opendevreview | Merged openstack/manila master: Fix Manila API error message https://review.opendev.org/c/openstack/manila/+/873087 | 06:59 |
---|---|---|
*** chuanm4 is now known as chuanm | 08:35 | |
opendevreview | Maurice Escher proposed openstack/manila master: tox4 needs allowlist_externals for fast8 https://review.opendev.org/c/openstack/manila/+/874645 | 09:48 |
opendevreview | Liron Kuchlani proposed openstack/manila-tempest-plugin master: Change the expected status response from Forbidden to NotFound https://review.opendev.org/c/openstack/manila-tempest-plugin/+/872260 | 11:08 |
opendevreview | Nahim Alves de Souza proposed openstack/manila master: NetApp ONTAP: REST transition - Data motion related functions https://review.opendev.org/c/openstack/manila/+/871104 | 11:53 |
opendevreview | Nahim Alves de Souza proposed openstack/manila master: NetApp ONTAP: REST transition - DHSS True functions https://review.opendev.org/c/openstack/manila/+/871105 | 11:53 |
opendevreview | kiran pawar proposed openstack/manila master: Add 'state' column in 'services' table. https://review.opendev.org/c/openstack/manila/+/859301 | 12:13 |
opendevreview | Maurice Escher proposed openstack/manila master: allow overide of basepython with TOX_PYTHON env https://review.opendev.org/c/openstack/manila/+/874647 | 12:16 |
opendevreview | Hassan Asghar proposed openstack/python-manilaclient master: Fix share network create command with the AZ option https://review.opendev.org/c/openstack/python-manilaclient/+/837116 | 16:13 |
opendevreview | Merged openstack/manila master: allow overide of basepython with TOX_PYTHON env https://review.opendev.org/c/openstack/manila/+/874647 | 17:39 |
carloss | nahimsouza[m]: I was checking the dhss=true REST change and realized we're missing the default_ad_site functionality there | 17:58 |
nahimsouza[m] | carloss: we checked the documentation and it states that the field "default-site" has no equivalent in REST API yet, the API team said they will include this only in ONTAP 9.13.1, that's why we didn't include this change, but we will include it later | 18:04 |
nahimsouza[m] | sorry, I forgot to mention this earlier | 18:07 |
carloss | okay, nahimsouza[m] - thanks | 18:07 |
gouthamr | felipe_rodrigues: o/ | 19:29 |
gouthamr | felipe_rodrigues: i left a couple of comments for you on the metadata change: https://review.opendev.org/c/openstack/manila/+/869708 | 19:29 |
felipe_rodrigues | ok. I will try to solve it as soon as possible to avoid missing the FFE | 19:30 |
gouthamr | yeah, i'm not hoping it'd be a huge rewrite | 19:30 |
gouthamr | felipe_rodrigues: i also assume you've already tested this and it works like you expect.. | 19:32 |
felipe_rodrigues | Yeah.. I need to test again with this change. The CI does not test this kind of feature yet.. | 19:39 |
felipe_rodrigues | You have 3 points: | 19:46 |
felipe_rodrigues | 1. Remove the field "network_type" from metadata, given it is redundant. The "set_vlan" is enough to pass the ID and the network type ==> Fine | 19:46 |
felipe_rodrigues | 2. Get the mtu once instead of for each allocation ==> Enhancement | 19:46 |
felipe_rodrigues | 3. "PLEASE RAISE A USER MESSAGE HERE. With...." => I couldn't get your point. What did you mean ? The code is raising an exception when vlan is out of the range | 19:46 |
gouthamr | on point 2. - that's an enhancement sure, but i'm also asking for the MTU to default to the network plugin provided value like carloss suggested earlier | 19:57 |
gouthamr | before ofcourse defaulting to the hard coded value in the driver | 19:57 |
gouthamr | on 3. i'd like the driver to raise a user message - either directly or via the share manager | 20:00 |
gouthamr | 3 can come as an enhancement that we can backport | 20:00 |
gouthamr | https://docs.openstack.org/manila/latest/contributor/user_messages.html | 20:00 |
gouthamr | felipe_rodrigues: something like this: https://opendev.org/openstack/manila/src/commit/209b8ccab158e5097d2dacd5a9cfae1420af5082/manila/share/drivers/netapp/dataontap/cluster_mode/lib_multi_svm.py#L2151-L2160 (thanks carloss!) | 20:05 |
felipe_rodrigues | nice | 20:19 |
felipe_rodrigues | about setting the net plugin provided mtu, the idea of this feature is setting L2 values manually because neutron cannot. otherwise I should also check the VLAN from Neutron and so on.. I don't want to cover the case of Neutron L2 and manual L2 working together, since it makes the solution more complex and it is not what the customer desires (at least he didn't mention that).. I want to let it clear: either set | 20:25 |
felipe_rodrigues | L2 from manual or from Neutron.. | 20:25 |
felipe_rodrigues | I can add the plugin L2 mtu value in the game.. if you think it worths.. I just want to make sure that there is a use case for that.. Why the solution is ignoring neutron segmentation and network type and it would take the mtu in consideration ? they are related to the same L2 infra.. | 20:30 |
gouthamr | felipe_rodrigues: sorry was on a call | 21:52 |
gouthamr | felipe_rodrigues: > Why the solution is ignoring neutron segmentation and network type and it would take the mtu in consideration ? they are related to the same L2 infra.. | 21:52 |
gouthamr | we wouldn't be ignoring it - we're allowing overrides | 21:53 |
gouthamr | right now, that's what's happening - if you set_vlan, you override whatever the plugin is setting | 21:53 |
opendevreview | Merged openstack/manila master: NetApp ONTAP: REST transition - DHSS False Basic operations https://review.opendev.org/c/openstack/manila/+/871103 | 23:11 |
felipe_rodrigues | I see.. no problem.. I can add the MTU from neutron.. Just need some time to test the solution manually again | 23:27 |
felipe_rodrigues | Trying to do right now.. Then, I will submit a new patchset | 23:27 |
gouthamr | great thanks felipe_rodrigues | 23:41 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!