openstackgerrit | Andre Beltrami proposed openstack/python-manilaclient master: Add commands for share server migration https://review.opendev.org/747204 | 00:32 |
---|---|---|
openstackgerrit | Merged openstack/manila-tempest-plugin master: Add uuid for tests https://review.opendev.org/701108 | 01:20 |
*** LinPeiWen has joined #openstack-manila | 01:23 | |
openstackgerrit | Douglas Viroel proposed openstack/python-manilaclient master: Add commands for share server migration https://review.opendev.org/747204 | 01:39 |
openstackgerrit | Merged openstack/manila master: Update "Create and manage shares" doc https://review.opendev.org/745291 | 02:07 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/manila-ui master: User messages panel https://review.opendev.org/742550 | 03:20 |
openstackgerrit | Merged openstack/manila master: Graduate share replication feature https://review.opendev.org/747073 | 03:44 |
*** carthaca has quit IRC | 04:32 | |
openstackgerrit | Merged openstack/manila stable/ussuri: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750179 | 04:38 |
openstackgerrit | Lin PeiWen proposed openstack/manila stable/train: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750180 | 05:12 |
openstackgerrit | Lin PeiWen proposed openstack/manila stable/train: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750180 | 05:15 |
*** Reepicheep has quit IRC | 05:42 | |
*** Reepicheep has joined #openstack-manila | 05:42 | |
*** vishalmanchanda has joined #openstack-manila | 06:14 | |
openstackgerrit | Lin PeiWen proposed openstack/manila stable/train: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750180 | 06:30 |
openstackgerrit | Lin PeiWen proposed openstack/manila stable/stein: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750182 | 06:31 |
*** Reepicheep has quit IRC | 06:43 | |
*** Reepicheep has joined #openstack-manila | 06:47 | |
*** AnaghaR has joined #openstack-manila | 07:31 | |
*** tosky has joined #openstack-manila | 07:57 | |
*** AnaghaR has quit IRC | 08:14 | |
*** Reepicheep has quit IRC | 08:32 | |
*** Reepicheep has joined #openstack-manila | 08:33 | |
*** carthaca has joined #openstack-manila | 08:39 | |
*** ianychoi__ has quit IRC | 08:40 | |
*** AnaghaR has joined #openstack-manila | 09:12 | |
*** Reepicheep has quit IRC | 09:15 | |
*** Reepicheep has joined #openstack-manila | 09:24 | |
openstackgerrit | Lin PeiWen proposed openstack/manila master: [Glusterfs] Fix delete share, Couldn't find the 'gluster_used_vols' https://review.opendev.org/750067 | 09:31 |
*** AnaghaR has quit IRC | 09:45 | |
*** Reepicheep has quit IRC | 10:33 | |
*** Reepicheep has joined #openstack-manila | 10:38 | |
openstackgerrit | Carlos Eduardo proposed openstack/manila master: Update share replica show docstrings https://review.opendev.org/750622 | 11:17 |
*** raildo has joined #openstack-manila | 11:30 | |
*** AnaghaR has joined #openstack-manila | 11:54 | |
vkmc | AnaghaR, o/ | 12:24 |
vkmc | could you submit your fix? | 12:24 |
*** andrebeltrami has joined #openstack-manila | 12:36 | |
*** Reepicheep has quit IRC | 13:06 | |
*** Reepicheep has joined #openstack-manila | 13:08 | |
*** kaisers_ has joined #openstack-manila | 13:22 | |
openstackgerrit | Merged openstack/python-manilaclient master: Graduate share replication feature https://review.opendev.org/747588 | 13:25 |
carloss | \o/ | 13:27 |
dviroel | carloss++ | 13:29 |
vkmc | carloss++ | 13:38 |
vkmc | :) | 13:38 |
*** AnaghaR has quit IRC | 13:43 | |
*** Reepicheep has quit IRC | 14:11 | |
*** Reepicheep has joined #openstack-manila | 14:15 | |
*** Reepicheep has quit IRC | 14:35 | |
*** Reepicheep has joined #openstack-manila | 14:37 | |
*** lpetrut has joined #openstack-manila | 15:12 | |
openstackgerrit | Merged openstack/manila-tempest-plugin master: Remove experimental flag from share replication feature https://review.opendev.org/747593 | 15:22 |
openstackgerrit | Merged openstack/manila stable/train: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750180 | 15:30 |
openstackgerrit | Merged openstack/manila master: Update share replica show docstrings https://review.opendev.org/750622 | 15:30 |
*** Reepicheep has quit IRC | 15:37 | |
*** Reepicheep has joined #openstack-manila | 15:39 | |
*** lpetrut has quit IRC | 15:59 | |
carloss | \o/ - thank you folks for reviewing these changes :) | 16:07 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/python-manilaclient master: Implement OSC share quota commands https://review.opendev.org/703992 | 16:16 |
openstackgerrit | Tom Barron proposed openstack/manila stable/stein: [Glusterfs] Fix delete share, mount point not disconnected https://review.opendev.org/750182 | 16:21 |
*** Reepicheep has quit IRC | 16:22 | |
*** Reepicheep has joined #openstack-manila | 16:23 | |
openstackgerrit | Ghanshyam Mann proposed openstack/manila master: [goal] Migrate tox based testing to ubuntu focal https://review.opendev.org/743137 | 16:28 |
*** Reepicheep has quit IRC | 16:35 | |
*** Reepicheep has joined #openstack-manila | 16:38 | |
*** Reepicheep has quit IRC | 17:03 | |
openstackgerrit | Douglas Viroel proposed openstack/manila master: [Container driver] Adds share and share server migration https://review.opendev.org/740831 | 17:04 |
*** Reepicheep has joined #openstack-manila | 17:05 | |
*** andrebeltrami has quit IRC | 17:06 | |
*** Reepicheep has quit IRC | 17:12 | |
*** Reepicheep has joined #openstack-manila | 17:16 | |
gmann | gouthamr: do you know what causing this error in https://review.opendev.org/743137 "ERROR: InvocationError for command /home/zuul/src/opendev.org/openstack/manila/tools/check_exec.py manila (exited with code 2)' | 17:55 |
gmann | here https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_863/743137/5/check/openstack-tox-pep8/863a599/job-output.txt | 17:55 |
gouthamr | gmann: looking | 17:58 |
gouthamr | gmann: i wonder if the interpreter doesn't exist.. | 18:03 |
gouthamr | we can specifically look for python3 | 18:03 |
gmann | seems so, testing with python3 | 18:03 |
*** Reepicheep has quit IRC | 18:04 | |
openstackgerrit | Goutham Pacha Ravi proposed openstack/manila master: Set check_exec interpretor to python3 https://review.opendev.org/750770 | 18:04 |
gouthamr | ^ gmann - you can fold this into your change if you wish | 18:05 |
*** Reepicheep has joined #openstack-manila | 18:05 | |
gmann | gouthamr: yeah we need to squash as 750770 fail on lower constraints, i will do | 18:17 |
openstackgerrit | Ghanshyam Mann proposed openstack/manila master: [goal] Migrate tox based testing to ubuntu focal https://review.opendev.org/743137 | 18:21 |
gmann | gouthamr: ^^ updted | 18:21 |
openstackgerrit | Tom Barron proposed openstack/manila master: [Container driver] Adds share and share server migration https://review.opendev.org/740831 | 18:29 |
gouthamr | ty gmann | 18:32 |
*** LinPeiWen has quit IRC | 18:36 | |
*** Reepicheep has quit IRC | 18:47 | |
*** Reepicheep has joined #openstack-manila | 18:48 | |
*** kaisers_ has quit IRC | 19:27 | |
openstackgerrit | Tom Barron proposed openstack/manila master: [NetApp] Add support for share server migration https://review.opendev.org/747048 | 19:29 |
*** tosky has quit IRC | 19:37 | |
*** vishalmanchanda has quit IRC | 19:45 | |
*** openstackgerrit has quit IRC | 20:17 | |
*** openstackgerrit has joined #openstack-manila | 20:18 | |
openstackgerrit | Goutham Pacha Ravi proposed openstack/python-manilaclient master: Add noqa markup to intentionally redefined methods https://review.opendev.org/750810 | 20:18 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/manila-ui master: User messages panel https://review.opendev.org/742550 | 20:24 |
gouthamr | carloss dviroel: an API question posted to https://review.opendev.org/#/c/747043/7/manila/share/api.py | 20:31 |
* dviroel looks | 20:37 | |
dviroel | gouthamr: i got your point, agree that admin will have some difficulties to identify to identify the destination share-server-id. | 20:49 |
dviroel | gouthamr: your idea on get_progress will work, we can search in the db for the 'source_share_server_id' and in the success scenario, the destination will be there | 20:50 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Stop using shaman https://review.opendev.org/749726 | 20:51 |
dviroel | gouthamr: and yes, we can also return the destination-share-server-id in 'get_progress', sounds good too | 20:51 |
gouthamr | dviroel: great, so your thought is to allow using either the "migrating" or the "migrating_to" share server IDs and still report the progress? | 20:52 |
gouthamr | dviroel: we can make the get_progress generic: "source_share_server_id" and "destination_share_server_id" can both be present in the get_progress response, wdyt? | 20:53 |
dviroel | gouthamr: it is possible, will need to add more code only | 20:54 |
gouthamr | ^ if we allow using either IDs | 20:54 |
dviroel | at this moment, all operations are supported only for the source-id | 20:54 |
dviroel | admin will need to still use the source-id to complete the migration | 20:54 |
gouthamr | dviroel: yes, i don't mind if you want to leave "get_progress" as is - when the source gets deleted, the API responds with a 404 | 20:55 |
gouthamr | dviroel: but, make sure to provide "destination_share_server_id" in all the API responses perhaps | 20:55 |
dviroel | about that ^ | 20:56 |
gouthamr | dviroel: migration_start, migration_get_progress and migration_complete | 20:56 |
dviroel | migration_start for instance, we will need to change the rpc cast to call, and wait for it to be processed by the destination | 20:57 |
dviroel | since the share-server creation is handled in the manager | 20:57 |
dviroel | get_progress is easy to fix that | 20:58 |
dviroel | complete is also easy since we already have the destination-id | 20:58 |
gouthamr | i see | 20:59 |
gouthamr | that still makes sense | 20:59 |
gouthamr | lets do it for get_progress and complete then | 21:00 |
dviroel | ack | 21:00 |
gouthamr | ty dviroel | 21:02 |
dviroel | carloss: lets work together in order to solve more issues | 21:03 |
carloss | sure dviroel | 21:03 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/manila-ui master: User messages panel https://review.opendev.org/742550 | 21:13 |
openstackgerrit | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Stop using shaman https://review.opendev.org/749726 | 21:18 |
vkmc | zuul down? | 21:35 |
vkmc | or is it me? | 21:35 |
* vkmc hits f5 | 21:36 | |
gouthamr | vkmc: ah, needs a restart apparently, per #openstack-infra | 21:48 |
openstackgerrit | Goutham Pacha Ravi proposed openstack/manila master: Update user doc index https://review.opendev.org/745292 | 22:12 |
*** raildo has quit IRC | 22:22 | |
openstackgerrit | Merged openstack/python-manilaclient master: Add noqa markup to intentionally redefined methods https://review.opendev.org/750810 | 22:38 |
dviroel | gouthamr: o/ | 22:56 |
gouthamr | hey dviroel | 22:56 |
dviroel | gouthamr: in get_progress, if the source was already deleted I'm looking for a destination server that is already Active, which means that will only works for success scenario. | 22:57 |
dviroel | gouthamr: won't be possible to check others status, since we can have failed destinations and so on, doesn't seems to make sens | 22:58 |
gouthamr | dviroel: how do you identify which one's the destination at that point? | 22:58 |
dviroel | this db search will only happen if the source_server doesn't exists anymore | 22:58 |
dviroel | gouthamr: we are not erasing anymore the source_share_server_id info | 22:59 |
gouthamr | dviroel: oh, is that new? | 22:59 |
dviroel | there is no need for that anymore | 22:59 |
dviroel | no, its already in the PS that you are reviewing | 22:59 |
dviroel | In the api we block a new migration only if: | 23:00 |
dviroel | there is already another destination_share_server with status 'server_migrating_to' that points to the same source | 23:00 |
dviroel | so it is possible to start a new migration if the previous one failed. | 23:01 |
dviroel | and this won't be a problem for the manager since it only search for status 'migrating_to' | 23:01 |
dviroel | and ignore the other ones | 23:01 |
dviroel | if we erase the 'source_share_server_id' of those failed destinations, the admin won't be able to identify the source after any errot | 23:02 |
dviroel | let us know if you think that there is a gap on that approacj | 23:02 |
dviroel | but this was changed after the collab review, i think that was something that was discussed at the meeting | 23:03 |
gouthamr | dviroel: yeah trying to think, i was intune with your previous approach where teh admin would be forced to ack and delete any failed migrations | 23:03 |
dviroel | And just to remember: all destinations are new share servers, the is no reuse.Any failure (or cancelled) migration will move the share-server to Inactive | 23:05 |
gouthamr | dviroel: okay, now you reset the status of the failed migration destination to "error"? | 23:05 |
dviroel | let me check that again | 23:05 |
gouthamr | inactive would work too | 23:06 |
dviroel | cancelled destination goes to inactive, failed destination goes to error (both 1st and 2nd phase) | 23:07 |
gouthamr | good - this sorta illustrates a reason for instances | 23:08 |
dviroel | both will be ignored by the 'provide_share_server_for_share' | 23:08 |
gouthamr | :) | 23:08 |
dviroel | oh, and the client won't be able to do a 'get-progress' on the already deleted source-share-server, since it does a '_find_share_server' first | 23:09 |
gouthamr | dviroel: ack, i'll take a further look.. your approach for get_progress makes sense | 23:09 |
dviroel | but look that works when testing with curl | 23:09 |
gouthamr | dviroel: yeah, we can remove that call its kinda redundant | 23:10 |
gouthamr | dviroel: afaiu the "find" methods in the CLI are useful when resolving names | 23:10 |
gouthamr | dviroel: share servers won't have names, so its a wasted lookup | 23:11 |
gouthamr | dviroel: not a release stopper however, we can optimize later | 23:12 |
dviroel | gouthamr: i need to take a further look in the code, but looks a little but tied with the ShareServer class in the client | 23:12 |
gouthamr | dviroel: ack | 23:12 |
dviroel | s/but/bit | 23:12 |
dviroel | tks | 23:13 |
openstackgerrit | Goutham Pacha Ravi proposed openstack/python-manilaclient master: Implement the first half of OSC share snapshots commands https://review.opendev.org/738691 | 23:20 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!