok, i'll move it back to consume from volume | 03:24 |
| winston-d | your code didn't do that, which is good. | 03:24 |
| winston-d | leave it to consume_from_volume() | 03:24 |
| winston-d | so yeah, don't remove it in consume_from_volume() | 03:25 |
| xyang1 | winston-d: So provisioned cap will be reported by driver every time, will it be problem if we add it in consume from vol? | 03:25 |
| xyang1 | winston-d: Volume_gb in consume from volume is the new volume to be created, correct? | 03:27 |
| *** annegentle has quit IRC | 03:28 | |
| winston-d | xyang1: not at all, consume_from_vol() is to make sure this scheduler instance has the latest stats of backend *after* it makes a decision and *before* it receives next update from backend. | 03:28 |
| xyang1 | winston-d: Ok, that works. Will fix it | 03:29 |
| winston-d | xyang1: basically, scheduler use this function to update its cache of backend stat. and this cache will always be refreshed by latest stats from backend. | 03:29 |
| winston-d | so no conflict | 03:30 |
| xyang1 | winston-d: that is good. About total capacity, should I always return True if it is infinite? | 03:30 |
| *** vilobhmm has quit IRC | 03:30 | |
| *** tellesnobrega_ has joined #openstack-cinder | 03:31 | |
| winston-d | xyang1: it depends. I left comment in your patch. | 03:31 |
| xyang1 | winston-d: Ok | 03:31 |
| anish | thingee: not sure, will put them on dropbox as well | 03:32 |
| anish | thingee: here you go https://www.dropbox.com/s/2pgytxrwpnpaqqt/stacktraces?dl=0 | 03:33 |
| xyang1 | winston-d: so handle it the same way as if free capacity is infinite | 03:33 |
| *** markvoelker has joined #openstack-cinder | 03:33 | |
| *** hemna_ has joined #openstack-cinder | 03:33 | |
| DaveChen | DuncanT: hi Duncan, ping? | 03:34 |
| winston-d | xyang1: well, if total_cap is infinite, and reserved_percentage>0, it should not pass cap filter | 03:34 |
| *** bkopilov has quit IRC | 03:37 | |
| xyang1 | winston-d: Ok, will return False if total is infinite and reserved percent >0. Then if either free or total is infinite or unknown, will give lowest weight | 03:37 |
| xyang1 | winston-d: Sounds right? | 03:38 |
| thingee | anish: so it looks like TestEncryptedCinderVolumes actually is a real failure in your target driver | 03:39 |
| thingee | "Details: (TestEncryptedCinderVolumes:_run_cleanups) Failed to delete volume 612aee97-4e48-43de-9f92-b9629dccbac0 within the required time (196 s)" | 03:39 |
| thingee | anish: that's from the stacktraces in the cert log | 03:39 |
| winston-d | xyang1: yes, full version is if total_cap is infinite, free_cap is actual value, and reserved_percentage>0, it should not pass cap filter; otherwise whe total/free are all infinite, give it lowest weight | 03:40 |
| anish | is there any way to run the tests individually ? | 03:40 |
| thingee | anish: 2015-02-02 18:06:30.410 DEBUG cinder.volume.targets.cxt [req-e433a9d2-eb71-4ce8-a2de-ab94c6ee81a7 883d5dece1d743df91cf0ef99f582431 c6666e06549f4179a8647ba395972996] Failed to open config for volume-612aee97-4e48-43de-9f92-b9629dccbac0: [Errno 2] No such file or directory: u'/opt/stack/data/cinder/volumes/cxt/volume-612aee97-4e48-43de-9f92-b9629dccbac0' from (pid=18457) _get_target_chap_auth | 03:40 |
| thingee | /opt/stack/cinder/cinder/volume/targets/cxt.py:120 | 03:40 |
| anish | that's not an error | 03:40 |
| anish | that's acceptable | 03:40 |
| anish | if you see how the code is written, it checks if a previous persistent file was written and tries to get chap from that | 03:40 |
| anish | if not, it creates chap credential | 03:40 |
| xyang1 | winston-d: Thanks! | 03:41 |
| *** Guest81568 has joined #openstack-cinder | 03:41 | |
| thingee | anish: ah ok | 03:42 |
| winston-d | xyang1: sure | 03:42 |
| thingee | anish: ah found it | 03:42 |
| *** jungleboyj has joined #openstack-cinder | 03:43 | |
| thingee | 2015-02-02 18:07:07.341 TRACE oslo_messaging.rpc.dispatcher ISCSITargetRemoveFailed: Failed to remove iscsi target for volume 612aee97-4e48-43de-9f92-b9629dccbac0. | 03:43 |
| thingee | that's the same uuid of the volume | 03:43 |
| anish | gotcha | 03:44 |
| anish | is there any way to run the tests manually ? | 03:44 |
| anish | err | 03:44 |
| anish | I mean individually ? | 03:44 |
| *** hflai has quit IRC | 03:46 | |
| thingee | anish: if you take a look at that cinder_driver_cert.sh file, it's just doing ./tools/pretty_tox.sh volume. | 03:47 |
| thingee | you can replace volume with a dot path to an individual test I believe | 03:47 |
| thingee | tempest.scenario.test_encrypted_cinder_volumes.TestEncryptedCinderVolumes.test_encrypted_cinder_volumes_cryptsetup | 03:47 |
| anish | thanks a lot, trying that now | 03:47 |
| thingee | http://blog.jasonmeridth.com/posts/how-to-run-a-single-test-with-tox/ | 03:48 |
| thingee | :) | 03:48 |
| *** davechen_ has joined #openstack-cinder | 03:48 | |
| anish | btw I am seeing tempest.api.volume.admin.test_volume_types.VolumeTypesV1Test.test_create_get_delete_volume_with_volume_type_and_extra_specs | 03:48 |
| anish | and tempest.api.volume.admin.test_volume_types.VolumeTypesV2Test.test_create_get_delete_volume_with_volume_type_and_extra_specs fail with tgt as well | 03:48 |
| thingee | anish: if tgt was failing, we'd being seeing failures in OpenStack gate http://status.openstack.org/zuul/ :) | 03:50 |
| thingee | that means it's a environment issue. did you set those TEMPEST_* vars in local.conf as I mentioned earlier? | 03:50 |
| anish | this might just be that my vm is not good enough | 03:50 |
| anish | yes, set them | 03:50 |
| anish | though, I am not using a volume_driver, so that is not set | 03:51 |
| anish | lvm looks to be default, vendor/protocol are set | 03:51 |
| *** davechen_ has quit IRC | 03:53 | |
| *** rushiagr_away is now known as rushiagr | 03:53 | |
| thingee | I don't have your c-api c-sch logs, but it definitely doesn't reach c-vol. | 03:53 |
| anish | I'm setting this : | 03:54 |
| anish | #TEMPEST_VOLUME_DRIVER=foo | 03:54 |
| anish | TEMPEST_VOLUME_VENDOR="Chelsio Inc" | 03:54 |
| anish | TEMPEST_STORAGE_PROTOCOL=iSCSI | 03:54 |
| thingee | which is what I suspected. I don't it's making it past the scheduler | 03:54 |
| thingee | ah there's the problem | 03:54 |
| anish | need a VOLUME_DRIVER ? | 03:54 |
| thingee | no, sorry for not explaining this earlier. the volume_vedor comes from the lvm driver. | 03:54 |
| thingee | anish: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/lvm.py#L180 | 03:55 |
| anish | oh so they're always gonna get overridden | 03:56 |
| thingee | anish: oh man. Actually I led you down a bad path...it sets "Open Source" by default | 03:57 |
| thingee | anish: so actually just comment out both of those | 03:57 |
| thingee | =/ | 03:57 |
| *** primechuck has quit IRC | 03:58 | |
| anish | will go | 03:58 |
| thingee | anish: https://github.com/openstack-dev/devstack/blob/master/lib/tempest#L72 | 03:58 |
| thingee | anish: doh | 03:58 |
| anish | but first, I have a noobish question | 03:58 |
| anish | tox -e py27 -- /opt/stack/tempest/tempest/scenario/test_encrypted_cinder_volumes.py:tempest.scenario.test_encrypted_cinder_volumes.TestEncryptedCinderVolumes | 03:58 |
| anish | seems to be the wrong command to run | 03:58 |
| *** Longgeek has quit IRC | 03:59 | |
| *** hflai has joined #openstack-cinder | 03:59 | |
| *** cchipman has joined #openstack-cinder | 04:00 | |
| thingee | anish: I'm not sure with running individual tests. I always have to refigure it out when I want to do it and brute force it until it works. ... test runners always confuse me because they are different in how they do this | 04:00 |
| thingee | tox, nosetest, testr | 04:00 |
| anish | yeah, I'm definitely worse at this than you | 04:01 |
| thingee | anish: I'm just bad with computers. period. | 04:01 |
| anish | haha | 04:01 |
| *** chipmanc has quit IRC | 04:01 | |
| anish | so L72 that you pointed out is what I followd in not setting VOLUME_DRIVER | 04:01 |
| anish | but I just grepped all the logs, no TEMPEST_ set anywhere | 04:02 |
| *** annashen has joined #openstack-cinder | 04:02 | |
| thingee | anish: so I was just correcting myself from earlier, if you don't set those TEMPEST_VOLUME_VENDOR and TEMPEST_STORAGE_PROTOCOL, it'll default to the right thing you want. | 04:03 |
| thingee | from the local.conf | 04:03 |
| anish | I did want to vendor, mostly because I thought it would be a good idea for hte logs to be tagged by vendor | 04:03 |
| *** Longgeek has joined #openstack-cinder | 04:04 | |
| thingee | anish: yeah but the vendor thing gets used for something specific with volume drivers in cinder. It should match "Open Source" for the LVM driver. | 04:04 |
| thingee | I'm pretty sure it's why your extra_spec test in tempest is failing | 04:04 |
| anish | aah | 04:04 |
| thingee | because the scheduler is doing this comparison right now "Chelsio Inc" == "Open Source" | 04:05 |
| anish | will remove that, don't particularly care | 04:05 |
| *** Longgeek has quit IRC | 04:05 | |
| thingee | anish: again sorry about that. Pretty sure that'll take of that issue though. | 04:05 |
| anish | sounds great, will try now | 04:05 |
| anish | I may have found another possible issue | 04:07 |
| anish | the driver cert script starts tgt automatically | 04:07 |
| anish | this prevents our target driver from running, as tgt is bound to port 3260 already | 04:08 |
| *** harlowja has quit IRC | 04:08 | |
| anish | easy enough hack | 04:09 |
| *** Longgeek has joined #openstack-cinder | 04:09 | |
| hemna_ | man | 04:10 |
| hemna_ | the EMC driver changes are painful | 04:10 |
| thingee | hemna_: ha I wondered what happened to you | 04:10 |
| hemna_ | not only is there a 7000 line driver change, but it's also dependent on 3 other patches that have boatloads of changes | 04:11 |
| hemna_ | *sigh* | 04:11 |
| hemna_ | heh I had to drive home and cook dinner | 04:11 |
| hemna_ | slammed at work with meetings and such. bleh | 04:11 |
| thingee | alright, I'll bb | 04:13 |
| *** primechuck has joined #openstack-cinder | 04:14 | |
| thingee | hemna_: and whoa. let me know when drivers out of tree sound good | 04:14 |
| hemna_ | lol, yah | 04:14 |
| hemna_ | can we make that happen now please ? :P | 04:14 |
| thingee | hemna_: no, we have to all argue first. | 04:14 |
| *** Longgeek has quit IRC | 04:17 | |
| *** Longgeek has joined #openstack-cinder | 04:17 | |
| *** Longgeek has quit IRC | 04:18 | |
| thingee | hemna_: I recommend telling xyang1 that emc owes you a few cold ones at the next summit. | 04:18 |
| hemna_ | yah I guess we don't feel right about stuff until we argue first. :) | 04:19 |
| xyang1 | hemna_: beer at the summit?:) | 04:21 |
| xyang1 | hemna_: By the way, the dependent patches are bug fixes | 04:21 |
| xyang1 | hemna_: Thanks a lot for spending time reviewing! | 04:22 |
| *** kaufer has quit IRC | 04:25 | |
| anish | thingee: 2x extra_specs errors still there, but all others are mysteriously gone | 04:26 |
| anish | would it make sense to run this on a real machine to avoid any transient issues | 04:26 |
| anish | (currently on a vm) | 04:26 |
| winston-d | hemna_: you can close eyes and hit the mouse button, see which score you can get, be it -1, +1 or +2, once you have that, spend a few mintues on writing the reasoning for that score and done. :D | 04:27 |
| hemna_ | xyang1, looks like one of them has a bunch of refactoring in them as well. | 04:28 |
| hemna_ | lots and lots and lots of code changes to go through. I hope you guys don't do this again in L. | 04:28 |
| *** _cjones_ has joined #openstack-cinder | 04:28 | |
| hemna_ | winston-d, lol | 04:28 |
| xyang1 | hemna_: winston-d good idea:) | 04:29 |
| hemna_ | ah bummer. -2! | 04:29 |
| hemna_ | :P | 04:29 |
| xyang1 | hemna_: We won't submit one big update, but there will still be patches:) | 04:29 |
| xyang1 | hemna_: 700 line each x 10 patches. Still a lot:( | 04:31 |
| hemna_ | :( | 04:33 |
| anish | just remove the zeros | 04:33 |
| winston-d | xyang1: why a bug fix be 7k big? | 04:33 |
| *** _cjones_ has quit IRC | 04:34 | |
| xyang1 | winston-d: 7000 is a driver update. It added support for a newer version of the array | 04:34 |
| winston-d | xyang1: ah, ok | 04:35 |
| xyang1 | winston-d: hemna_ probably 1/3 is docstring | 04:35 |
| *** tsekiyama has quit IRC | 04:38 | |
| *** cchipman has quit IRC | 04:41 | |
| *** coolsvap|afk is now known as coolsvap | 04:42 | |
| hemna_ | phew | 04:43 |
| hemna_ | man | 04:43 |
| *** tsekiyama has joined #openstack-cinder | 04:44 | |
| *** Mandell has joined #openstack-cinder | 04:48 | |
| *** Longgeek has joined #openstack-cinder | 04:50 | |
| *** annashen has quit IRC | 04:52 | |
| xyang1 | winston-d: So if i added volume_gb in consume from volume, i should remove volume size in capacity weigher and filter, right? Line 83 in weights/capacity.py and line 88 in filters/capacity_filter.py | 04:53 |
| *** avishay has joined #openstack-cinder | 04:55 | |
| winston-d | xyang1: no, those are OK, you didn't change the value in host_state. | 04:55 |
| xyang1 | winston-d: Does provisioned cap already contain volume_gb ? | 04:56 |
| xyang1 | winston-d: Am i adding it twice? | 04:57 |
| *** nellysmitt has joined #openstack-cinder | 04:57 | |
| *** anshul has joined #openstack-cinder | 04:59 | |
| xyang1 | winston-d: In host_manager.py, i added volume_gb to provisioned cap. So in filter, when i retrieve provisioned cap from host state, it seems that it already has volume size in it? | 04:59 |
| winston-d | where in host_manager did you add volume_gb to provisioned cap? consume_from_volume? | 05:01 |
| *** nellysmitt has quit IRC | 05:02 | |
| xyang1 | winston-d: Yes | 05:02 |
| winston-d | consume_from_volume() is called *AFTER* filtering/weighing | 05:09 |
| *** bkopilov has joined #openstack-cinder | 05:10 | |
| xyang1 | winston-d: Ok, then I still need volume size in filter and weigher | 05:10 |
| winston-d | filtering/weighing shouldn't touch host_state at all, thingk about those backends can't pass filter and didn't get picked after weighing | 05:10 |
| *** takedakn has joined #openstack-cinder | 05:15 | |
| *** BharatK has joined #openstack-cinder | 05:15 | |
| thingee | hemna_, xyang1: IMO, this is where CI and just doing checks for bad things like touching db comes in. Anything like style should be enforced via hacking/pep. | 05:16 |
| thingee | and how you're interacting with the volume manager layer. | 05:16 |
| xyang1 | thingee: ? | 05:17 |
| thingee | xyang1: wrt to big patches. | 05:17 |
| thingee | for volume drivers | 05:17 |
| thingee | these drivers are already being reported via ci, so that's a good indicator | 05:18 |
| thingee | xyang1: and you're still up :) | 05:18 |
| xyang1 | thingee: Ok, you mean CI works gives reviewer confidence | 05:18 |
| hemna_ | thingee, yah I tend to agree | 05:18 |
| xyang1 | thingee: Ya, will take a break soon | 05:19 |
| thingee | xyang1: yeah. I care about that, db access, interaction with volume manager...and that's about it | 05:19 |
| hemna_ | although it makes me wonder what the reasoning is behind a 7000 line patch | 05:19 |
| hemna_ | :( | 05:19 |
| xyang1 | hemna_: Has to do with array architecture and smis:( | 05:19 |
| xyang1 | hemna_: And lots of docstring | 05:20 |
| thingee | should get jungleboyj to do a hacking check on volume drivers touching the db...though that's happening today | 05:20 |
| hemna_ | thingee, I think that would be a great idea | 05:21 |
| hemna_ | :) | 05:21 |
| winston-d | hemna_: +2 | 05:21 |
| *** nileshb_ has joined #openstack-cinder | 05:22 | |
| avishay | definitely agree | 05:23 |
| *** xyang has quit IRC | 05:23 | |
| thingee | avishay: heeey welcome to the party | 05:23 |
| openstackgerrit | xing-yang proposed openstack/cinder: Support over subscription in thin provisioning https://review.openstack.org/142171 | 05:24 |
| avishay | thingee: weeeee | 05:24 |
| *** takedakn has quit IRC | 05:24 | |
| xyang1 | winston-d: addressed your comments. Let me know if anything else needs to be changed | 05:26 |
| xyang1 | Good night everyone | 05:27 |
| thingee | xyang1: night | 05:27 |
| winston-d | xyang1: will take a look later, i'm now on something else and have to get some lunch after that. | 05:27 |
| winston-d | xyang1: godnite | 05:27 |
| avishay | xyang1: good night | 05:27 |
| *** takedakn has joined #openstack-cinder | 05:30 | |
| *** takedakn has quit IRC | 05:31 | |
| nileshb_ | Hi Cinder reviewers: please help review GPFS Consistency Group support: https://review.openstack.org/#/c/143987/ | 05:34 |
| thingee | nileshb_: we're currently focusing on priorities for k-2: https://etherpad.openstack.org/p/cinder-k2-priorities | 05:34 |
| nileshb_ | sure, thanks | 05:35 |
| thingee | nileshb_: but after next week free game :) | 05:35 |
| nileshb_ | yeah, got it now ... thanks for updating | 05:36 |
| tsekiyama | hemna_: Haven't we agreed that we should abort the operation if multipathd is not running, instead of silently ignore the multipath=True settings by administrators? | 05:36 |
| hemna_ | tsekiyama, so | 05:37 |
| hemna_ | I'm on the fence about that | 05:37 |
| hemna_ | we don't puke today if multipathd isn't running | 05:38 |
| *** annashen has joined #openstack-cinder | 05:38 | |
| openstackgerrit | Jeegn Chen proposed openstack/cinder: EMC VNX Cinder Driver Update https://review.openstack.org/141629 | 05:38 |
| tsekiyama | hemna_: But the operations will fail if cinder is configured to use multipath, doesn't it? | 05:40 |
| hemna_ | actually no | 05:40 |
| hemna_ | today the driver doesn't get a flag to tell it if there is multipathd running | 05:40 |
| hemna_ | so it's up to the driver to decide if it wants to export a volume via multiple paths | 05:41 |
| hemna_ | the brick side, just looks for the first volume to show up and uses it. | 05:41 |
| tsekiyama | Hmm, even if multipath device is not found? | 05:42 |
| *** nileshb_ has quit IRC | 05:42 | |
| hemna_ | so | 05:42 |
| hemna_ | yes | 05:42 |
| hemna_ | the volume entries all show up in /dev/disk/by-path | 05:42 |
| hemna_ | if the same volume shows up under multiple paths, they still all show up there. | 05:43 |
| hemna_ | if multipathd is running, it recognizes and associates the same volume to a multipath device | 05:43 |
| hemna_ | that multipath device is what is used if it exists. | 05:43 |
| hemna_ | if not, the first volume discovered is used. | 05:43 |
| hemna_ | so it all still works | 05:43 |
| hemna_ | but | 05:43 |
| hemna_ | if multipath isn't running | 05:44 |
| hemna_ | and multiple entries show up for the same volume | 05:44 |
| hemna_ | only the first one to shown up is used | 05:44 |
| hemna_ | and use likewise removed at detach time | 05:44 |
| hemna_ | so any others will get left behind | 05:44 |
| hemna_ | and unmanaged | 05:44 |
| *** anshul has quit IRC | 05:44 | |
| tsekiyama | interesting | 05:45 |
| hemna_ | so multipathd not being there or running doesn't prevent the attach/detach process from working | 05:45 |
| hemna_ | so | 05:45 |
| hemna_ | now if someone has configured cinder to use multipath | 05:45 |
| hemna_ | and it's not running | 05:46 |
| hemna_ | what should we do? | 05:46 |
| hemna_ | I think we should log a warning/error and set the flag to False | 05:46 |
| hemna_ | that still allows cinder to continue the process | 05:46 |
| hemna_ | of doing the attach | 05:46 |
| hemna_ | the drivers will see the flag as False, and won't export the volume more than once. | 05:47 |
| *** anshul has joined #openstack-cinder | 05:47 | |
| hemna_ | Nova side might be different | 05:47 |
| tsekiyama | yeah.. | 05:47 |
| hemna_ | because the attaches that nova does and cinder are for different reasons. | 05:47 |
| hemna_ | cinder attaches volumes to do a quick copy volume <---> image operation | 05:47 |
| hemna_ | then detaches and goes away. | 05:47 |
| tsekiyama | but in nova side, it is directly affects reliability/stability | 05:48 |
| hemna_ | nova attaches volumes for end user facing reasons. | 05:48 |
| hemna_ | yup | 05:48 |
| hemna_ | so maybe we need another additional option | 05:48 |
| hemna_ | like | 05:48 |
| hemna_ | enforce_multipath or something like that | 05:48 |
| hemna_ | nova can set that enforcement to true | 05:48 |
| hemna_ | cinder will default it to false | 05:48 |
| tsekiyama | that sounds reasonable to me | 05:49 |
| *** Mandell has quit IRC | 05:49 | |
| *** primechuck has quit IRC | 05:50 | |
| *** rwsu is now known as rwsu-afk | 05:51 | |
| tsekiyama | OK, i will add the parameter to get_connector_properties() to determine it should fall back to single path | 05:52 |
| tsekiyama | and maybe cinder.conf option to specify the parameter | 05:52 |
| *** Mandell has joined #openstack-cinder | 05:56 | |
| *** lpetrut has joined #openstack-cinder | 05:58 | |
| anish | thingee: I was able to pinpoint the issues. for some reason the call to remove_iscsi_target comes in twice, succeeds the first time and fails the seond | 06:01 |
| anish | any pointers as to why this would be happening ? | 06:01 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex https://review.openstack.org/152401 | 06:08 |
| *** oomichi has quit IRC | 06:09 | |
| *** harlowja_at_home has joined #openstack-cinder | 06:10 | |
| *** xyang has joined #openstack-cinder | 06:10 | |
| hemna_ | ok night guys | 06:13 |
| *** annashen has quit IRC | 06:14 | |
| *** Longgeek has quit IRC | 06:16 | |
| *** Longgeek has joined #openstack-cinder | 06:16 | |
| avishay | hemna_: good night | 06:16 |
| *** Longgeek has quit IRC | 06:17 | |
| thingee | hemna_: seeya tomorrow | 06:17 |
| *** hemna_ has quit IRC | 06:18 | |
| *** ho has joined #openstack-cinder | 06:19 | |
| *** annashen has joined #openstack-cinder | 06:19 | |
| *** sweston has quit IRC | 06:21 | |
| *** sweston has joined #openstack-cinder | 06:22 | |
| *** tsekiyama has quit IRC | 06:24 | |
| *** Ilja has joined #openstack-cinder | 06:28 | |
| *** nshaikh has joined #openstack-cinder | 06:29 | |
| *** xyang has quit IRC | 06:30 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Kilo Consistency Group API update https://review.openstack.org/144561 | 06:32 |
| *** rushiagr is now known as rushiagr_away | 06:32 | |
| *** nlevinki has joined #openstack-cinder | 06:32 | |
| *** anshul has quit IRC | 06:34 | |
| *** dulek has joined #openstack-cinder | 06:51 | |
| *** nellysmitt has joined #openstack-cinder | 06:58 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: Add support for chiscsi iscsi helper https://review.openstack.org/147758 | 07:01 |
| openstackgerrit | Tina Tang proposed openstack/cinder: Pool-aware scheduler support in EMC VNX Cinder driver https://review.openstack.org/147408 | 07:01 |
| *** nellysmitt has quit IRC | 07:03 | |
| *** dulek_ has joined #openstack-cinder | 07:08 | |
| *** dulek has quit IRC | 07:11 | |
| *** paul-- has quit IRC | 07:12 | |
| *** harlowja_at_home has quit IRC | 07:14 | |
| *** nkrinner has joined #openstack-cinder | 07:15 | |
| *** paul-- has joined #openstack-cinder | 07:17 | |
| *** annashen has quit IRC | 07:26 | |
| *** takedakn has joined #openstack-cinder | 07:27 | |
| *** takedakn has quit IRC | 07:29 | |
| *** takedakn has joined #openstack-cinder | 07:30 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset snapshot image metadata https://review.openstack.org/149163 | 07:34 |
| *** takedakn has quit IRC | 07:34 | |
| *** tsekiyama has joined #openstack-cinder | 07:35 | |
| *** ronenkat has joined #openstack-cinder | 07:36 | |
| *** tsekiyama has quit IRC | 07:40 | |
| *** nuritv has joined #openstack-cinder | 07:40 | |
| *** anshul has joined #openstack-cinder | 07:43 | |
| openstackgerrit | Li Min Liu proposed openstack/cinder: Make storwize debug log more readable https://review.openstack.org/152358 | 07:43 |
| *** Miouge has joined #openstack-cinder | 07:44 | |
| *** Ilja has quit IRC | 07:52 | |
| *** afazekas has joined #openstack-cinder | 07:52 | |
| *** TobiasE has joined #openstack-cinder | 07:53 | |
| *** chlong has quit IRC | 07:56 | |
| *** ronis has joined #openstack-cinder | 07:57 | |
| *** tshefi has joined #openstack-cinder | 07:58 | |
| *** eharney has joined #openstack-cinder | 08:04 | |
| *** sgotliv has joined #openstack-cinder | 08:12 | |
| *** pradipta_away is now known as pradipta | 08:13 | |
| *** lpetrut has quit IRC | 08:18 | |
| *** dalgaaf has joined #openstack-cinder | 08:18 | |
| *** Miouge has quit IRC | 08:21 | |
| eharney | is there a list of which Cinder drivers have implemented replication? | 08:21 |
| *** Miouge has joined #openstack-cinder | 08:21 | |
| *** Raj_ has joined #openstack-cinder | 08:23 | |
| Raj_ | hey hi | 08:23 |
| Raj_ | guys | 08:23 |
| Raj_ | where i will get the code of latest kilo branch | 08:23 |
| Raj_ | or divers code which support the cinder | 08:23 |
| *** nellysmitt has joined #openstack-cinder | 08:24 | |
| *** wpf1 has quit IRC | 08:25 | |
| *** ronenkat has quit IRC | 08:25 | |
| *** wpf1 has joined #openstack-cinder | 08:26 | |
| *** lpetrut has joined #openstack-cinder | 08:29 | |
| *** Ilja has joined #openstack-cinder | 08:30 | |
| *** ronenkat has joined #openstack-cinder | 08:34 | |
| nikesh_vedams_ | Raj_ https://github.com/openstack/cinder/tree/master/cinder is this you want? | 08:46 |
| Raj_ | nikesh_vedams_ : all the recent code they want to | 08:47 |
| Raj_ | add is here only | 08:48 |
| nikesh_vedams_ | Raj_ : if code is merged in master than you can see here | 08:48 |
| Raj_ | ohk | 08:48 |
| nikesh_vedams_ | master represents current development branch | 08:49 |
| Raj_ | so this is the code which is going to be in kilo branch right ..??? | 08:49 |
| nikesh_vedams_ | which is kilo | 08:49 |
| nikesh_vedams_ | for this release | 08:49 |
| *** ndipanov has joined #openstack-cinder | 08:50 | |
| Raj_ | kilo is the next release of juno right | 08:50 |
| Raj_ | i want to chekc the code for kilo branch only | 08:51 |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset volume image metadata https://review.openstack.org/147726 | 08:52 |
| *** alecv has joined #openstack-cinder | 08:56 | |
| *** jordanP has joined #openstack-cinder | 08:57 | |
| *** lpetrut has quit IRC | 08:58 | |
| *** jistr has joined #openstack-cinder | 08:58 | |
| nikesh_vedams_ | Raj_ : yes | 08:59 |
| *** karimb has joined #openstack-cinder | 08:59 | |
| *** jpich has joined #openstack-cinder | 09:00 | |
| *** ronenkat has quit IRC | 09:01 | |
| *** ronenkat has joined #openstack-cinder | 09:01 | |
| *** mtanino has quit IRC | 09:03 | |
| *** Mandell has quit IRC | 09:14 | |
| openstackgerrit | Bob proposed openstack/cinder: Fix module import method https://review.openstack.org/152116 | 09:14 |
| openstackgerrit | Bob proposed openstack/cinder: Do a check before associate a LUN to a LUN group https://review.openstack.org/148144 | 09:15 |
| openstackgerrit | Bob proposed openstack/cinder: Fix problems under multipath https://review.openstack.org/150767 | 09:15 |
| *** junhongl__ is now known as junhongl | 09:18 | |
| anish | thingee: down to 1 failing test, and 2 that fail with every iscsi_helper, should have the results up by tomorrow, hope that's good enough | 09:20 |
| *** eharney_ has quit IRC | 09:20 | |
| *** ronenkat has quit IRC | 09:26 | |
| *** lpetrut has joined #openstack-cinder | 09:27 | |
| *** Miouge has quit IRC | 09:37 | |
| *** Miouge has joined #openstack-cinder | 09:39 | |
| openstackgerrit | Dave Chen proposed openstack/cinder: set/unset snapshot image metadata https://review.openstack.org/149163 | 09:46 |
| *** e0ne has joined #openstack-cinder | 09:47 | |
| *** ho has quit IRC | 09:53 | |
| openstackgerrit | Chao Zheng Li proposed openstack/cinder: Add support and Unittest of consistency group for XIV/DS8K driver proxy https://review.openstack.org/150281 | 09:59 |
| *** Yada has joined #openstack-cinder | 10:11 | |
| *** e0ne has quit IRC | 10:16 | |
| *** Yada has left #openstack-cinder | 10:22 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: Add support for chiscsi iscsi helper https://review.openstack.org/147758 | 10:27 |
| *** alonmarx_ has joined #openstack-cinder | 10:36 | |
| *** alonmarx has quit IRC | 10:36 | |
| *** Raj_ has quit IRC | 10:48 | |
| openstackgerrit | yatin proposed openstack/python-cinderclient: Added options that can be used while uploading volume to glance https://review.openstack.org/152460 | 10:49 |
| *** nellysmitt has quit IRC | 10:50 | |
| *** aix has joined #openstack-cinder | 10:51 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support https://review.openstack.org/149677 | 11:01 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Switch drivers to use execute from cinder utils https://review.openstack.org/151698 | 11:01 |
| openstackgerrit | Eric Harney proposed openstack/cinder: RemoteFS: Use nas_ip and nas_share_path options https://review.openstack.org/149765 | 11:03 |
| *** 21WAA8C7O has quit IRC | 11:04 | |
| openstackgerrit | Bharat Kumar Kobagana proposed openstack/cinder: Backup driver for GlusterFS storage https://review.openstack.org/152286 | 11:08 |
| *** Ilja has quit IRC | 11:10 | |
| *** e0ne has joined #openstack-cinder | 11:10 | |
| *** Ilja has joined #openstack-cinder | 11:11 | |
| openstackgerrit | yatin proposed openstack/cinder: Added more options while uploading volume to glance https://review.openstack.org/152475 | 11:18 |
| openstackgerrit | Eric Harney proposed openstack/cinder: Tests: Don't sleep for looping calls (eqlx) https://review.openstack.org/152476 | 11:19 |
| *** dalgaaf has quit IRC | 11:20 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support https://review.openstack.org/149677 | 11:20 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Switch drivers to use execute from cinder utils https://review.openstack.org/151698 | 11:20 |
| *** Ilja has quit IRC | 11:21 | |
| *** jistr has quit IRC | 11:21 | |
| *** Ilja has joined #openstack-cinder | 11:24 | |
| *** aix has quit IRC | 11:27 | |
| *** jistr has joined #openstack-cinder | 11:27 | |
| *** aix has joined #openstack-cinder | 11:28 | |
| openstackgerrit | Eric Harney proposed openstack/cinder: Tests: Don't sleep for looping calls (eqlx) https://review.openstack.org/152476 | 11:30 |
| *** BharatK has quit IRC | 11:34 | |
| *** dulek_ is now known as dulek | 11:35 | |
| *** diegows has joined #openstack-cinder | 11:43 | |
| *** ronenkat has joined #openstack-cinder | 11:46 | |
| *** Lee1092 has quit IRC | 11:50 | |
| *** e0ne is now known as e0ne_ | 12:01 | |
| *** e0ne_ is now known as e0ne | 12:01 | |
| *** chlong has joined #openstack-cinder | 12:06 | |
| *** EmilienM|afk is now known as EmilienM | 12:08 | |
| *** boris-42 has quit IRC | 12:22 | |
| *** boris-42 has joined #openstack-cinder | 12:22 | |
| *** aix has quit IRC | 12:24 | |
| *** Longgeek has joined #openstack-cinder | 12:29 | |
| *** tellesnobrega__ has joined #openstack-cinder | 12:31 | |
| *** tellesnobrega_ has quit IRC | 12:32 | |
| *** Longgeek has quit IRC | 12:33 | |
| *** aix has joined #openstack-cinder | 12:37 | |
| *** IanGovett has joined #openstack-cinder | 12:43 | |
| *** markvoelker has quit IRC | 12:45 | |
| *** markvoelker has joined #openstack-cinder | 12:46 | |
| *** lpetrut1 has joined #openstack-cinder | 12:49 | |
| *** markvoelker has quit IRC | 12:50 | |
| *** diegows has quit IRC | 12:50 | |
| *** markvoelker has joined #openstack-cinder | 12:56 | |
| *** lpetrut has quit IRC | 12:57 | |
| *** jungleboyj has quit IRC | 12:57 | |
| *** lan_ has quit IRC | 12:57 | |
| *** kbyrne has quit IRC | 12:57 | |
| *** jungleboyj has joined #openstack-cinder | 12:57 | |
| *** lan_ has joined #openstack-cinder | 12:58 | |
| *** kbyrne has joined #openstack-cinder | 12:58 | |
| *** Yogi1 has joined #openstack-cinder | 13:06 | |
| *** diegows has joined #openstack-cinder | 13:08 | |
| *** Yogi1 has quit IRC | 13:12 | |
| *** Mandell has joined #openstack-cinder | 13:14 | |
| *** bswartz has quit IRC | 13:15 | |
| *** Mandell has quit IRC | 13:19 | |
| *** gtt116__ has joined #openstack-cinder | 13:20 | |
| *** tbarron has joined #openstack-cinder | 13:20 | |
| *** tbarron has quit IRC | 13:20 | |
| *** IanGovett has quit IRC | 13:20 | |
| *** IanGovett has joined #openstack-cinder | 13:21 | |
| *** gtt116_ has quit IRC | 13:23 | |
| *** nshaikh has left #openstack-cinder | 13:24 | |
| *** IanGovett has quit IRC | 13:26 | |
| *** IanGovett has joined #openstack-cinder | 13:27 | |
| *** jasondotstar has joined #openstack-cinder | 13:28 | |
| *** gtt116_ has joined #openstack-cinder | 13:31 | |
| *** bkopilov has quit IRC | 13:32 | |
| *** akerr has joined #openstack-cinder | 13:33 | |
| *** IanGovett has quit IRC | 13:34 | |
| *** gtt116__ has quit IRC | 13:35 | |
| *** Longgeek has joined #openstack-cinder | 13:35 | |
| *** IanGovett has joined #openstack-cinder | 13:36 | |
| *** alonmarx_ has quit IRC | 13:37 | |
| *** alonmarx has joined #openstack-cinder | 13:38 | |
| *** alexpilotti has joined #openstack-cinder | 13:40 | |
| *** alonmarx has quit IRC | 13:41 | |
| openstackgerrit | yatin proposed openstack/python-cinderclient: Added options that can be used while uploading volume to glance https://review.openstack.org/152460 | 13:45 |
| *** mtanino has joined #openstack-cinder | 13:52 | |
| *** alonmarx has joined #openstack-cinder | 13:53 | |
| *** IanGovett1 has joined #openstack-cinder | 13:53 | |
| *** Longgeek has quit IRC | 13:54 | |
| *** bswartz has joined #openstack-cinder | 13:55 | |
| *** IanGovett has quit IRC | 13:55 | |
| *** obutenko has quit IRC | 13:57 | |
| *** mwichmann has joined #openstack-cinder | 14:01 | |
| *** sgotliv has quit IRC | 14:01 | |
| *** tellesnobrega has quit IRC | 14:01 | |
| *** bill_az_ has joined #openstack-cinder | 14:02 | |
| *** annegentle has joined #openstack-cinder | 14:05 | |
| *** jungleboyj has quit IRC | 14:05 | |
| *** dulek has quit IRC | 14:06 | |
| *** dulek has joined #openstack-cinder | 14:06 | |
| *** jasondotstar has quit IRC | 14:08 | |
| *** Yogi1 has joined #openstack-cinder | 14:08 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support https://review.openstack.org/149677 | 14:09 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Switch drivers to use execute from cinder utils https://review.openstack.org/151698 | 14:09 |
| *** sgotliv has joined #openstack-cinder | 14:10 | |
| *** kaufer has joined #openstack-cinder | 14:10 | |
| *** dustins has joined #openstack-cinder | 14:11 | |
| *** mwichmann has quit IRC | 14:11 | |
| *** tbarron has joined #openstack-cinder | 14:13 | |
| *** markvoelker has quit IRC | 14:16 | |
| *** markvoelker has joined #openstack-cinder | 14:16 | |
| *** jasondotstar has joined #openstack-cinder | 14:20 | |
| *** markvoelker has quit IRC | 14:21 | |
| openstackgerrit | Scott DAngelo proposed openstack/python-cinderclient: reset-state should warn that it is DB only https://review.openstack.org/152278 | 14:24 |
| *** mwichmann has joined #openstack-cinder | 14:24 | |
| *** mtanino has quit IRC | 14:25 | |
| *** Xacs has joined #openstack-cinder | 14:28 | |
| *** mriedem has joined #openstack-cinder | 14:29 | |
| Xacs | Hi everyone. I heard about the performance degradation when configuring ceph to use ext4 file system instead of XFS. | 14:30 |
| Xacs | Is it? | 14:30 |
| *** boris-42 has quit IRC | 14:32 | |
| *** eharney has quit IRC | 14:32 | |
| *** primechu_ has joined #openstack-cinder | 14:34 | |
| *** anshul has quit IRC | 14:34 | |
| *** changbl has quit IRC | 14:35 | |
| *** Lee1092 has joined #openstack-cinder | 14:35 | |
| Xacs | In other words, which featured will be missed if I use ext4 instead of XFS in my ceph/cinder deployment? | 14:37 |
| *** obutenko has joined #openstack-cinder | 14:38 | |
| *** mwichmann has quit IRC | 14:40 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Group test files in the right package in cinder.tests https://review.openstack.org/151232 | 14:41 |
| *** IanGovett1 has quit IRC | 14:44 | |
| *** afazekas has quit IRC | 14:45 | |
| *** IanGovett has joined #openstack-cinder | 14:46 | |
| *** dulek has quit IRC | 14:46 | |
| *** dulek has joined #openstack-cinder | 14:47 | |
| *** Mandell has joined #openstack-cinder | 14:47 | |
| *** dulek_ has joined #openstack-cinder | 14:48 | |
| *** dulek has quit IRC | 14:48 | |
| Xacs | Is there any performance comparison between these two file system when configuring to work in ceph? | 14:49 |
| *** rushiagr_away is now known as rushiagr | 14:49 | |
| *** dalgaaf has joined #openstack-cinder | 14:49 | |
| *** Longgeek has joined #openstack-cinder | 14:51 | |
| guitarzan | Xacs: sounds like a great question for #ceph :) | 14:51 |
| *** mwichmann has joined #openstack-cinder | 14:53 | |
| *** jecarey has quit IRC | 14:53 | |
| *** mwichmann has quit IRC | 14:53 | |
| *** jcru has joined #openstack-cinder | 14:53 | |
| Xacs | guitarzan: No one answer there. I asked it before restate it in this channel. | 14:54 |
| guitarzan | Xacs: I'm just giving you a hard time :) Some folks in here probably use ceph | 14:54 |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Group test files in the right package in cinder.tests https://review.openstack.org/151232 | 14:55 |
| avishay | Xacs: ceph uses extended attributes heavily, and I think xfs is more tuned for that | 14:56 |
| *** mwichmann has joined #openstack-cinder | 14:57 | |
| *** nlevinki_ has joined #openstack-cinder | 15:00 | |
| *** markvoelker has joined #openstack-cinder | 15:01 | |
| *** coolsvap is now known as coolsvap|afk | 15:01 | |
| *** zhipeng has joined #openstack-cinder | 15:02 | |
| *** nlevinki has quit IRC | 15:02 | |
| *** dustins_ has joined #openstack-cinder | 15:02 | |
| *** bswartz has quit IRC | 15:03 | |
| *** eharney has joined #openstack-cinder | 15:04 | |
| *** dustins has quit IRC | 15:05 | |
| *** Yogi11 has joined #openstack-cinder | 15:06 | |
| *** bswartz has joined #openstack-cinder | 15:07 | |
| *** ronenkat has quit IRC | 15:08 | |
| openstackgerrit | Jordan Pittier proposed openstack/cinder: Scality SOFS : Use ensure_tree from fileutils https://review.openstack.org/152565 | 15:08 |
| *** Yogi1 has quit IRC | 15:08 | |
| Xacs | avishay: thanks. Configuring ceph to use XFS, yields backward compatibility problems to me. I want to know exactly the penalties when using ext4. | 15:11 |
| *** Longgeek has quit IRC | 15:11 | |
| *** IanGovett has quit IRC | 15:12 | |
| *** Mandell has quit IRC | 15:13 | |
| *** haomaiwang has joined #openstack-cinder | 15:14 | |
| *** IanGovett has joined #openstack-cinder | 15:18 | |
| *** haomaiwang has quit IRC | 15:19 | |
| *** kvidvans has joined #openstack-cinder | 15:19 | |
| *** baffle_ is now known as baffle | 15:20 | |
| avishay | Xacs: i don't know the numbers, sorry | 15:20 |
| *** fischerw has joined #openstack-cinder | 15:20 | |
| *** rushil has joined #openstack-cinder | 15:23 | |
| *** jungleboyj has joined #openstack-cinder | 15:24 | |
| jungleboyj | eharney: I think right now it is just Storwize, XIV and DS8k that have replication working. | 15:24 |
| *** aviramb has joined #openstack-cinder | 15:24 | |
| *** haomaiwang has joined #openstack-cinder | 15:25 | |
| *** jecarey has joined #openstack-cinder | 15:26 | |
| *** xyang has joined #openstack-cinder | 15:26 | |
| *** nellysmitt has joined #openstack-cinder | 15:27 | |
| *** tellesnobrega has joined #openstack-cinder | 15:28 | |
| *** Yogi1 has joined #openstack-cinder | 15:30 | |
| *** mtanino has joined #openstack-cinder | 15:31 | |
| *** Yogi11 has quit IRC | 15:32 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Add rootwrap daemon mode support https://review.openstack.org/149677 | 15:33 |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Switch drivers to use execute from cinder utils https://review.openstack.org/151698 | 15:33 |
| *** bkopilov has joined #openstack-cinder | 15:34 | |
| *** mwichmann has quit IRC | 15:36 | |
| *** e0ne is now known as e0ne_ | 15:37 | |
| *** e0ne_ is now known as e0ne | 15:38 | |
| *** tellesnobrega__ has quit IRC | 15:40 | |
| jordanP | asselin, Hi. Do you have 5min to talk about a third party CI for cinder ? I have a question | 15:41 |
| jordanP | anyway, here the question : | 15:41 |
| *** mwichmann has joined #openstack-cinder | 15:42 | |
| jordanP | is it possible, using the gerrit-jenkins trigger plugin to launch 2 jenkins job but report the result all at once ? We have 2 drivers here at Scality (we don't use Zuul) so we have 2 jobs. According to some doc', it"s not allowed for a single CI jenkins user to post 2 comments on review.openstack.org | 15:43 |
| jordanP | so we need to somehow launch 2 jobs but "aggregate" the results in one comment | 15:44 |
| *** haomaiwang has quit IRC | 15:45 | |
| *** coolsvap|afk is now known as coolsvap | 15:46 | |
| openstackgerrit | Mikhail Khodos proposed openstack/cinder: Fix Nexenta NFS driver mounts https://review.openstack.org/149380 | 15:46 |
| *** rongzez has joined #openstack-cinder | 15:47 | |
| *** rongzez has quit IRC | 15:48 | |
| *** coolsvap is now known as coolsvap_ | 15:50 | |
| *** thangp has joined #openstack-cinder | 15:50 | |
| jordanP | asselin, eikke I think I have my response : http://strongspace.com/rtyler/public/gerrit-jenkins-notes.pdf grep for "Multiple jobs with the same trigger criteria" | 15:51 |
| *** coolsvap_ is now known as coolsvap|afk | 15:52 | |
| *** diegows_ has joined #openstack-cinder | 15:53 | |
| *** coolsvap|afk is now known as swapnilkulkarni | 15:54 | |
| *** vnagpal has joined #openstack-cinder | 15:54 | |
| *** swapnilkulkarni is now known as swapnil | 15:54 | |
| *** swapnil is now known as swapnilk | 15:54 | |
| *** vnagpal has quit IRC | 15:55 | |
| mtanino | anish: hi | 15:56 |
| *** vnagpal_ has joined #openstack-cinder | 15:56 | |
| *** vnagpal_ has quit IRC | 15:56 | |
| *** rmesta has joined #openstack-cinder | 15:56 | |
| *** vnagpal has joined #openstack-cinder | 15:57 | |
| *** diegows has quit IRC | 15:57 | |
| openstackgerrit | Jordan Pittier proposed openstack/cinder: Scality SOFS : Use ensure_tree from fileutils https://review.openstack.org/152565 | 15:57 |
| *** Mandell has joined #openstack-cinder | 15:58 | |
| *** Mandell has quit IRC | 16:00 | |
| openstackgerrit | Navneet Singh proposed openstack/cinder: NetApp eseries implementation for manage/unmanage https://review.openstack.org/138093 | 16:07 |
| *** annegentle has quit IRC | 16:08 | |
| *** dkingshott has joined #openstack-cinder | 16:10 | |
| *** diegows has joined #openstack-cinder | 16:14 | |
| *** remote_morgan_ has joined #openstack-cinder | 16:16 | |
| *** tsekiyama has joined #openstack-cinder | 16:16 | |
| *** bswartz has quit IRC | 16:19 | |
| *** dannywilson has joined #openstack-cinder | 16:19 | |
| *** annegentle has joined #openstack-cinder | 16:19 | |
| *** rhagarty has quit IRC | 16:20 | |
| *** rhagarty has joined #openstack-cinder | 16:20 | |
| Swanson | Anyone have any suggestions for where to look on "volume service is down" errors? | 16:21 |
| Swanson | I've been getting these since I enabled multiple backends. (Not all the time. Just often enough to screw me.) | 16:21 |
| *** severson_ has quit IRC | 16:22 | |
| *** severson_ has joined #openstack-cinder | 16:23 | |
| *** lpabon has joined #openstack-cinder | 16:23 | |
| *** bswartz has joined #openstack-cinder | 16:23 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Support over subscription in thin provisioning https://review.openstack.org/142171 | 16:24 |
| *** xyang has quit IRC | 16:24 | |
| *** severson_ has quit IRC | 16:25 | |
| *** rhagarty has quit IRC | 16:25 | |
| *** asselin_ has quit IRC | 16:25 | |
| *** asselin has quit IRC | 16:25 | |
| *** fischerw has quit IRC | 16:25 | |
| *** rhagarty has joined #openstack-cinder | 16:25 | |
| *** asselin_ has joined #openstack-cinder | 16:25 | |
| *** fischerw has joined #openstack-cinder | 16:25 | |
| *** asselin has joined #openstack-cinder | 16:25 | |
| *** ronis has quit IRC | 16:25 | |
| *** sseverson has joined #openstack-cinder | 16:25 | |
| *** asselin_ has quit IRC | 16:26 | |
| *** asselin has quit IRC | 16:26 | |
| *** rhagarty has quit IRC | 16:26 | |
| *** ronenkat has joined #openstack-cinder | 16:26 | |
| *** sseverson has quit IRC | 16:26 | |
| *** asselin_ has joined #openstack-cinder | 16:26 | |
| *** asselin has joined #openstack-cinder | 16:26 | |
| *** rhagarty has joined #openstack-cinder | 16:26 | |
| *** sseverson has joined #openstack-cinder | 16:26 | |
| *** cinderella has joined #openstack-cinder | 16:27 | |
| cinderella | hi | 16:27 |
| *** rhagarty has quit IRC | 16:27 | |
| *** asselin_ has quit IRC | 16:27 | |
| *** asselin has quit IRC | 16:27 | |
| cinderella | can someone help me? I'm having problems with cinder | 16:27 |
| *** fischerw has quit IRC | 16:27 | |
| cinderella | [17:22] <cinderella> I'm able to create volumes up to 1 gb of space [17:22] <cinderella> but when I exceed this 1 GB, I get errors [17:22] <cinderella> it's not a free space problem | 16:27 |
| *** sseverson has quit IRC | 16:27 | |
| *** asselin_ has joined #openstack-cinder | 16:27 | |
| *** rhagarty has joined #openstack-cinder | 16:27 | |
| *** asselin has joined #openstack-cinder | 16:27 | |
| *** fischerw has joined #openstack-cinder | 16:27 | |
| *** sseverson has joined #openstack-cinder | 16:27 | |
| *** rhagarty has quit IRC | 16:28 | |
| *** fischerw has quit IRC | 16:28 | |
| *** sseverson has quit IRC | 16:28 | |
| *** asselin has quit IRC | 16:28 | |
| *** asselin_ has quit IRC | 16:28 | |
| *** asselin_ has joined #openstack-cinder | 16:28 | |
| *** fischerw has joined #openstack-cinder | 16:28 | |
| *** asselin has joined #openstack-cinder | 16:28 | |
| *** rhagarty has joined #openstack-cinder | 16:28 | |
| *** sseverson has joined #openstack-cinder | 16:28 | |
| *** vnagpal has quit IRC | 16:29 | |
| *** gary-smith has joined #openstack-cinder | 16:29 | |
| *** fischerw has quit IRC | 16:29 | |
| *** fischerw has joined #openstack-cinder | 16:30 | |
| *** fischerw has quit IRC | 16:30 | |
| *** sseverson has quit IRC | 16:30 | |
| *** fischerw has joined #openstack-cinder | 16:30 | |
| *** sseverson has joined #openstack-cinder | 16:30 | |
| avishay | cinderella: more info? which driver? what command are you running exactly? | 16:31 |
| openstackgerrit | xing-yang proposed openstack/cinder: Fix detach volume from host problem in VMAX driver https://review.openstack.org/143891 | 16:31 |
| *** fischerw has quit IRC | 16:31 | |
| openstackgerrit | Eric Harney proposed openstack/cinder: RemoteFS: Use nas_ip and nas_share_path options https://review.openstack.org/149765 | 16:31 |
| *** fischerw has joined #openstack-cinder | 16:32 | |
| cinderella | avishay: I'm working with horizon dashboard | 16:32 |
| *** fischerw has quit IRC | 16:32 | |
| cinderella | in the logs I can see the error: No valid host found | 16:32 |
| *** sseverson has quit IRC | 16:33 | |
| *** fischerw has joined #openstack-cinder | 16:33 | |
| avishay | cinderella: why do you say it's not a free space issue? | 16:33 |
| cinderella | avishay: I'm working on my server, I'm pretty sure I have >900GB free | 16:33 |
| *** rushil has quit IRC | 16:34 | |
| avishay | cinderella: how did you set it up? devstack? | 16:34 |
| cinderella | avishay: followed this guide: https://github.com/fornyx/OpenStack-Havana-Install-Guide/blob/master/OpenStack-Havana-Install-Guide.rst | 16:34 |
| avishay | cinderella: oh what a bad guide :) | 16:35 |
| avishay | cinderella: it told you to create a 2GB loopback device for your storage | 16:36 |
| avishay | cinderella: that's why... | 16:36 |
| cinderella | yes avishay, now I created one of 10 GB | 16:36 |
| *** diemt has joined #openstack-cinder | 16:36 | |
| cinderella | (deleted all the old ones) | 16:36 |
| avishay | cinderella: with the same name? | 16:36 |
| cinderella | deleted the old cinder-volumes with vgremove | 16:37 |
| cinderella | and created a new one, mounted on loop5, of 10 GB | 16:37 |
| cinderella | now I'm able to instantiate bigger volumes :) | 16:37 |
| cinderella | but can't attach them to instances | 16:37 |
| cinderella | avishay: can you clarify me how all of this works? I'm a bit confused | 16:37 |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX driver Kilo update https://review.openstack.org/141729 | 16:39 |
| *** AlkaD has joined #openstack-cinder | 16:39 | |
| cinderella | avishay: is that guide bad? why? :) | 16:40 |
| avishay | cinderella: it's not bad per se, it just didn't really explain what was going on - this is not a "real" setup | 16:42 |
| avishay | cinderella: basically you are using an LVM backend, which will put the storage on local drives | 16:42 |
| avishay | cinderella: instead of using a full drive, it creates a "fake" one using a 2GB file | 16:42 |
| avishay | cinderella: if you have a spare partition, you can use that instead | 16:43 |
| cinderella | avishay: if I want to use a NAS, how does that work? | 16:43 |
| avishay | cinderella: which NAS? | 16:43 |
| cinderella | avishay: I don't know the exact model :( but I've heard of "buffalo" | 16:44 |
| cinderella | I'm banging my head with cinder because soon I should configure the Openstack on this NAS | 16:45 |
| *** changbl has joined #openstack-cinder | 16:45 | |
| avishay | cinderella: as far as i know, buffalo isn't supported | 16:46 |
| avishay | cinderella: (there is no driver) | 16:46 |
| Swanson | avishay: is there a generic smb driver? Buffalo is consumer level so I assume it coughs up smb shares. | 16:47 |
| cinderella | we should be able to use it via iscsi, but not so sure about that - we're pretty new in this | 16:47 |
| *** diegows_ has quit IRC | 16:48 | |
| *** diegows has quit IRC | 16:49 | |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX driver Kilo update https://review.openstack.org/141729 | 16:50 |
| *** annegentle has quit IRC | 16:51 | |
| avishay | Swanson: cinderella: ah, there is some windows SMB driver, not familiar with it though | 16:51 |
| cinderella | avishay: If I want to associate a real partition to Cinder, what should I do? | 16:51 |
| cinderella | (instead of a file alike I did till now) | 16:51 |
| *** AlkaD has quit IRC | 16:52 | |
| avishay | cinderella: in the guide, skip the part with 'dd', and in the next part use the partition instead of /dev/loop2 | 16:52 |
| avishay | cinderella: just make sure it has no important data on it! | 16:53 |
| cinderella | it's ok avishay, thank you! | 16:53 |
| cinderella | what about using volumes with instances? | 16:53 |
| avishay | cinderella: what do you mean? | 16:54 |
| cinderella | for example, to install a generic linux iso as I would do with a "real" pc | 16:54 |
| cinderella | avishay: do you get what I mean? I know my english isn't very good :) | 16:56 |
| *** Longgeek has joined #openstack-cinder | 16:57 | |
| cinderella | avishay: wanna give a volume to an instance such that that VM has an hard drive to work on | 16:58 |
| cinderella | for example, to install a linux distro | 16:58 |
| *** xyang has joined #openstack-cinder | 16:58 | |
| avishay | cinderella: so you can boot a VM, attach (use nova volume-attach), and then install onto the secondary device | 16:59 |
| *** bswartz has quit IRC | 16:59 | |
| xyang | thingee: ping | 16:59 |
| cinderella | avishay: fine, thanks. volumes survive till the partition is live, right? because, as you can see in that guide, it says that "the volume group is lost after a reboot" | 17:01 |
| *** boris-42 has joined #openstack-cinder | 17:02 | |
| *** Longgeek has quit IRC | 17:03 | |
| *** TobiasE has quit IRC | 17:05 | |
| *** changbl has quit IRC | 17:06 | |
| *** jpich has quit IRC | 17:08 | |
| *** annegentle has joined #openstack-cinder | 17:09 | |
| *** eharney has quit IRC | 17:11 | |
| *** lpabon has quit IRC | 17:11 | |
| openstackgerrit | Abel Lopez proposed openstack/cinder: Purge deleted rows https://review.openstack.org/146766 | 17:12 |
| cinderella | avishay: another question and I won't stress you anymore :) what is the exact difference between cinder and swift? | 17:12 |
| avishay | cinderella: cinder manages block storage (does not actually store data and is not on the data path) | 17:13 |
| avishay | cinderella: swift implements object storage (GET/PUT/DELETE) interface | 17:13 |
| *** diegows has joined #openstack-cinder | 17:14 | |
| *** nkrinner has quit IRC | 17:14 | |
| *** rwsu-afk is now known as rwsu | 17:15 | |
| *** dustins has joined #openstack-cinder | 17:16 | |
| cinderella | avishay: very gentle | 17:16 |
| *** annashen has joined #openstack-cinder | 17:16 | |
| cinderella | thanks :) | 17:16 |
| *** zhipeng_ has joined #openstack-cinder | 17:17 | |
| *** ronenkat_ has joined #openstack-cinder | 17:18 | |
| *** dustins_ has quit IRC | 17:18 | |
| *** rushil has joined #openstack-cinder | 17:19 | |
| *** ronenkat has quit IRC | 17:19 | |
| *** zhipeng has quit IRC | 17:20 | |
| *** markstur has joined #openstack-cinder | 17:21 | |
| *** markstur has left #openstack-cinder | 17:21 | |
| *** _cjones_ has joined #openstack-cinder | 17:21 | |
| *** sgotliv has quit IRC | 17:23 | |
| avishay | cinderella: good luck! | 17:25 |
| *** lpetrut1 has quit IRC | 17:25 | |
| *** n1 has joined #openstack-cinder | 17:25 | |
| n1 | hi cinder gurus ... this openstack cinder cookbook has a bug in multibackend mode : https://github.com/stackforge/cookbook-openstack-block-storage/blob/dad180e458b41c60e8f06f2ffa04ee56fdcced3b/recipes/volume.rb | 17:26 |
| n1 | rbd_secret_uuid in each of the sections is getting generated as nil | 17:26 |
| n1 | is there any workaround or fix for this ? | 17:26 |
| *** dulek_ has quit IRC | 17:28 | |
| *** dulek_ has joined #openstack-cinder | 17:28 | |
| openstackgerrit | yatin proposed openstack/cinder: Added more options while uploading volume to glance https://review.openstack.org/152475 | 17:28 |
| cinderella | avishay: when I attach, I'm required to insert a Device Name (for example, /dev/vdc). What should I insert there | 17:30 |
| cinderella | ? | 17:30 |
| *** annegentle has quit IRC | 17:33 | |
| avishay | cinderella: /dev/vdc is fine | 17:33 |
| avishay | cinderella: it's a hint of what the device in the guest should be | 17:34 |
| cinderella | avishay: it is the mount point that I'll see in VM? | 17:34 |
| avishay | cinderella: not guaranteed, but it tries | 17:34 |
| *** leeantho has joined #openstack-cinder | 17:34 | |
| cinderella | avishay: on horizon, I see "attaching" for a few minutes, and then it simply does not attach | 17:35 |
| avishay | cinderella: i'm very sorry, but i am currently busy with something else right now - perhaps somebody else can help you | 17:36 |
| cinderella | avishay: no problem, thanks anyway :) | 17:36 |
| avishay | cinderella: sure, good luck | 17:37 |
| *** Mandell has joined #openstack-cinder | 17:37 | |
| *** diegows has quit IRC | 17:37 | |
| *** ronis has joined #openstack-cinder | 17:38 | |
| *** jistr has quit IRC | 17:38 | |
| *** vilobhmm has joined #openstack-cinder | 17:40 | |
| *** jordanP has quit IRC | 17:46 | |
| *** tshefi has quit IRC | 17:47 | |
| *** nlevinki_ has quit IRC | 17:48 | |
| *** krtaylor has quit IRC | 17:48 | |
| *** ronenkat__ has joined #openstack-cinder | 17:53 | |
| *** karimb has quit IRC | 17:54 | |
| *** krtaylor has joined #openstack-cinder | 17:54 | |
| openstackgerrit | Seiji Aguchi proposed openstack/cinder: Add manage/unmanage methods for Hitachi Block Storage Driver https://review.openstack.org/141315 | 17:54 |
| *** dulek_ has quit IRC | 17:54 | |
| *** dulek has joined #openstack-cinder | 17:55 | |
| *** ronenkat_ has quit IRC | 17:56 | |
| *** sseverson has joined #openstack-cinder | 17:57 | |
| *** akerr has quit IRC | 17:58 | |
| *** nellysmitt has quit IRC | 17:58 | |
| *** akerr has joined #openstack-cinder | 17:58 | |
| *** lpetrut has joined #openstack-cinder | 18:01 | |
| *** dustins_ has joined #openstack-cinder | 18:01 | |
| *** tbarron has quit IRC | 18:02 | |
| *** changbl has joined #openstack-cinder | 18:04 | |
| *** dustins has quit IRC | 18:04 | |
| *** rushil has quit IRC | 18:08 | |
| *** bill_az_ has quit IRC | 18:09 | |
| thingee | xyang: hi | 18:11 |
| *** tbarron has joined #openstack-cinder | 18:12 | |
| xyang | thingee: hi, I'm in a meeting now. will ping you a little later | 18:12 |
| *** harlowja has joined #openstack-cinder | 18:15 | |
| openstackgerrit | Julie Gravel proposed openstack/cinder-specs: Validate volume type extra specs https://review.openstack.org/131280 | 18:18 |
| *** Mandell has quit IRC | 18:21 | |
| *** cinderella has quit IRC | 18:21 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Allow archiving deleted rows to shadow tables, for performance https://review.openstack.org/131182 | 18:24 |
| xyang | thingee: still there? | 18:25 |
| xyang | thingee: just noticed that things are moved out of K-2 | 18:25 |
| *** patrickeast has joined #openstack-cinder | 18:25 | |
| thingee | xyang: yeah, traction on reviews isn't great right now | 18:25 |
| xyang | thingee: ok | 18:26 |
| xyang | thingee: Can we break the CG update into two? Modify CG is ready | 18:26 |
| *** tbarron1 has joined #openstack-cinder | 18:27 | |
| *** Mandell has joined #openstack-cinder | 18:28 | |
| *** n1 has quit IRC | 18:30 | |
| *** avishay has quit IRC | 18:30 | |
| *** tbarron has quit IRC | 18:30 | |
| Swanson | "model server went away" ringing any bells? Looks like there are two RPC servers being created for service cinder-volume. (One for each backend I presume.) Then I get that message. | 18:34 |
| *** dustins_ has quit IRC | 18:37 | |
| hemna | :( | 18:38 |
| hemna | I did a bunch last night | 18:38 |
| *** jecarey has quit IRC | 18:39 | |
| *** pradipta is now known as pradipta_away | 18:40 | |
| xyang | hemna: that's right | 18:40 |
| hemna | xyang, your vmax update is failing jenkins fwiw | 18:40 |
| openstackgerrit | nikeshmahalka proposed openstack/cinder: Add iSCSI SCST Target support to cinder https://review.openstack.org/145778 | 18:40 |
| xyang | hemna: I'm fixing it now. some extra space, etc | 18:40 |
| nikesh_vedams_ | thingee jdg: commited new patch no 12 for "https://review.openstack.org/145778". | 18:41 |
| thingee | nikesh_vedams_: I moved scst to k-3. no time is left right now, sorry. | 18:42 |
| xyang | hemna: if the dependent patch updated, do you just do a "git review -d <parent change id>", and then "git checkout <your branch>", then "git rebase -i <the parent review branch>"? | 18:43 |
| thingee | nikesh_vedams_: we can't move forward on this until we remove this idea of walking through all volumes in the database. | 18:43 |
| nikesh_vedams_ | thingee jdg: addressed previous reviews comments and also removed database concept | 18:43 |
| thingee | but I think jdg has some ideas with that, if they weren't already posted in the review | 18:43 |
| nikesh_vedams_ | thingee :removed database dependency | 18:43 |
| *** dustins has joined #openstack-cinder | 18:44 | |
| nikesh_vedams_ | now we are using scst commands to get LUNs | 18:44 |
| nikesh_vedams_ | so no walking through all volumes in the database. | 18:44 |
| thingee | nikesh_vedams_: this is still not right. A target driver should not inherit from volume driver classes. | 18:45 |
| *** xyang has quit IRC | 18:46 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Fix detach volume from host problem in VMAX driver https://review.openstack.org/143891 | 18:46 |
| *** Longgeek has joined #openstack-cinder | 18:48 | |
| nikesh_vedams_ | thingee: By using SCST commands to get LUNs info,have to use grouping mechanism. Under that group we have to mention which LUNs are exposed to which initiator. In this way we need to have initiator iqn. So we are inheriting volume driver class to our SCST target driver class. | 18:50 |
| thingee | nikesh_vedams_: Again this is not right. | 18:51 |
| nikesh_vedams_ | Apart from this way we are unable to get initiator name. | 18:51 |
| thingee | nikesh_vedams_: I'll revisit it in k-3 | 18:51 |
| *** delatte has joined #openstack-cinder | 18:51 | |
| *** jecarey has joined #openstack-cinder | 18:51 | |
| thingee | a target driver is a *target*. If you're working around that, something is not right. | 18:51 |
| thingee | if you're turning the target class into a volume driver, that is not right | 18:51 |
| thingee | I can't make it anymore clear than that | 18:52 |
| openstackgerrit | Tom Barron proposed openstack/cinder: Refactor Swift backup driver and introduce chunking driver https://review.openstack.org/149725 | 18:53 |
| openstackgerrit | xing-yang proposed openstack/cinder: EMC VMAX driver Kilo update https://review.openstack.org/141729 | 18:54 |
| *** delattec has quit IRC | 18:55 | |
| erlon | thingee: hi, Mike | 18:56 |
| erlon | thingee: I asked about the inclusion of our bp on k2 on the ml | 18:57 |
| erlon | https://www.mail-archive.com/openstack-dev@lists.openstack.org/msg44848.html | 18:57 |
| *** rushiagr is now known as rushiagr_away | 18:57 | |
| *** ronenkat__ has quit IRC | 18:57 | |
| erlon | thingee: just to refresh, is there any reason for that blueprints not be included on k2? | 18:57 |
| *** xyang has joined #openstack-cinder | 18:59 | |
| *** ronenkat has joined #openstack-cinder | 18:59 | |
| thingee | erlon: not enough people reviewing. It's in k-3 | 18:59 |
| *** aix has quit IRC | 18:59 | |
| *** delatte has quit IRC | 19:00 | |
| openstackgerrit | Tom Barron proposed openstack/cinder: NFS backup driver https://review.openstack.org/149726 | 19:00 |
| anish | mtanino: you pinged ? | 19:00 |
| *** dalgaaf has quit IRC | 19:00 | |
| *** vilobhmm has quit IRC | 19:01 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: Add support for chiscsi iscsi helper https://review.openstack.org/147758 | 19:02 |
| *** e0ne is now known as e0ne_ | 19:02 | |
| mtanino | anish: Ah, pinged, but I commented on the gerrit | 19:03 |
| erlon | thingee: hmm, Ok, is it possible to get at least the bug target? I need to backport it and dont know if after k3 will have time left for that | 19:03 |
| erlon | thingee: https://review.openstack.org/141868 | 19:03 |
| anish | mtanino: yes, just updated the commit message with link to certification | 19:03 |
| erlon | thingee: it is blocked only by the unit tests which, by the way where merged already | 19:04 |
| *** Mandell has quit IRC | 19:04 | |
| mtanino | anish: :) | 19:04 |
| *** changbl has quit IRC | 19:05 | |
| thingee | erlon: not sure why that review is topic of a bp/* | 19:05 |
| thingee | instead of bug/* | 19:05 |
| thingee | nit pick | 19:06 |
| *** vilobhmm has joined #openstack-cinder | 19:06 | |
| *** e0ne_ is now known as e0ne | 19:06 | |
| erlon | thingee: this last one is a bug, the ssh feature could be a bug, and the pool aware certainly is a bp | 19:06 |
| *** e0ne has quit IRC | 19:07 | |
| thingee | erlon: in the future, if you say closes-bug in the commit message, it's a bug. If you're closing a bug in a blueprint, it shouldn't be a bug | 19:07 |
| thingee | erlon: is there any update on the HDS CI? | 19:09 |
| thingee | erlon: steve quinn responded to me back in 1/15 for both hitachi and hds | 19:09 |
| erlon | thingee: yes, it is woking in silent mode, but still not 100% stable | 19:09 |
| thingee | erlon: excellent, glad to hear it! :) | 19:10 |
| *** Longgeek has quit IRC | 19:10 | |
| erlon | thingee: we where about to put it in check, but as marcusvrn went in holydays AND the deadline was extended :) we are trying to make it faster and stable | 19:11 |
| thingee | erlon: haha, ok | 19:11 |
| thingee | erlon: your bug is targeted. | 19:12 |
| *** swapnilk is now known as coolsvap_ | 19:15 | |
| *** coolsvap_ is now known as coolsvap | 19:19 | |
| *** jungleboyj has quit IRC | 19:25 | |
| *** jungleboyj has joined #openstack-cinder | 19:25 | |
| *** Guest81568 is now known as annegentle | 19:29 | |
| xyang | thingee: I've made CG update two entries in the etherpad and moved ModifyCG to the K3 section. Create CG from snapshot is in the Not Ready section. | 19:30 |
| *** jasondotstar has quit IRC | 19:30 | |
| thingee | xyang: this is just for k-2 | 19:31 |
| thingee | the etherpad that is | 19:32 |
| ^rcaskey | so...been reading up on cinder's architecture. Cartoon understanding: drives run on the cinder server, connectors run on the compute nodes? | 19:32 |
| xyang | thingee: ok | 19:32 |
| *** changbl has joined #openstack-cinder | 19:34 | |
| ^rcaskey | err drivers, typo there, srry | 19:34 |
| thingee | ^rcaskey: compute node vms are the initiators. cinder is keeping track in this multi-tenant envrionment who owns what volumes in all the backends that could exist. In addition cinder is providing connection information (iscsi, fibrechannel) to the initiator | 19:36 |
| thingee | ^rcaskey: after that information is handed off, it's completely just vm talking to the storage backend. | 19:37 |
| thingee | ^rcaskey: cinder isn't involved in the data path | 19:37 |
| *** kvidvans has quit IRC | 19:39 | |
| thingee | the point is to provide this storage (it might be vendor a and/or vendor b) from a single interface. the end user doesn't care who it is, just provide the storage, and maybe have some qos requirements, etc. | 19:39 |
| ^rcaskey | yup, but the thinking that has to get done on the clients end happens in the connector specified by the scheduler? | 19:40 |
| *** Lee1092 has quit IRC | 19:40 | |
| *** coolsvap is now known as coolsvap_ | 19:41 | |
| *** ronenkat has quit IRC | 19:41 | |
| thingee | ^rcaskey: the connector is up to the volume driver. For example, it's lvm running iscsi, it's going to give back iscsi connector information. it's then up to nova to interpret that information to connect to that volume. | 19:42 |
| thingee | however, it is possible to have the scheduler filter backends based on what protocol they use. | 19:42 |
| ^rcaskey | is there anthing that runs....elsewhere....for say doing some kind of lowish level map-reducish thinking about volumes | 19:46 |
| *** bswartz has joined #openstack-cinder | 19:46 | |
| ^rcaskey | dunno what that would be...some kind of terrible dedupe middleware or something | 19:47 |
| *** nellysmitt has joined #openstack-cinder | 19:48 | |
| *** e0ne has joined #openstack-cinder | 19:49 | |
| *** e0ne is now known as e0ne_ | 19:49 | |
| erlon | thingee: thanks, how about the BPs? | 19:50 |
| *** e0ne_ is now known as e0ne | 19:50 | |
| thingee | erlon: the one that was in k-2 moved to k-3. let me know any others that are missed | 19:50 |
| thingee | ^rcaskey: map reduce for what? | 19:50 |
| thingee | heh makes me think of that lil john turn down for what song. Map reduce for WHAT?! | 19:51 |
| ^rcaskey | thingee, i dunno, I just am trying to make sure 100% of the backend-specific thinking is happening either in the initiator's connector or in the driver on the storage node | 19:51 |
| ^rcaskey | thingee, so I can filter out all the 1s from my block device in a distributed manner :p | 19:51 |
| thingee | ^rcaskey: hate to disappoint, but we use mysql with where clauses. nothing fancy here :) | 19:52 |
| anish | thingee: passing all certs now, updated the cert results as asked, for whenever you have time | 19:52 |
| thingee | maybe an index on the project_id column | 19:52 |
| erlon | thingee: only these 2: hds-hnas-pool-aware-sched and hds-hnas-ssh-backend | 19:52 |
| thingee | anish: I saw that thanks | 19:53 |
| thingee | anish: yeah the code is pretty straight forward to me. I'll get back to it in a minute, but I think I'm fine with it | 19:53 |
| anish | excellent ! | 19:53 |
| thingee | anish: I hope you found tempest tests useful too | 19:53 |
| anish | yes ! did need a fix | 19:54 |
| thingee | since these are suppose to be actual ways openstack users will use your target driver. | 19:54 |
| anish | makes sense, there was one test that seemed to be making calls simultaneously, which I hadn't been able to quite reproduce locally | 19:55 |
| anish | + figured out how to run them individually btw | 19:55 |
| anish | it's just ./tools/pretty_tox.sh <test_name> | 19:55 |
| anish | thingee: I did have one question. Should iscsi_helpers set data["vendor_name"] ? | 19:58 |
| thingee | anish: no, because that's pushed up from the volume driver to the scheduler. | 19:58 |
| anish | aah, gotcha | 19:59 |
| *** shakamunyi has joined #openstack-cinder | 20:02 | |
| *** chlong has quit IRC | 20:05 | |
| thingee | erlon: bps targeted | 20:07 |
| erlon | thingee: great, thanks! | 20:07 |
| *** theanalyst has quit IRC | 20:07 | |
| thingee | k-3 is already at 21 bps...might need to rethink some things here. | 20:09 |
| thingee | I'm guessing only 14 bps will be doable for k-2. | 20:09 |
| thingee | anyways, bbl | 20:10 |
| scottda | Alright, I'm a bit confused. | 20:10 |
| scottda | in cinder/api/contrib/admin_actions.py I see _force_detach | 20:10 |
| scottda | which calls into the driver volume_api.terminate_connection and volume_api.detach | 20:10 |
| *** theanalyst has joined #openstack-cinder | 20:11 | |
| scottda | Shouldn't this be sufficient for any rollback or cleanup from within the code, like timeout in initialize_connection? | 20:11 |
| scottda | It just lacks a call to the DB to set the state to 'available' | 20:11 |
| *** boris-42 has quit IRC | 20:12 | |
| scottda | And it could be called from python-cinderclient for a 'force-detach' command, along with DB cleanup if the driver calls succeed. | 20:12 |
| *** tellesnobrega_ has joined #openstack-cinder | 20:15 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/cinder: Allow archiving deleted rows to shadow tables, for performance https://review.openstack.org/131182 | 20:17 |
| *** afazekas has joined #openstack-cinder | 20:21 | |
| *** afazekas is now known as afazekas_drunk | 20:21 | |
| *** sgotliv has joined #openstack-cinder | 20:23 | |
| *** shakamunyi has quit IRC | 20:23 | |
| *** alecv has quit IRC | 20:25 | |
| *** jasondotstar has joined #openstack-cinder | 20:25 | |
| *** Ilja has quit IRC | 20:30 | |
| *** anteaya has quit IRC | 20:34 | |
| *** Mandell has joined #openstack-cinder | 20:36 | |
| *** MentalRay_ has joined #openstack-cinder | 20:38 | |
| *** gary-smith has quit IRC | 20:39 | |
| *** gary-smith has joined #openstack-cinder | 20:40 | |
| *** Mandell has quit IRC | 20:43 | |
| *** afazekas_drunk has quit IRC | 20:44 | |
| *** asselin_ has quit IRC | 20:46 | |
| *** asselin has quit IRC | 20:46 | |
| *** rhagarty has quit IRC | 20:46 | |
| *** MIDENN_ has quit IRC | 20:46 | |
| *** kmartin has quit IRC | 20:46 | |
| *** fischerw has quit IRC | 20:46 | |
| *** asselin_ has joined #openstack-cinder | 20:46 | |
| *** fischerw has joined #openstack-cinder | 20:46 | |
| *** asselin has joined #openstack-cinder | 20:46 | |
| *** rhagarty has joined #openstack-cinder | 20:46 | |
| *** kmartin has joined #openstack-cinder | 20:46 | |
| *** mdenny has joined #openstack-cinder | 20:46 | |
| *** gary-smith has quit IRC | 20:47 | |
| *** gary-smith has joined #openstack-cinder | 20:48 | |
| *** Yogi1 has quit IRC | 20:49 | |
| *** openstackgerrit has quit IRC | 20:50 | |
| *** jasondotstar has quit IRC | 20:50 | |
| *** openstackgerrit has joined #openstack-cinder | 20:51 | |
| *** annashen has quit IRC | 20:52 | |
| *** afazekas has joined #openstack-cinder | 20:53 | |
| *** anteaya has joined #openstack-cinder | 20:53 | |
| *** tellesnobrega_ has quit IRC | 20:54 | |
| *** Longgeek has joined #openstack-cinder | 20:56 | |
| *** afazekas is now known as fazekas_drunk | 20:57 | |
| *** fazekas_drunk is now known as afazekas_drunk | 20:57 | |
| *** tellesnobrega_ has joined #openstack-cinder | 20:59 | |
| *** tellesnobrega_ has quit IRC | 20:59 | |
| *** Longgeek has quit IRC | 21:00 | |
| *** jcru has quit IRC | 21:01 | |
| *** annegent_ has joined #openstack-cinder | 21:02 | |
| *** diegows has joined #openstack-cinder | 21:02 | |
| *** sgotliv has quit IRC | 21:03 | |
| openstackgerrit | abhiram moturi proposed openstack/cinder: Enhance unit tests for zfssa drivers https://review.openstack.org/152676 | 21:04 |
| *** Xacs has quit IRC | 21:04 | |
| *** Miouge has quit IRC | 21:05 | |
| *** nellysmitt has quit IRC | 21:09 | |
| *** annegent_ has quit IRC | 21:18 | |
| rhe00_ | install_slave.sh is trying to pull the install_puppet.sh script from https://git.openstack.org/cgit/openstack-infra/config/plain/install_puppet.sh | 21:20 |
| rhe00_ | sorry wrong channel | 21:20 |
| *** Yogi1 has joined #openstack-cinder | 21:22 | |
| *** annashen has joined #openstack-cinder | 21:27 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: Change oslo.utils to oslo_utils https://review.openstack.org/152687 | 21:27 |
| *** jseiler__ has quit IRC | 21:29 | |
| openstackgerrit | Tom Barron proposed openstack/cinder: NFS backup driver https://review.openstack.org/149726 | 21:35 |
| openstackgerrit | Tom Barron proposed openstack/cinder: Refactor Swift backup driver and introduce chunking driver https://review.openstack.org/149725 | 21:35 |
| *** jseiler has joined #openstack-cinder | 21:35 | |
| thingee | rhe00_: happy to see that accidental message :) | 21:35 |
| openstackgerrit | Tomoki Sekiyama proposed openstack/cinder: Enhance iSCSI multipath support https://review.openstack.org/134681 | 21:36 |
| *** obutenko has quit IRC | 21:39 | |
| *** aarefiev has quit IRC | 21:39 | |
| *** obutenko has joined #openstack-cinder | 21:41 | |
| *** aarefiev has joined #openstack-cinder | 21:42 | |
| *** obutenko has quit IRC | 21:42 | |
| *** rushil has joined #openstack-cinder | 21:44 | |
| *** jnrao has joined #openstack-cinder | 21:46 | |
| openstackgerrit | Patrick East proposed openstack/cinder: Make PureISCSIDriver iSCSI port discovery more flexible https://review.openstack.org/152693 | 21:51 |
| *** ronis has quit IRC | 21:51 | |
| openstackgerrit | Anish Bhatt proposed openstack/cinder: Change oslo.* to oslo_* https://review.openstack.org/152687 | 21:53 |
| *** jungleboyj has quit IRC | 21:54 | |
| *** lpetrut has quit IRC | 21:54 | |
| *** tbarron1 has quit IRC | 21:56 | |
| patrickeast | anyone have a good example of unit tests that are testing a method that has the @retry decorator on it they could point me to? should i just mock the decorator? | 21:56 |
| *** afazekas_drunk is now known as afeakas|oyua | 22:00 | |
| *** e0ne has quit IRC | 22:00 | |
| *** akerr has quit IRC | 22:03 | |
| openstackgerrit | Tomoki Sekiyama proposed openstack/cinder: Enhance iSCSI multipath support https://review.openstack.org/134681 | 22:04 |
| *** Mandell has joined #openstack-cinder | 22:06 | |
| thingee | patrickeast: the tests themselves for @retry https://github.com/openstack/cinder/blob/master/cinder/tests/test_utils.py#L1397 | 22:06 |
| anish | hemna: I just had a iscsiadm related patch land in nova, should I make one for brick as well ? or is that automated ? | 22:06 |
| thingee | anish: it is not automated. He's just trying to keep up with the changes by hand. | 22:07 |
| *** arif-ali has joined #openstack-cinder | 22:09 | |
| patrickeast | thingee: thanks, so the rety part of it is taken care of | 22:10 |
| *** e0ne has joined #openstack-cinder | 22:10 | |
| patrickeast | if i understand correctly when some unit test that calls the method triggers the retry it would actually sleep and retry, is there any easy to disable that or fake the sleep time on it? | 22:10 |
| *** bswartz has quit IRC | 22:11 | |
| patrickeast | for example i test specifically for the behavior when it throws an exception (the same one it would retry on) | 22:11 |
| anish | thingee: hmm, any suggestions how to go about it ? I was thinking I'll rebase my patch onto brick, but it seems a few commits behind | 22:11 |
| patrickeast | hmm maybe i’ll just change my test around | 22:12 |
| patrickeast | easier that way ;) | 22:12 |
| *** jaypipes has quit IRC | 22:15 | |
| openstackgerrit | abhiram moturi proposed openstack/cinder: Enhance unit tests for zfssa drivers https://review.openstack.org/152676 | 22:15 |
| thingee | hemna: ^ | 22:17 |
| *** Yogi1 has quit IRC | 22:17 | |
| thingee | anish: I defer to hemna for the rebasing part | 22:18 |
| *** jasondotstar has joined #openstack-cinder | 22:20 | |
| anish | ko ! | 22:21 |
| *** e0ne has quit IRC | 22:22 | |
| *** harlowja is now known as harlowja_away | 22:28 | |
| *** thangp has quit IRC | 22:31 | |
| *** Mandell has quit IRC | 22:45 | |
| *** jasondotstar has quit IRC | 22:46 | |
| *** tellesnobrega has quit IRC | 22:47 | |
| *** Mandell has joined #openstack-cinder | 22:48 | |
| *** kaufer has quit IRC | 22:48 | |
| *** tellesnobrega has joined #openstack-cinder | 22:49 | |
| *** jungleboyj has joined #openstack-cinder | 22:54 | |
| *** dulek has quit IRC | 22:59 | |
| openstackgerrit | Rajini Ram proposed openstack/cinder: Fixes the EQL driver CI tests AttributeError https://review.openstack.org/152705 | 23:00 |
| openstackgerrit | Patrick East proposed openstack/cinder: Make PureISCSIDriver iSCSI port discovery more flexible https://review.openstack.org/152693 | 23:01 |
| *** EmilienM is now known as EmilienM|afk | 23:03 | |
| *** rushil has quit IRC | 23:06 | |
| *** changbl has quit IRC | 23:06 | |
| *** jaypipes has joined #openstack-cinder | 23:07 | |
| *** briancurtin has quit IRC | 23:08 | |
| *** zhiyan has quit IRC | 23:08 | |
| *** ameade has quit IRC | 23:09 | |
| *** Mandell has quit IRC | 23:10 | |
| *** nellysmitt has joined #openstack-cinder | 23:10 | |
| *** MentalRay_ has quit IRC | 23:10 | |
| *** Mandell has joined #openstack-cinder | 23:10 | |
| *** eharney has joined #openstack-cinder | 23:12 | |
| arif-ali | Hi all, I am setting up gpfs driver, I'm able to create the volume in GPFS, but when I assign the volume to an instance, I get the error as shown in the following paste http://paste.openstack.org/show/166461/, any ideas on how to fix this would be great | 23:12 |
| *** Mandell has quit IRC | 23:14 | |
| *** nellysmitt has quit IRC | 23:14 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/152091 | 23:15 |
| *** jasondotstar has joined #openstack-cinder | 23:15 | |
| hemna | back | 23:15 |
| hemna | anish, nothing is automated | 23:15 |
| hemna | anish, just go ahead and push the change up in gerrit for brick | 23:16 |
| hemna | I'll manage pulling it into the brick lib. | 23:16 |
| anish | hemna: only my change ? | 23:20 |
| anish | some of the functions definitions don't match up if I do that directly | 23:20 |
| *** zhiyan has joined #openstack-cinder | 23:21 | |
| hemna | I'm not sure I follow you | 23:21 |
| *** ameade has joined #openstack-cinder | 23:21 | |
| hemna | the code between nova's libvirt and brick is slightly different | 23:22 |
| hemna | but should match in capability | 23:22 |
| hemna | you'll have to figure out that piece of brick's initiator | 23:22 |
| hemna | what nova patch is yours that you are referencing ? | 23:22 |
| anish | this https://review.openstack.org/#/c/146233/ | 23:22 |
| *** Mandell has joined #openstack-cinder | 23:22 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-cinderclient: Updated from global requirements https://review.openstack.org/149975 | 23:22 |
| hemna | ah that one | 23:23 |
| hemna | ok | 23:23 |
| *** fischerw has quit IRC | 23:23 | |
| anish | so I can rebase this onto brick easey-peasey | 23:23 |
| hemna | so yah, you should submit a brick patch for that one. | 23:23 |
| hemna | but you should still have a cinder-spec and bp for it as well. | 23:23 |
| *** Mandell has quit IRC | 23:24 | |
| anish | but it will miss some functionality, for eg brick does not seem to have discovery authentication yet | 23:24 |
| anish | (I might just be reading the code wrong) | 23:24 |
| *** Mandell has joined #openstack-cinder | 23:24 | |
| hemna | wouldn't surprise me that there was stuff added in libvirt that didn't make it into brick just yet | 23:24 |
| hemna | we'll have to deal with that | 23:24 |
| anish | hemna: can do. I presume this would not be bound by k2/k3 deadlines yes ? | 23:24 |
| hemna | so, we might want to wait until brick is standalone | 23:24 |
| hemna | just file the BP and we can figure out where to target it | 23:25 |
| hemna | still needs a cinder-spec as well | 23:25 |
| anish | can do | 23:25 |
| hemna | if other things are missing in brick's initiator code, then we'll have to deal with that first | 23:25 |
| *** zhipeng_ has quit IRC | 23:26 | |
| anish | nothing that affects my patch directly, I just had to rebase on discovery-auth support, hence I asked | 23:26 |
| anish | hemna: unrelated, any plans for isns support/mutual auth in brick ? | 23:27 |
| *** diemt has quit IRC | 23:27 | |
| hemna | anish, currently nothing on the plan | 23:27 |
| *** triggerz has quit IRC | 23:28 | |
| anish | hemna: specs/kilo is okay ? since there seems to be no future-work directory | 23:28 |
| *** Mandell has quit IRC | 23:29 | |
| *** triggerz has joined #openstack-cinder | 23:29 | |
| *** Mandell has joined #openstack-cinder | 23:29 | |
| *** Mandell has quit IRC | 23:30 | |
| *** Mandell has joined #openstack-cinder | 23:31 | |
| hemna | anish, sure that's fine. | 23:32 |
| *** harlowja_away is now known as harlowja | 23:32 | |
| *** briancurtin has joined #openstack-cinder | 23:33 | |
| openstackgerrit | Eric Harney proposed openstack/cinder: NFS snapshots https://review.openstack.org/147186 | 23:34 |
| *** enterprisedc has quit IRC | 23:41 | |
| *** enterprisedc has joined #openstack-cinder | 23:42 | |
| *** jasondotstar has quit IRC | 23:45 | |
| thingee | hemna: so I punted the vmax and vnx stuff to k-3. | 23:45 |
| thingee | the only thing I have you for is oversubscription thin provisioning https://review.openstack.org/#/c/142171/ | 23:46 |
| thingee | along with jdg | 23:46 |
| thingee | if we're not able to sign off on it early tomorrow, it'll have to be punted. | 23:47 |
| hemna | thingee, I saw that earlier today when the etherpad got rearranged :) | 23:49 |
| thingee | anybody else noticing this tempest issue when running devstack? http://paste.openstack.org/show/166476/ | 23:49 |
| *** annashen has quit IRC | 23:52 | |
| *** IanGovett has quit IRC | 23:52 | |
| openstackgerrit | Jay Bryant proposed openstack/cinder-specs: Fix the volume replication spec https://review.openstack.org/152735 | 23:55 |
| *** dustins has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!