Earlier  
Posted Nick Remark
#openstack-nova - 2019-01-09
04:07:10 openstackgerrit Takashi NATSUME proposed openstack/nova stable/queens: Add description of custom resource classes https://review.openstack.org/619125
07:38:32 openstackgerrit Zhenyu Zheng proposed openstack/nova-specs master: Amend the detach-boot-volume design https://review.openstack.org/619161
08:33:46 openstackgerrit Yikun Jiang proposed openstack/nova master: Per aggregate scheduling weight https://review.openstack.org/628163
08:58:43 gibi stephenfin: left a question in https://review.openstack.org/#/c/629281
09:23:51 sean-k-mooney gibi: i replied to your question. i think is a safe assumtion to make if they have enabled ram oversubcription that the smallest pagesize on the host is swappable and therefor can be over subsribed
09:24:34 sean-k-mooney the linux kernel does not allow you to reserve all memory on a node for hugepages so the smallest pagesize will always be the system default small page size
09:40:20 openstackgerrit Maciej Jozefczyk proposed openstack/nova master: Add fill_virtual_interface_list online_data_migration script https://review.openstack.org/614167
09:40:21 openstackgerrit Maciej Jozefczyk proposed openstack/nova master: Force refresh instance info_cache during heal https://review.openstack.org/591607
09:41:20 maciejjozefczyk sean-k-mooney: I found a bug in https://review.openstack.org/#/c/614167
09:42:09 gibi sean-k-mooney: thanks for the explanation
09:48:19 openstackgerrit Jan Gutter proposed openstack/nova master: Convert vrouter legacy plugging to os-vif https://review.openstack.org/571325
09:51:29 kashyap msg = _("Setting both 'live_migration_tunnelled' and"
09:51:29 kashyap Is this too verbose a message?
09:51:30 kashyap "libvirt and QEMU versions, and have TLS"
09:51:30 kashyap "time is invalid. If you have the relevant"
09:51:30 kashyap "'live_migration_with_native_tls' at the same"
09:51:32 kashyap "configured in your environment, pick"
09:51:35 kashyap "'live_migration_with_native_tls'.")
09:53:02 kashyap stephenfin: ^
09:54:22 kashyap (Or anyone interested in wordsmithing)
09:55:37 gibi kashyap: I like this message but I'm not a native English speaker
09:56:18 kashyap I consider myself an "English geek" (although it's my 3rd language). I went for maximum clarity for non-native speakers.
09:56:28 kashyap Because: "explicit is better than implicit".
09:56:49 kashyap And I don't like error messages that tells me what is not valid (that I already see from the error); give me something actionable.
09:57:27 kashyap I'll go with the above (with the second occurrence of "have" removed), and will see if anyone can come up w/ something better.
09:59:18 sean-k-mooney kashyap: it reads fine to me. its hard to be too verbose when explaining that you have conflicting config options set
09:59:37 sean-k-mooney you even provided a recommendation on what option to prefer so i looks good to me too
09:59:49 kashyap Okido; thanks for the review.
09:59:54 sean-k-mooney maciejjozefczyk: was it related to the socket timeout or unrelated.
10:00:03 kashyap As I noted, my bar for error messages is: "give me something actionable"
10:00:19 kashyap It is really frustrating to just see "error: please figure it out"
10:00:42 sean-k-mooney you mean like "no valid host"
10:00:46 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: libvirt: Support native TLS for migration and disks over NBD https://review.openstack.org/625216
10:00:47 kashyap sean-k-mooney: Heh
10:01:06 sean-k-mooney its my favorite nova error message
10:01:17 kashyap Yeah, it's the most awful message.
10:01:25 kashyap One day, if I feel angry enough, I might clean them all up
10:03:00 sean-k-mooney well its availd error message but the issue is 90% of the time it because you hit the retry filter limit and you hit the retry filter because or a different error which was lost in the logs
10:03:22 kashyap Yeah, I recall that it's "best we could do"
10:35:43 maciejjozefczyk sean-k-mooney: nope, different topic
10:35:48 maciejjozefczyk already fixed in new patch
10:36:22 sean-k-mooney cool i had only tried the followon patch as i was testing on master so did not need the db change
10:38:50 maciejjozefczyk for me I needed to update about 7k vifs only for one db
11:08:54 nehaalhat gibi: Hi, in case of functional test cases, how nova sync with placement service? Do you any idea about this?
11:11:32 gibi nehaalhat: in functional test nova imports placement code via https://github.com/openstack/nova/blob/e3eb5f916580a9bab8f67b0fd685c6b3b23a97b7/nova/tests/functional/fixtures.py#L41
11:12:42 nehaalhat like aggregates created in nova, how they are synced in placement service?
11:22:28 nehaalhat gibi: and do you have idea how to mock a method from placement in nova functional test cases where placementfixture is used like: https://github.com/openstack/nova/blob/e3eb5f916580a9bab8f67b0fd685c6b3b23a97b7/nova/tests/functional/test_aggregates.py#L385
11:31:47 gibi nehaalhat: do you have a patch with a functional test which doesn't work without mocking something in placement?
11:32:08 gibi nehaalhat: in general mocking in functional test is discuraged
11:32:26 gibi nehaalhat: but I can look at your patch to understand more
11:34:21 nehaalhat gibi: No, i dont have a patch on community gerrit right now
11:35:02 gibi nehaalhat: let me try to dig out where nova sync the aggregates to placement...
11:35:18 nehaalhat ok
11:36:01 gibi nehaalhat: there is a nova-manage CLI command for aggregates sync https://github.com/openstack/nova/blob/da17e0ed8a9b9ddc79b6d2004e6fb941c6163fbf/nova/cmd/manage.py#L2158
11:37:02 gibi nehaalhat: and this is the patch that introduced the automatic syncing https://review.openstack.org/#/c/553597/
11:37:11 nehaalhat gibi: actually I want to add a FT under: https://github.com/openstack/nova/blob/e3eb5f916580a9bab8f67b0fd685c6b3b23a97b7/nova/tests/functional/test_aggregates.py#L385 where placement fixture is used, and I want to mock this method in my FT
11:38:01 gibi nehaalhat: and this is the point where nova adds a host to a placement aggregate https://review.openstack.org/#/c/553597/15/nova/compute/api.py@5147
11:38:54 gibi nehaalhat: so I think if your functional test adds a host to an aggregate then nova will automatically add that host to the placement aggregate as weel
11:38:57 gibi well
11:39:30 nehaalhat gibi: ok thanks
11:40:18 gibi nehaalhat: but if you push a wip patch to gerrit then I can help making it work
11:41:11 nehaalhat gibi: I want to add a FT under: https://github.com/openstack/nova/blob/e3eb5f916580a9bab8f67b0fd685c6b3b23a97b7/nova/tests/functional/test_aggregates.py#L385 where placement fixture is used
11:41:25 nehaalhat and I want to mock this method in my FT : https://github.com/openstack/placement/blob/master/placement/util.py#L236
11:42:05 nehaalhat actually I cant push on community gerrit without internal approval
11:42:07 gibi nehaalhat: why do you need to mock that?
11:42:12 openstackgerrit Jan Gutter proposed openstack/nova master: Add support for vrouter HW datapath offloads https://review.openstack.org/572082
11:43:01 gibi nehaalhat: what is the use case you want to tests?
11:46:09 ygk_12345 hi all
11:46:46 ygk_12345 I am seeing this error on the compute node
11:46:53 ygk_12345 HypervisorUnavailable: Connection to the hypervisor is broken on host: compute-7
11:47:01 ygk_12345 i cant do a virsh list command
11:47:14 nehaalhat gibi: working on this feature: https://review.openstack.org/#/c/609960/3/specs/stein/approved/placement-req-filter-forbidden-aggregates.rst
11:47:20 sean-k-mooney ygk_12345: that usually means the libvirtd daemon has exited
11:47:33 ygk_12345 sean-k-mooney: what does that mean ?
11:47:56 sean-k-mooney the libvirtd service has crashed/stopped on the compute node
11:48:12 jangutter ygk_12345 what distro and which version are you running?
11:48:33 ygk_12345 jangutter: Ubuntu 16.04.3 LTS \n \l
11:48:40 sean-k-mooney ygk_12345: try systemctl status libvirtd
11:49:03 ygk_12345 sean-k-mooney: I see this
11:49:05 ygk_12345 sean-k-mooney: 2019-01-09 11:48:44.365+0000: 16119: error : virPidFileAcquirePath:422 : Failed to acquire pid file '/var/run/libvirtd.pid': Resource temporarily unavailable
11:49:19 ygk_12345 sean-k-mooney: i did libvirtd -l
11:49:41 ygk_12345 sean-k-mooney: Failed to open file '/proc/28288/stat': No such file or directory
11:49:41 sean-k-mooney ygk_12345: that will try to start a new libvirtd
11:49:50 sean-k-mooney that is not the same thing as runing status
11:50:08 ygk_12345 sean-k-mooney: the libvirtd is running finr though
11:50:16 ygk_12345 sean-k-mooney: failed to connect to monitor socket: No such process
11:50:36 sean-k-mooney you said you cant run virsh command so its obviouly in an error state
11:50:41 nehaalhat gibi: and this spec is dependent on https://review.openstack.org/#/c/603352/4/specs/stein/approved/negative-aggregate-membership.rst
11:50:52 ygk_12345 sean-k-mooney: virsh list command is hanging
11:50:56 ygk_12345 sean-k-mooney: no output
11:51:04 gibi nehaalhat: OK. Do you need to mock something from placement because the placement change that supports your feature is not merged yet?
11:51:11 sean-k-mooney yes because the libvirt daemon is not responding to requests
11:51:17 nehaalhat gibi: yes
11:51:31 f10d4 hi
11:51:46 ygk_12345 sean-k-mooney: now after sometime the vrish command is working but the flaoting IP of the instances is not pingable
11:52:15 ygk_12345 sean-k-mooney: it seems tot fix itself after sometime
11:52:22 ygk_12345 sean-k-mooney: bit strange
11:52:23 sean-k-mooney virsh is just a client that publishes request to the libvirt daemon to execute commands via dbus
11:52:24 jangutter ygk_12345: Ubuntu 16.04 doesn't always use systemd: you might have to look at libvirt's logs directly. Check your system logs to see if everything's healthy or something isn't restarting all the time.
11:52:33 gibi nehaalhat: if you have both the placement change an the nova change locally then you have to instruct the nova test to use your local placement change for the functional test

Earlier   Later