December 2020 - openstack-discuss

[tacker] Next meeting Tuesday 12 Jan
by Yasufumi Ogawa 31 Dec '20

31 Dec '20

Hi team, As we agreed in the last meeting, we will skip IRC meeting next week for new year holidays. Thank you & happy holidays! Yasufumi

1 0

[horizon] No meetings until January, 13th
by Ivan Kolodyazhny 30 Dec '20

30 Dec '20

Hi team, As agreed during the last IRC meeting [1], we'll skip the next two meetings due to the Christmas and New Year holidays. [1] http://eavesdrop.openstack.org/meetings/horizon/2020/horizon.2020-12-23-15.… Regards, Ivan Kolodyazhny, http://blog.e0ne.info/

1 0

Openstack instance management prive
by Ranjeet Kumar 30 Dec '20

30 Dec '20

Hi What could be the average unit price for the life cycle management of an instance in open stack environment?

1 0

Instance on provider network not working
by mohammad mighani 30 Dec '20

30 Dec '20

Hello I use openstack Ussuri in Ubuntu 18 and one controller node and one compute node with two interface, management and Provider that connected to internet i install keystone, glance, placement, nova, neutron on self service option *what i can * i can create provider network and self service network and router between them i can ping router i can launch instance on self service network and it has access to internet and everything looks fine *what i can NOT* I can't launch instances on the provider network. When I launch an instance on the provider network, the provider interface of the compute node will be corrupted and not respond and the instance will not be created. error in instance : *Message *Build of instance 2a86a84b-50fc-415c-a15c-cb87398f37ea aborted: Failed to allocate the network(s), not rescheduling. *Code *500 *Details *Traceback (most recent call last): File "/usr/lib/python3/dist-packages/nova/virt/libvirt/driver.py", line 6529, in _create_domain_and_network post_xml_callback=post_xml_callback) File "/usr/lib/python3.6/contextlib.py", line 88, in __exit__ next(self.gen) File "/usr/lib/python3/dist-packages/nova/compute/manager.py", line 513, in wait_for_instance_event actual_event = event.wait() File "/usr/lib/python3/dist-packages/eventlet/event.py", line 125, in wait result = hub.switch() File "/usr/lib/python3/dist-packages/eventlet/hubs/hub.py", line 298, in switch return self.greenlet.switch() eventlet.timeout.Timeout: 300 seconds During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/lib/python3/dist-packages/nova/compute/manager.py", line 2378, in _build_and_run_instance accel_info=accel_info) File "/usr/lib/python3/dist-packages/nova/virt/libvirt/driver.py", line 3663, in spawn cleanup_instance_disks=created_disks) File "/usr/lib/python3/dist-packages/nova/virt/libvirt/driver.py", line 6552, in _create_domain_and_network raise exception.VirtualInterfaceCreateException() nova.exception.VirtualInterfaceCreateException: Virtual Interface creation failed During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/lib/python3/dist-packages/nova/compute/manager.py", line 2200, in _do_build_and_run_instance filter_properties, request_spec, accel_uuids) File "/usr/lib/python3/dist-packages/nova/compute/manager.py", line 2444, in _build_and_run_instance reason=msg) nova.exception.BuildAbortException: Build of instance 2a86a84b-50fc-415c-a15c-cb87398f37ea aborted: Failed to allocate the network(s), not rescheduling. *Created *Dec. 29, 2020, 9:18 a.m. *After this I can't ping the computer from the provider network interface.* I reinstalled openstack so many times and also I tried the train and victoria version and have the same problem. I attached log files of nova compute and neutron linux bridge from compute node and neutron server from controller node. thanks.

2 1

[neutron] Team meetings this year cancelled
by Slawek Kaplonski 30 Dec '20

30 Dec '20

Hi, As we discussed during our last meeting, we are going to cancel our team meetings in next 2 weeks. Have a great holidays and see You all on the meeting at 5.01.2021 -- Slawek Kaplonski Principal Software Engineer Red Hat

2 1

[Nova][queens] disk.config not found issue
by Ignazio Cassano 30 Dec '20

30 Dec '20

Hello All, after a compute node failure if I try to reset state and hard reboot an instance it fails because disk.config file does not exist. I am using cinder with netapp nfs driver and I am using disk config drive. It happens also on stein. Any help, please ? It can be' solved creating the disk.config with touch command but it does not seem a good solution . Ignazio

1 1

[osc][neutron] security group list slow in Stein
by Eric K. Miller 29 Dec '20

29 Dec '20

Hi, We are working on upgrade planning for our Stein deployments, but since we have some time before this will be finalized, I thought I'd ask if there has been any performance improvement/fixes in the performance of the "security group list" command in the OpenStack Client and/or Neutron API. "security group rule list" is nearly instant (when authenticated to a project), and returns more info than security group list, but "security group list" can take up to 29 seconds in the worst case when authenticated to a project. Oddly enough, when logged in as cloud admin, "security group list" only takes about 14 seconds, even though it is displaying every security group in the cloud (in our case, quite a few!). Also, just wondering if anyone else has this issue. Thanks! Eric

1 0

[neutron] bug deputy report December 21st to 27th
by Miguel Lavalle 29 Dec '20

29 Dec '20

Hi, This week Slawek and I cooperated to perform the bug deputy duties. It was a relatively quiet week. These are the bugs triaged during the week: High ==== https://bugs.launchpad.net/neutron/+bug/1908957 iptable rules collision deployed with k8s iptables kube-proxy enabled. Owner: Norman Shen https://bugs.launchpad.net/neutron/+bug/1909038 [ovn] TypeError: lrp_set_options() takes 2 positional arguments but 3 were given. Owner: Flavio Fernandes https://bugs.launchpad.net/neutron/+bug/1909234 [fullstack] "test_min_bw_qos_port_removed" failing randomly. Owner: Rodolfo Alonso RFE === https://bugs.launchpad.net/neutron/+bug/1909100 [RFE]add new vnic type "cyborg"

1 0

Re: something abount ironic-python-agent
by Kaifeng Wang 28 Dec '20

28 Dec '20

On Sun, Dec 27, 2020 at 8:39 PM Ankele zhang <ankelezhang(a)gmail.com> wrote: > Hi Kaifeng: > Before receiving your reply, I have downloaded > the ronic-python-agent-builder-2.3.0, but I don't know how to build it and > how to use it, I can't find the direction for use. > And I try 'python setup.py install' to install it. and the following is my > execution. > > I don't know the optional values of 'distribution', 'release', 'element', > I hope you can help me. > > distribution refers to the base image that the ramdisk will be built on, release is used to specify the distribution variant. element is a concept of diskimage-builder, each element is handling part of the building process, you can take a look at the diskimage-builder documentation for a complete list [1]. In case further customization on the ramdisk is required, elements can be included by the argument. We have some examples on how to use the IPA-B at [2]. [1] https://docs.openstack.org/diskimage-builder/latest/elements.html [2] https://docs.openstack.org/ironic-python-agent-builder/latest/admin/dib.html Ankele > > Kaifeng Wang <kaifeng.w(a)gmail.com> 于2020年12月27日周日下午2:04写道： > >> >> >> On Sun, Dec 27, 2020 at 12:12 AM Ankele zhang <ankelezhang(a)gmail.com> >> wrote: >> >>> Hi ~ >>> I have deployed OpenStack platform in Rocky version and integreted >>> Ironic(Rocky) into it according to the official documents. I download the >>> coreOS vmlinuz and ramdisk images from >>> https://tarballs.opendev.org/openstack/ironic-python-agent/coreos/files/. >>> and if I don't clean my devices I can deploy baremetal node server >>> currently. When I cleaning my disk using "openstack baremetal node clean >>> --clean-steps '[{"interface": "deploy", "step": "erase_devices"}]' my_bm", >>> I got an error message "Clean step failed: Error performing clean_step >>> erase_devices: No HardwareManager found to handle method: Could not find >>> method: erase_block_device" from ironic-conductor.log and I got the same >>> error in coreOS by "sudo journalctl -u ironic-python-agent -f". I download >>> the Rocky ironic-python-agent source code from github and I search >>> 'erase_block_devices' ,it returned from hardware.py +781. So, the first >>> problem , I don't know why I got this error message. >>> >> >> CoreOS based ramdisk was not maintained for a while, you could try a >> rocky release at >> https://tarballs.opendev.org/openstack/ironic-python-agent/dib/ >> >> >>> I downloaded the source code of the ironic-python-agent and tried to >>> build the coreOS image by 'Makefile', but I always got error as following, >>> this is my second problem. >>> >>> >> Building an IPA ramdisk is moved from diskimage-builder to >> ironic-python-agent-builder, you can find the document here: >> https://docs.openstack.org/ironic-python-agent-builder/latest/ >> >> >>> >>> Finally, I want to code my own 'custom HardwareManager' to support 'raid >>> config' for the ironic-python-agent, but I don't know how to get started. >>> >>> >> For an in-band RAID support, you'll need to create a new hardware manager >> and implement some deploy steps to support RAID configuration, >> specifically, the create_configuration and delete_configuration steps. >> The following links may provide some starting points for you. >> https://docs.openstack.org/ironic/latest/contributor/deploy-steps.html >> >> https://docs.openstack.org/ironic/latest/admin/cleaning.html#raid-interface >> >> Hope this helps. >> >> >>> Looking forward to your help. Thank you. >>> >>> Ankele >>> >>

1 0

something abount ironic-python-agent
by Ankele zhang 27 Dec '20

27 Dec '20

Hi ~ I have deployed OpenStack platform in Rocky version and integreted Ironic(Rocky) into it according to the official documents. I download the coreOS vmlinuz and ramdisk images from https://tarballs.opendev.org/openstack/ironic-python-agent/coreos/files/. and if I don't clean my devices I can deploy baremetal node server currently. When I cleaning my disk using "openstack baremetal node clean --clean-steps '[{"interface": "deploy", "step": "erase_devices"}]' my_bm", I got an error message "Clean step failed: Error performing clean_step erase_devices: No HardwareManager found to handle method: Could not find method: erase_block_device" from ironic-conductor.log and I got the same error in coreOS by "sudo journalctl -u ironic-python-agent -f". I download the Rocky ironic-python-agent source code from github and I search 'erase_block_devices' ,it returned from hardware.py +781. So, the first problem , I don't know why I got this error message. I downloaded the source code of the ironic-python-agent and tried to build the coreOS image by 'Makefile', but I always got error as following, this is my second problem. [image: image.png] Finally, I want to code my own 'custom HardwareManager' to support 'raid config' for the ironic-python-agent, but I don't know how to get started. Looking forward to your help. Thank you. Ankele

2 1