| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Share migration Newton improvements https://review.openstack.org/328431 | 00:04 |
|---|---|---|
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Add share type change to Share Migration https://review.openstack.org/345142 | 00:04 |
| *** yangyapeng has quit IRC | 00:11 | |
| openstackgerrit | Jay Mehta proposed openstack/manila: HPE 3PAR driver pool support https://review.openstack.org/329552 | 00:20 |
| *** Suyi has quit IRC | 00:22 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Manila install guide: Fix post-install steps https://review.openstack.org/359491 | 00:22 |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Manila install guide: Fix post-install steps https://review.openstack.org/359491 | 00:25 |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Manila install guide: Fix post-install steps https://review.openstack.org/359491 | 00:26 |
| *** tpsilva has quit IRC | 00:28 | |
| *** qeelee has joined #openstack-manila | 00:29 | |
| *** dustins has quit IRC | 00:30 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Manila install guide: Fix post-install steps https://review.openstack.org/359491 | 00:41 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/python-manilaclient: Update Share Migration CLI https://review.openstack.org/328437 | 00:45 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/python-manilaclient: Add share type change to Share Migration CLI https://review.openstack.org/345125 | 00:45 |
| *** kaisers_ has joined #openstack-manila | 00:53 | |
| *** kaisers_ has quit IRC | 00:58 | |
| *** JoseMello has quit IRC | 01:10 | |
| *** mtanin___ has quit IRC | 01:12 | |
| openstackgerrit | Cedric Zhuang proposed openstack/manila: Add documentation for EMC Unity Driver for Manila https://review.openstack.org/358576 | 01:19 |
| *** cknight has joined #openstack-manila | 01:19 | |
| *** chlong has joined #openstack-manila | 01:20 | |
| *** cknight1 has joined #openstack-manila | 01:22 | |
| *** cknight has quit IRC | 01:24 | |
| *** yangyapeng has joined #openstack-manila | 01:28 | |
| *** wangqun has joined #openstack-manila | 01:45 | |
| zhongjun_ | gouthamr: ping | 02:10 |
| gouthamr | zhongjun_: pong. | 02:23 |
| *** gouthamr_ has joined #openstack-manila | 02:25 | |
| *** gouthamr has quit IRC | 02:28 | |
| *** kaisers_ has joined #openstack-manila | 02:42 | |
| openstackgerrit | Yingzhe Zeng proposed openstack/manila: Implement replication support in huawei driver https://review.openstack.org/331586 | 02:43 |
| *** kaisers_ has quit IRC | 02:46 | |
| zhongjun_ | gouthamr: In https://review.openstack.org/#/c/359316/ | 02:49 |
| zhongjun_ | gouthamr: Does Netapp can pass the ipv6 tempest test? | 02:51 |
| gouthamr_ | zhongjun_: looks like it works for the generic driver | 02:52 |
| gouthamr_ | zhongjun_: just ran the job on the CI.. teh NetApp CI waits for Jenkins to vote +1 before kicking off tests.. | 02:52 |
| gouthamr_ | zhongjun_: ignore my comments on the generic driver.. looks like those tests were skipped.. | 02:53 |
| gouthamr_ | zhongjun_: but it looks like the tests passed on the ZFS driver.. http://logs.openstack.org/16/359316/2/check/gate-manila-tempest-dsvm-postgres-zfsonlinux/d2bb73f/console.html | 02:55 |
| gouthamr_ | zhongjun_: some access rule related failures though. worth digging in.. | 02:55 |
| zhongjun_ | gouthamr: some access rule related failures, because I change access_type ipv6 to ip | 02:58 |
| *** yangyape_ has joined #openstack-manila | 02:59 | |
| gouthamr_ | zhongjun_: maybe remove cifs from the list... some of the errors are because of that | 03:00 |
| *** yangyapeng has quit IRC | 03:02 | |
| gouthamr_ | zhongjun_: also take a look at http://logs.openstack.org/16/359316/2/check/gate-manila-tempest-minimal-dsvm-lvm/b896df6/logs/screen-m-shr.txt.gz?level=TRACE#_2016-08-23_18_23_59_328 | 03:02 |
| zhongjun_ | gouthamr_: Many driver CI not run, so I could not see all of feedback | 03:02 |
| gouthamr_ | zhongjun_: that's the log from teh LVM driver | 03:03 |
| gouthamr_ | zhongjun_: we're using netaddr to validate the range ... and the validation is failing | 03:03 |
| *** gouthamr_ is now known as gouthamr | 03:05 | |
| zhongjun_ | gouthamr: The validation is failing, Because I passed all of the ip. | 03:08 |
| *** mtanino has joined #openstack-manila | 03:08 | |
| zhongjun_ | gouthamr: https://review.openstack.org/#/c/359316/2/manila/api/v1/shares.py | 03:08 |
| gouthamr | zhongjun_: oh | 03:09 |
| gouthamr | zhongjun_: those are "expected" failures then :) | 03:09 |
| zhongjun_ | gouthamr: yes | 03:09 |
| gouthamr | np.. this is a good test patch.. we can still get some insights.. | 03:09 |
| *** yangyapeng has joined #openstack-manila | 03:12 | |
| *** yangyape_ has quit IRC | 03:12 | |
| *** yangyape_ has joined #openstack-manila | 03:13 | |
| *** yangyapeng has quit IRC | 03:13 | |
| openstackgerrit | Peter Wang proposed openstack/manila: Use job for NFS share creation https://review.openstack.org/359567 | 03:20 |
| openstackgerrit | Peter Wang proposed openstack/manila: Use job for NFS share creation https://review.openstack.org/359567 | 03:22 |
| *** gouthamr has quit IRC | 03:36 | |
| *** gouthamr has joined #openstack-manila | 03:37 | |
| *** ganso has quit IRC | 03:41 | |
| *** gouthamr has quit IRC | 03:45 | |
| *** cknight1 has quit IRC | 03:48 | |
| *** cknight has joined #openstack-manila | 03:49 | |
| openstackgerrit | zhongjun proposed openstack/manila: [DNM] Test IP(ipv4 and ipv6) rules with all drivers https://review.openstack.org/359316 | 03:52 |
| *** shausy has joined #openstack-manila | 04:29 | |
| *** kaisers_ has joined #openstack-manila | 04:31 | |
| openstackgerrit | Peter Wang proposed openstack/manila: VNX: Use job for NFS share creation https://review.openstack.org/359567 | 04:32 |
| *** kaisers_ has quit IRC | 04:35 | |
| *** cknight has quit IRC | 04:39 | |
| *** mtanino has quit IRC | 04:42 | |
| markstur | bswartz, et. al. I'm on jury duty this week. Will miss the weekly meeting and be away from IRC. | 04:55 |
| *** markstur is now known as markstur_jury_du | 04:56 | |
| *** lpetrut has joined #openstack-manila | 05:20 | |
| *** yangyape_ has quit IRC | 05:32 | |
| *** nkrinner_afk is now known as nkrinner | 05:43 | |
| *** yangyapeng has joined #openstack-manila | 05:45 | |
| *** sandanar has joined #openstack-manila | 06:02 | |
| openstackgerrit | gecong proposed openstack/manila: Correct reraising of exception https://review.openstack.org/359612 | 06:18 |
| *** kaisers_ has joined #openstack-manila | 06:19 | |
| *** lpetrut has quit IRC | 06:20 | |
| *** aovchinnikov has joined #openstack-manila | 06:22 | |
| *** lpetrut has joined #openstack-manila | 06:22 | |
| *** kaisers_ has quit IRC | 06:24 | |
| *** FL1SK has quit IRC | 06:27 | |
| *** lpetrut has quit IRC | 06:33 | |
| *** rraja has joined #openstack-manila | 06:38 | |
| *** lpetrut has joined #openstack-manila | 06:46 | |
| *** kaisers_ has joined #openstack-manila | 06:47 | |
| *** tovchinnikova has joined #openstack-manila | 06:53 | |
| *** lpetrut has quit IRC | 07:03 | |
| *** nkrinner has quit IRC | 07:21 | |
| *** nkrinner has joined #openstack-manila | 07:25 | |
| openstackgerrit | gecong proposed openstack/manila: Correct reraising of exception https://review.openstack.org/359612 | 07:27 |
| *** nkrinner has quit IRC | 07:30 | |
| *** nkrinner has joined #openstack-manila | 07:43 | |
| openstackgerrit | zhongjun proposed openstack/manila: [DNM] Test IP(ipv4 and ipv6) rules with all drivers https://review.openstack.org/359316 | 07:49 |
| *** FL1SK has joined #openstack-manila | 08:26 | |
| *** aovchinnikov has quit IRC | 08:27 | |
| *** pcaruana has joined #openstack-manila | 08:29 | |
| *** aovchinnikov has joined #openstack-manila | 09:32 | |
| *** lpetrut has joined #openstack-manila | 09:59 | |
| *** tovchinnikova has quit IRC | 10:13 | |
| *** lpetrut1 has joined #openstack-manila | 10:14 | |
| *** lpetrut has quit IRC | 10:15 | |
| *** lpetrut1 is now known as lpetrut | 10:15 | |
| *** yangyapeng has quit IRC | 10:19 | |
| *** kaisers_ has quit IRC | 10:30 | |
| *** wangqun has quit IRC | 10:34 | |
| *** shausy has quit IRC | 10:45 | |
| *** shausy has joined #openstack-manila | 10:45 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: [DNM] Tests for scenario tests https://review.openstack.org/340236 | 10:56 |
| *** JoseMello has joined #openstack-manila | 11:05 | |
| openstackgerrit | Ramana Raja proposed openstack/manila: glusterfs: handle new cli XML format https://review.openstack.org/352167 | 11:09 |
| *** alyson_ has joined #openstack-manila | 11:15 | |
| *** timcl has joined #openstack-manila | 11:18 | |
| *** tpsilva has joined #openstack-manila | 11:34 | |
| *** yangyapeng has joined #openstack-manila | 11:47 | |
| *** dustins has joined #openstack-manila | 12:02 | |
| *** xyang1 has joined #openstack-manila | 12:14 | |
| *** kaisers_ has joined #openstack-manila | 12:18 | |
| *** kaisers_ has quit IRC | 12:23 | |
| *** gouthamr has joined #openstack-manila | 12:26 | |
| *** xyang1 has quit IRC | 12:34 | |
| *** xyang1 has joined #openstack-manila | 12:36 | |
| *** porrua has joined #openstack-manila | 12:37 | |
| *** ganso has joined #openstack-manila | 12:40 | |
| ganso | xyang1, gouthamr, markstur_jury_du, bswartz, tbarron: Hello folks, could you please a look at the scenario job fix? :) #link https://review.openstack.org/#/c/359166 | 12:41 |
| gouthamr | ganso: how do we know 50.2.5.x isn't part of a tenant network allocation? | 12:53 |
| bswartz | ganso: where does the 50.2.5.0/24 range come from? | 12:53 |
| bswartz | you can't just make up subnet ranges | 12:53 |
| ganso | bswartz: it can be overwritten from local.conf | 12:54 |
| ganso | bswartz: FIXED_RANGE is defined in there, ADMIN_NET_RANGE is not | 12:55 |
| ganso | gouthamr: in CI it is not part of any existing tenant | 12:58 |
| gouthamr | ganso: for the CI specifically, setting the opt in contrib/ci/pre_test_hook.sh makes sense | 12:59 |
| gouthamr | just like $FIXED_RANGE, if not specified, you can't use that setting.. rather than have a network range for a normal devstack | 13:00 |
| tbarron | oh, you guys are asking just hat I asked on the review :) | 13:01 |
| tbarron | ganso: thanks for figuring out what was going on, btw | 13:02 |
| ganso | gouthamr: ok, I will move and export it in pre_test_hook, and add conditions to not the config if not defined | 13:04 |
| ganso | gouthamr: not *do the config | 13:04 |
| gouthamr | ganso: +1 thanks... | 13:04 |
| ganso | gouthamr: thanks for the feedback | 13:05 |
| ganso | gouthamr: forgot that plugin.sh is also used by other installations, not only CI | 13:05 |
| gouthamr | ganso: np.. running tempest against the new optimized migration fixes; how brave do you feel this morning? :D | 13:06 |
| ganso | gouthamr: lol | 13:06 |
| ganso | gouthamr: speaking of which | 13:06 |
| ganso | gouthamr: I think I will have to kidnap Valeriy's dummy driver migration support patch | 13:07 |
| gouthamr | * with DHSS=True... something vponomaryov didn't add in a "real" driver. :P | 13:07 |
| *** eharney has joined #openstack-manila | 13:07 | |
| gouthamr | ganso: nice.. i think the ZFS patch needs an update too.. i guess bswartz was going to help with that | 13:08 |
| gouthamr | ganso: quick q | 13:15 |
| gouthamr | ganso: didn't understand your answer for "why can't tempest migrate between two pools" | 13:16 |
| *** akerr has joined #openstack-manila | 13:17 | |
| ganso | gouthamr: testing with dummy driver had 4 backends, 2 DHSS=True and 2 DHSS=False | 13:19 |
| ganso | gouthamr: tempest isn't smart enough to choose the one that has a compatible share type | 13:19 |
| ganso | gouthamr: like, it creates the share originally with DHSS=True, so, for that share type, it can only be migrated to another DHSS=True pool | 13:20 |
| ganso | gouthamr: and it doesn't know how to choose, there is no API that exposes the pool capabilities to tempest | 13:20 |
| ganso | gouthamr: so I used the etc/tempest.conf [share] backend_names config option to determine which tempest can choose | 13:21 |
| gouthamr | ganso: self.shares_client.list_pools()['pools'] is teh API to expose pool capabilities | 13:21 |
| ganso | gouthamr: humm if I am not mistaken, it doesn't show capabilities | 13:22 |
| gouthamr | ganso: it does, use 'details=True' :) | 13:22 |
| ganso | gouthamr: I probably did not try that | 13:22 |
| ganso | gouthamr: ok I can refactor that, and then it won't need the enabled_backends option anymore | 13:23 |
| gouthamr | ganso: https://github.com/openstack/manila/blob/master/manila_tempest_tests/tests/api/base.py#L602 <-- example | 13:23 |
| bswartz | ganso: why not 10.2.5.0/24 ? | 13:23 |
| ganso | bswartz: I don't know, why 10.2.5.0/24 ? for CI I believe it doesn't really matter | 13:24 |
| ganso | bswartz: I'll move from plugins.sh to pre_test_hool.sh btw | 13:25 |
| bswartz | ganso: it matters | 13:25 |
| bswartz | ganso: 50.2.5.0/24 is routable, 10.2.5.0/24 is not | 13:25 |
| *** openstackgerrit has quit IRC | 13:26 | |
| bswartz | ganso: read RFC 1918 if you don't know the difference | 13:26 |
| ganso | bswartz: Yea I just google on that. Addresses that are supposed to be reached from the internet | 13:27 |
| *** openstackgerrit has joined #openstack-manila | 13:27 | |
| ganso | bswartz: I will include that in the upcoming patch | 13:28 |
| *** rraja has quit IRC | 13:33 | |
| *** eharney has quit IRC | 13:38 | |
| *** qeelee has quit IRC | 13:38 | |
| *** eharney has joined #openstack-manila | 13:43 | |
| *** rraja has joined #openstack-manila | 13:44 | |
| *** rraja has quit IRC | 13:45 | |
| *** sandanar has quit IRC | 13:45 | |
| *** rraja has joined #openstack-manila | 13:45 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: [DNM] Tests for scenario tests https://review.openstack.org/340236 | 13:50 |
| *** aovchinnikov has quit IRC | 13:52 | |
| *** kaisers_ has joined #openstack-manila | 14:07 | |
| *** kaisers_ has quit IRC | 14:11 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Fix connectivity problem in Scenario job https://review.openstack.org/359166 | 14:22 |
| *** kaisers_ has joined #openstack-manila | 14:32 | |
| *** dsariel has joined #openstack-manila | 14:36 | |
| *** eharney has quit IRC | 14:36 | |
| *** shausy has quit IRC | 14:40 | |
| *** rraja has quit IRC | 14:44 | |
| *** cknight has joined #openstack-manila | 14:48 | |
| *** shausy has joined #openstack-manila | 14:51 | |
| *** dustins has quit IRC | 14:51 | |
| *** kaisers_ has quit IRC | 14:52 | |
| *** eharney has joined #openstack-manila | 14:55 | |
| *** rraja has joined #openstack-manila | 14:56 | |
| *** rraja has quit IRC | 14:57 | |
| *** vbellur has quit IRC | 15:02 | |
| *** kaisers_ has joined #openstack-manila | 15:08 | |
| *** nkrinner is now known as nkrinner_afk | 15:16 | |
| *** david-lyle_ has joined #openstack-manila | 15:25 | |
| *** david-lyle_ has quit IRC | 15:26 | |
| *** vbellur has joined #openstack-manila | 15:28 | |
| *** shausy has quit IRC | 15:29 | |
| *** dsariel has quit IRC | 15:35 | |
| openstackgerrit | zhongjun proposed openstack/manila: [DNM] Test IP(ipv4 and ipv6) rules with all drivers https://review.openstack.org/359316 | 15:41 |
| *** dsariel has joined #openstack-manila | 15:47 | |
| *** Suyi has joined #openstack-manila | 16:32 | |
| tpsilva | cknight, tbarron, xyang1: hey folks! could you please take a look on this hitachi driver patch? I think it's ready for merge for quite a while: https://review.openstack.org/#/c/356049/ | 16:33 |
| tpsilva | thank you! | 16:33 |
| *** kaisers_ has quit IRC | 16:34 | |
| cknight | tpsilva: Looks good, thanks for doing this. | 16:39 |
| bswartz | ganso: I'm testing your patch now, and I put a comment on it | 16:49 |
| bswartz | ganso: can you explain how data_node_access_ip works? | 16:49 |
| ganso | bswartz: which patch? | 16:49 |
| bswartz | https://review.openstack.org/359166 | 16:49 |
| *** mtanino has joined #openstack-manila | 16:50 | |
| tpsilva | cknight: thank you | 16:50 |
| ganso | bswartz: data_node_access_ip is the IP of the data node in the admin network, which should be given access to so shares can be mounted in the data node | 16:50 |
| tbarron | tpsilva: should you trigger a recheck on hitachi-hsp ? | 16:50 |
| bswartz | ganso: your patch sets its value to a subnet though | 16:51 |
| ganso | bswartz: yes, since I am not employing any mechanism to detect the node IP, I am using the subnet, as I was already doing before | 16:51 |
| bswartz | but in a devstack-gate environment, what IP does it actually use? | 16:51 |
| ganso | bswartz: the subnet range specified is able to be used both for the driver admin network and the data node | 16:52 |
| bswartz | ganso: does something create an IP in that subnet for the manila node to use? | 16:52 |
| bswartz | I just don't understand how the packets are flowing | 16:52 |
| tpsilva | tbarron: this patch is regarding hitachi hnas | 16:52 |
| ganso | bswartz: the IP is generated at run time | 16:53 |
| tpsilva | tbarron: doesn't touch hsp driver code | 16:53 |
| ganso | bswartz: so the node running the test has many interfaces | 16:53 |
| tpsilva | tbarron: but I see it failed... will check the hsp CI | 16:53 |
| ganso | bswartz: it has eth0, eth1, etc... and it has a tap interface which is assigned an IP in the admin_net | 16:53 |
| tpsilva | tbarron: hnas CI is ok | 16:53 |
| bswartz | ganso: can you point to the code that creates the tap interface? | 16:53 |
| ganso | bswartz: we do not know which IP that tap interface will have, the generic driver will create it | 16:53 |
| tbarron | tpsilva: yeah, that's what I thought, but the optics :) | 16:53 |
| ganso | bswartz: yes, just a sec | 16:54 |
| tpsilva | tbarron: but hey, thanks... I'll check what's going on with that CI | 16:54 |
| bswartz | ganso: the tap interface that the generic driver creates would be the server IP, no the client IP | 16:54 |
| bswartz | ganso: access rule should specify the client IP or IP range | 16:54 |
| ganso | bswartz: https://github.com/openstack/manila/blob/master/manila/share/drivers/service_instance.py#L892 | 16:54 |
| tbarron | tpsilva: I'll go on and +2, the code looks fine. Dunno if you need that other CI passing for +W, will let the cores with more experience here decide | 16:55 |
| bswartz | ganso: thanks, but as I said above that's an IP of the NFS server not the client | 16:55 |
| ganso | bswartz: yes, there is another one, I am looking for it | 16:56 |
| tpsilva | tbarron: thank you | 16:56 |
| tbarron | tpsilva: bswartz woops, pressed +W by mistake, but I think it's fine :) | 16:56 |
| ganso | bswartz: there are 2 IPs, one the share server will have (the one I linked you) and the one in the host | 16:56 |
| bswartz | tbarron: >_< | 16:56 |
| bswartz | tbarron: which patch | 16:56 |
| tbarron | tpsilva: if your other driver breaks for some reason I can't fathom b/c of this patch, it's on you | 16:56 |
| cknight | tbarron: I think it's fine, I asked them to move this driver when they added the HSP one. | 16:56 |
| tbarron | everybocy else is passing | 16:56 |
| tpsilva | tbarron, bswartz: it's fine... think it's an isolated failure of that CI | 16:57 |
| tbarron | cknight: +1 | 16:57 |
| tbarron | bswartz: https://review.openstack.org/#/c/356049/ | 16:57 |
| bswartz | tbarron: looks fine it got 2 +2s | 16:58 |
| ganso | bswartz: https://github.com/openstack/manila/blob/master/manila/share/drivers/service_instance.py#L1013 | 16:58 |
| ganso | bswartz: that one goes to the host | 16:58 |
| ganso | so the share server and the host have IPs, they can reach each other | 16:58 |
| ganso | bswartz: the host is actually the data node itself | 16:58 |
| bswartz | tbarron: everyone makes that mistake eventually -- if you do it on a patch that really shouldn't merge you know you can throw it out of the gate by removing the +W and pushing a new patchset right? | 16:58 |
| ganso | bswartz: so when it is invoked "allow_access" with the host IP, the host can mount it | 16:59 |
| ganso | bswartz: since it is very hard to determine at run time, for CI only, we use the subnet | 16:59 |
| tbarron | bswartz: yeah, that was one of valeriy's pro tips when i was inducted, but thanks for the reminder | 16:59 |
| bswartz | tbarron: it seemed timely | 16:59 |
| tbarron | :) | 17:00 |
| bswartz | ganso: running some experiments with that | 17:00 |
| ganso | bswartz: experiments? | 17:01 |
| bswartz | ganso: I want to avoid hacking core stuff to work around weird generic driver limitations | 17:04 |
| bswartz | so I'm trying to figure out if we can make generic more like everything else in this case | 17:05 |
| ganso | bswartz: actually, after we introduced "admin networks" to drivers, it made most of them consistent | 17:06 |
| ganso | bswartz: I remember talking to cknight about this back in Mitaka | 17:06 |
| ganso | bswartz: for any share that has an export in the admin network, you would just use the data node IP address instead of range, then it works | 17:07 |
| bswartz | ganso: yes but it seems that, weirdly, if you use generic driver you'll get one of these IP per share server | 17:08 |
| ganso | bswartz: the one in L1013 no | 17:08 |
| ganso | bswartz: unless, it is in different share networks | 17:09 |
| cknight | bswartz, ganso: My main objection in the mitaka code was that if I had an access rule for the same address as the data node address, the rule was deleted at the end of the migration. It's probably not a common use case, but it's annoying in a dev environment. | 17:09 |
| ganso | cknight: I may have fixed that. I don't remember | 17:09 |
| ganso | cknight: I remember the problem, but I don't remember if I fixed it | 17:09 |
| cknight | ganso: You didn't want to fix it in Mitaka. I haven't tried since then. | 17:09 |
| ganso | cknight: yes, but I may have fixed later | 17:10 |
| *** JoseMello has quit IRC | 17:10 | |
| *** dsariel has quit IRC | 17:17 | |
| *** aovchinnikov has joined #openstack-manila | 17:27 | |
| openstackgerrit | Merged openstack/manila: add access_key to share_access_map https://review.openstack.org/343306 | 17:32 |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Fix connectivity problem in Scenario job https://review.openstack.org/359166 | 17:39 |
| ganso | bswartz: ^ | 17:44 |
| ganso | bswartz: btw, I believe we are going to miss manila-ui FF for share migration | 17:52 |
| *** dustins has joined #openstack-manila | 17:52 | |
| *** timcl has quit IRC | 17:54 | |
| *** timcl has joined #openstack-manila | 17:54 | |
| gouthamr | bswartz: can you take a look and +1 this removal patch? https://review.openstack.org/#/c/359235/ | 17:58 |
| gouthamr | bswartz: we merged the install guide in-tree and I have the last few fixes here: https://review.openstack.org/#/c/359491/ | 17:58 |
| bswartz | ganso: what do we need to merge? | 17:58 |
| ganso | bswartz: for manila-ui? 3 big patches | 17:59 |
| ganso | bswartz: + manila-ui's | 17:59 |
| bswartz | ganso: the UI patches depend on the manila patches? | 17:59 |
| *** lpetrut has quit IRC | 17:59 | |
| * bswartz sighs | 17:59 | |
| bswartz | that's a disappointment | 17:59 |
| ganso | bswartz: they don't have depends-on in the commit message, but to work, they do | 18:00 |
| *** vbellur has quit IRC | 18:00 | |
| bswartz | yeah we can't release a manila UI that doesn't work | 18:00 |
| *** Yogi1 has joined #openstack-manila | 18:00 | |
| ganso | bswartz: can we remove a feature in manila-ui bugfix period (between FF and RC1) if we think it is not good? | 18:01 |
| openstackgerrit | Merged openstack/manila: Updated from global requirements https://review.openstack.org/359516 | 18:01 |
| bswartz | ganso: we could, but that's not an excuse to merge a feature that doesn't work yet | 18:01 |
| ganso | bswartz: it works (still needs to be updated with nondisruptive parameter I added yesterday to main share migration patch), but it may not be bug-free or results be satisfactory | 18:02 |
| ganso | bswartz: but yes, it is not right to merge it before share migration main patches | 18:02 |
| ganso | bswartz: if main share migration patches were merged, we could do that even with some bugs and fix later | 18:03 |
| bswartz | okay so we should just merge all the manila migration patches before the end of the day today? | 18:03 |
| bswartz | :-D | 18:03 |
| ganso | bswartz: but I highly advise people to try it and see what they think | 18:03 |
| ganso | bswartz: lol | 18:04 |
| ganso | bswartz: if could if they are all reviewed, have you taken a look at https://review.openstack.org/#/c/345142/ yet ? | 18:04 |
| ganso | bswartz: *we could if... | 18:04 |
| bswartz | ganso: needs a rebase | 18:05 |
| ganso | bswartz: ^ that one is a bit scary DB migration... seems like it hasn't broken anything | 18:05 |
| ganso | bswartz: yes, I know... will rebase it very soon, later today | 18:05 |
| ganso | bswartz: I still would like to update the dummy driver support patch | 18:08 |
| *** vbellur has joined #openstack-manila | 18:16 | |
| tbarron | tpsilva: one #openstack-infra mwhahaha just asked: Anyone know who the owner of 'Hitachi Manila HNAS CI' is and why it's trying to run on puppet-swift changes? | 18:16 |
| tpsilva | tbarron: hm, weird | 18:17 |
| tpsilva | tbarron: I'll check with our CI guy :) | 18:17 |
| tpsilva | tbarron: thanks for letting me know | 18:17 |
| tbarron | tpsilva: is it you they want to talk to? | 18:17 |
| tpsilva | tbarron: no... it's marcusvrn | 18:18 |
| tbarron | tpsilva: k, they are making noises about disabliing stuff over in #openstack-infra | 18:19 |
| *** aovchinnikov has quit IRC | 18:19 | |
| tpsilva | tbarron: I'll get in there to check that as well | 18:19 |
| tpsilva | thank you | 18:19 |
| *** marcusvrn_ has joined #openstack-manila | 18:20 | |
| tpsilva | tbarron: thanks a lot... sounds like if we missed that it could get disabled | 18:22 |
| *** akerr has quit IRC | 18:23 | |
| tbarron | tpsilva: yeah, I don't really know. Don't see the relevant parties on this channel to ask, just seemed safer to check. | 18:24 |
| *** akerr has joined #openstack-manila | 18:24 | |
| *** rooneym has quit IRC | 18:25 | |
| tpsilva | tbarron: I'll setup some keywords to get notifications from openstack-infra to avoid that happening again :) | 18:26 |
| tbarron | tpsilva: i have a ping on 'manila'. not too annoying yet. | 18:29 |
| *** aovchinnikov has joined #openstack-manila | 18:40 | |
| *** lpetrut has joined #openstack-manila | 18:43 | |
| bswartz | who is mwhahaha? | 19:10 |
| bswartz | ganso: no matter what I do, admin_net and default have the same IP range | 19:13 |
| * bswartz must be doing something wrong | 19:14 | |
| ganso | bswartz: so you are trying to override and it is not working? | 19:14 |
| bswartz | ganso: what local.conf options should I use to set them | 19:19 |
| bswartz | I was using FIXED_RANGE and MANILA_ADMIN_NET_RANGE | 19:19 |
| bswartz | only the value for FIXED_RANGE got used though | 19:20 |
| *** lpetrut has quit IRC | 19:26 | |
| bswartz | ganso: ah I was using your old patch not your new one | 19:27 |
| bswartz | trying again now | 19:27 |
| *** lpetrut has joined #openstack-manila | 19:30 | |
| *** lpetrut has quit IRC | 19:33 | |
| ganso | bswartz: I am not sure, but I believe it will only work if you export the env var... not if you set in local.conf prior to stacking | 19:35 |
| bswartz | ganso: o_O | 19:35 |
| ganso | bswartz: i've seen this for several env vars in my CI | 19:35 |
| bswartz | ganso: the whole point for local.conf is to set the variables that devstack consumes | 19:35 |
| bswartz | ganso: if that doesn't work we may have bugs in our devstack plugin | 19:36 |
| ganso | bswartz: yes, but I don't know if the script exports everything that is in local.conf before starting pre_test_hook. in that case it would work, but as I've seen this not work in my CI I am led to believe it does not do this | 19:36 |
| ganso | bswartz: this forces us to check if something is in local.conf before doing anything | 19:37 |
| bswartz | it shouldn't need to export... | 19:37 |
| bswartz | we should always read from localconf | 19:37 |
| ganso | bswartz: we currently just add to local.conf, regardless of what is there, so it may in fact end up overwriting local.conf I suppose | 19:37 |
| *** lpetrut has joined #openstack-manila | 19:38 | |
| ganso | bswartz: I mostly replicated what is already in pre_test_hook.sh | 19:38 |
| bswartz | ganso: there's only 1 export in there, and it's MANILA_TEMPEST_CONCURRENCY | 19:39 |
| ganso | bswartz: yes, but all other values such as echo "TEMPEST_CONCURRENCY=${MANILA_TEMPEST_CONCURRENCY:-8}" >> $localrc_path | 19:41 |
| akerr | a lot of values are set in post_test_hook because it runs after devstack and can use iniset to update the conf files | 19:41 |
| bswartz | right. most of that file is generating local.conf | 19:42 |
| ganso | bswartz: I am not sure if that is overriding if TEMPEST_CONCURRENCY is already set | 19:42 |
| bswartz | I don't use the pre test hooks or the post test hooks | 19:42 |
| bswartz | developers aren't supposed to use those -- they're ONLY for CI | 19:42 |
| bswartz | the devstack plugin is meant to stand on its own | 19:42 |
| gouthamr | ganso: so, migration-get-progress will remain a call through newton.. | 19:44 |
| gouthamr | ganso: while migration-start changes from RPC call to cast | 19:44 |
| *** aovchinnikov has quit IRC | 19:45 | |
| gouthamr | ganso: wanted to point out that i ran into an issue with this.. i was in the middle of a m-shr restart; and the RPC call timed out after a while | 19:45 |
| gouthamr | ganso: maybe, before you make the call, you need to check if the service is up and enabled | 19:46 |
| ganso | gouthamr: timeout is 60 seconds | 19:46 |
| ganso | gouthamr: hummm, could do that | 19:47 |
| gouthamr | ganso: yeah.. why wait, you can send an error immediately | 19:47 |
| bswartz | oookay my scripts are seriously buggy | 19:50 |
| bswartz | that's why none of this is working right | 19:50 |
| ganso | bswartz: for CI it is using pre_test_hook, which sets the variable in local.conf... plugin.sh reads what is in local.conf. This is what the user would use, so there is no problem here | 19:51 |
| ganso | bswartz: if a user is setting in local.conf, then it will work | 19:51 |
| bswartz | ganso: yeah ignore all my earlier comments until I stop being an idiot | 19:51 |
| bswartz | I'll get this right, then I'll complain if I'm still having an issue | 19:52 |
| ganso | bswartz: but the case where it might not work because local.conf is overriden as I said above is for CI as well, not user's | 19:52 |
| ganso | bswartz: as for "might not work" I mean, what is set in local.conf previously to devstack... pre_test_hook is ensuring the default is being applied | 19:52 |
| *** porrua has quit IRC | 20:06 | |
| openstackgerrit | Merged openstack/manila: Rename and move HNAS driver https://review.openstack.org/356049 | 20:07 |
| *** Yogi1 has quit IRC | 20:08 | |
| bswartz | ganso: okay sanity has returned | 20:14 |
| *** vbellur has quit IRC | 20:14 | |
| bswartz | I'm seeing the networks created correctly by devstack now | 20:14 |
| bswartz | only issue is that generic driver doesn't seem to have a network interface on the tenant subnet | 20:15 |
| *** akerr has quit IRC | 20:15 | |
| ganso | bswartz: AFAIK it shouldn't have, there is a config "connect_to_tenant_network" that needs to be set for that, but it is not by default. The default is to have a router connect the manila_service_network and the tenant_network... not a network interface | 20:17 |
| *** akerr has joined #openstack-manila | 20:45 | |
| *** akerr_ has joined #openstack-manila | 20:47 | |
| *** akerr has quit IRC | 20:50 | |
| bswartz | ganso: okay I'm having different problems with the generic driver now (related to cinder) but I think your patch is fine | 20:55 |
| ganso | bswartz: great, can we merge it then? :) | 20:56 |
| bswartz | anyone seen this failure w/ generic driver? http://paste.openstack.org/show/563138/ | 20:57 |
| bswartz | ganso: -1 from jenkins | 20:58 |
| ganso | bswartz: dang... I will check that once I get home | 20:58 |
| gouthamr | bswartz: manila service tenant issues? | 20:58 |
| bswartz | k it looks fine to me other than that | 20:58 |
| bswartz | gouthamr: yeah probably something changed and I didn't update my conf | 20:59 |
| bswartz | gouthamr: if you know what the change is it would help | 20:59 |
| gouthamr | bswartz: strange this is DHSS=True? | 21:00 |
| bswartz | yes, generic | 21:00 |
| gouthamr | bswartz: the nova VM got created fine, but the cinder API gave you an error; weird indeed.. nope never seen that issue.. 401 is the auth token response from the cinder wsgi app.. | 21:03 |
| openstackgerrit | Jay Mehta proposed openstack/manila: Enhance setup_server() unit-tests for Huawei, generic, share drivers https://review.openstack.org/360088 | 21:10 |
| *** timcl has quit IRC | 21:23 | |
| *** gouthamr has quit IRC | 21:23 | |
| *** dsariel has joined #openstack-manila | 21:38 | |
| *** gouthamr has joined #openstack-manila | 21:42 | |
| *** lpetrut has quit IRC | 21:55 | |
| *** alyson_ has quit IRC | 22:03 | |
| *** dsariel has quit IRC | 22:12 | |
| *** dustins has quit IRC | 22:18 | |
| *** cknight has quit IRC | 22:31 | |
| *** marcusvrn_ has quit IRC | 22:54 | |
| *** zigo has quit IRC | 22:56 | |
| *** zigo has joined #openstack-manila | 22:57 | |
| *** tpsilva has quit IRC | 23:08 | |
| *** faiz89 has joined #openstack-manila | 23:22 | |
| *** xyang1 has quit IRC | 23:31 | |
| *** cknight has joined #openstack-manila | 23:40 | |
| *** cknight1 has joined #openstack-manila | 23:46 | |
| *** cknight has quit IRC | 23:46 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!