July 2022 - openstack-discuss - lists.openstack.org

[tc] Reminder: August 2022 OpenInfra Board Sync
by Jeremy Stanley 07 Jul '22

07 Jul '22

If you're interested in participating in an informal discussion between the OpenStack TC, OpenInfra board members, and other interested community collaborators, don't forget to mark your preferred dates and times by Friday, 2022-07-15, so that I can let the board members know what our collective availability looks like: https://framadate.org/atdFRM8YeUtauSgC If you don't know what this is about, see my earlier post for details (I intentionally avoided replying to it in order to increase visibility for the reminder): https://lists.openstack.org/pipermail/openstack-discuss/2022-June/029352.ht… -- Jeremy Stanley

1 0

[manila] Stepping down from manila core
by Rodrigo Barbieri 07 Jul '22

07 Jul '22

Hello fellow zorillas, It has been a long time since I started to hope every day I'd able to dedicate more time to manila core activities and so far that hasn't happened and I don't see it happening in the near foreseeable future. I had been following the meetings notes weekly until ~2 months ago but I recently ended up dropping those as well. Therefore I am stepping down from the manila core role. I would like to thank everyone that I worked closely with from 2014 to 2019 on this project. I hold this project and all of you dear to my heart and I am extremely glad and grateful to have worked with you and met you on summits/PTGs, as the life memories around Manila are among the best I have from that period. If someday circumstances change, the manila project and its community will be ones I will be very happy to go back to working closely with again. Kind regards, -- Rodrigo Barbieri MSc Computer Scientist

3 2

Openstack instance is unreachable
by GEORGIOS PAPATHANAIL 07 Jul '22

07 Jul '22

I am facing the following issue with my new Openstack installation. The installation is a little bit weird and to elaborate more, I have a controller and a compute node running as VMs in XenServer. The compute node has nested virtualization enabled and uses qemu in order to provision VMs. I have a provider flat network that has a public range of IPs and VMs use this network. I am able to create instances and access them via console, but I can't ping them or ssh even if they have public IPs. I am not very familiar with Xen Server, is there any configuration that is needed (bridges, promisc mode, etc)? -- *George Papathanail* *Associate Researcher* *Department of Applied Informatics* *University of Macedonia*

3 5

Poor I/O performance on OpenStack block device (OpenStack Centos8:Ussuri)
by Vinh Nguyen Duc 07 Jul '22

07 Jul '22

I have a problem with I/O performance on Openstack block device HDD. *Environment:**Openstack version: Ussuri* - OS: CentOS8 - Kernel: 4.18.0-240.15.1.el8_3.x86_64 - KVM: qemu-kvm-5.1.0-20.el8 *CEPH version: Octopus * *15.2.8-0.el8.x84_64* - OS: CentOS8 - Kernel: 4.18.0-240.15.1.el8_3.x86_64 In CEPH Cluster we have 2 class: - Bluestore - HDD (only for cinder volume) - SSD (images, cinder volume) *Hardware:* - Ceph-client: 2x10Gbps (bond) MTU 9000 - Ceph-replicate: 2x10Gbps (bond) MTU 9000 *VM:* - Swapoff - non LVM *Issue*When create VM on Openstack using cinder volume HDD, have really poor performance: 60-85 MB/s writes. And when tests with ioping have high latency. *Diagnostic* 1. I have checked the performance between Compute Host (Openstack) and CEPH, and created an RBD (HDD class) mounted on Compute Host. And the performance is 300-400 MB/s. => So i think the problem is in the hypervisor But when I check performance on VM using cinder Volume SSD, the result equals performance when test RBD (SSD) mounted on a Compute host. 2. I already have to configure disk_cachemodes="network=writeback"(and enable rbd cache client) or test with disk_cachemodes="none" but nothing different. 3. Push iperf3 from compute host to random ceph host still has 20Gb traffic. 4. Compute Host and CEPH host connected to the same switch (layer2). Where else can I look for issues? Please help me in this case. Thank you.

3 4

[large-scale][cinder] backend_native_threads_pool_size option with rbd backend
by Arnaud Morin 07 Jul '22

07 Jul '22

Hey all, Is there any recommendation on the number of threads to use when using RBD backend (option backend_native_threads_pool_size)? The doc is saying that 20 is the default but it should be increased, specially for the RBD driver, but up to which value? Is there anyone tuning this parameter in their openstack deployments? If yes, maybe we can add some recommendations on openstack large-scale doc about it? Cheers, Arnaud.

4 6

[all][TC] Bare rechecks stats update
by Slawek Kaplonski 07 Jul '22

07 Jul '22

Hi, New stats from last 7 days about bare rechecks in each team are available in [1]: +--------------------+---------------+--------------+-------------------+ | Team | Bare rechecks | All Rechecks | Bare rechecks [%] | +--------------------+---------------+--------------+-------------------+ | skyline | 1 | 1 | 100.0 | | OpenStack Charms | 10 | 10 | 100.0 | | kolla | 92 | 92 | 100.0 (was 76.19%)| | sahara | 21 | 21 | 100.0 | | tacker | 5 | 5 | 100.0 | | keystone | 17 | 17 | 100.0 | | OpenStack-Helm | 15 | 15 | 100.0 (was 70%) | | rally | 1 | 1 | 100.0 | | barbican | 2 | 2 | 100.0 | | Telemetry | 3 | 3 | 100.0 | | ec2-api | 5 | 5 | 100.0 | | zun | 2 | 2 | 100.0 | | kuryr | 30 | 32 | 93.75 (was 92.5%) | | cinder | 89 | 102 | 87.25 (was 81.33%)| | Puppet OpenStack | 13 | 15 | 86.67 (was 85.29%)| | horizon | 17 | 20 | 85.0 (was 100%) | | Quality Assurance | 26 | 32 | 81.25 (was 64.44%)| | manila | 70 | 89 | 78.65 (was 74%) | | ironic | 77 | 101 | 76.24 (was 79.66%)| | tripleo | 230 | 318 | 72.33 (was 87.19%)| | swift | 5 | 7 | 71.43 (was 75%) | | glance | 8 | 12 | 66.67 (was 50%) | | designate | 9 | 14 | 64.29 (was 64.71%)| | nova | 10 | 18 | 55.56 (was 84.21%)| | octavia | 6 | 11 | 54.55 (was 16.67%)| | OpenStackSDK | 13 | 25 | 52.0 (was 74.19%) | | neutron | 26 | 54 | 48.15 (was 73.68%)| | heat | 3 | 8 | 37.5 (was 25%) | | requirements | 1 | 3 | 33.33 (was 87.5) | | oslo | 1 | 3 | 33.33 (was 0%) | | Release Management | 0 | 2 | 0.0 | | OpenStackAnsible | 0 | 34 | 0.0 (was 13.64%) | +--------------------+---------------+--------------+-------------------+ There are some teams which made progress since last check (especially *OpenStackAnsible* which had all rechecks done with given some comment at least - thx a lot for that :)) [1] https://etherpad.opendev.org/p/recheck-weekly-summary[1] -- Slawek Kaplonski Principal Software Engineer Red Hat -------- [1] https://etherpad.opendev.org/p/recheck-weekly-summary

1 0

[all][tc] Technical Committee next weekly meeting on 7 July 2022 at 1500 UTC
by Ghanshyam Mann 06 Jul '22

06 Jul '22

Hello Everyone, The technical Committee's next weekly meeting is scheduled for 7 July 2022, at 1500 UTC. If you would like to add topics for discussion, please add them to the below wiki page by Wednesday, 6 July at 2100 UTC. https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting -gmann

1 1

openstackclient: booting from image-created volume w/ delete_on_termination
by Kurt Garloff 06 Jul '22

06 Jul '22

Hi openstack CLI wizards, Having flavors without disks, I want volumes to be created from images on the fly that I can boot from. But I don't want to do bookkeeping for these volumes; they should disappear once the server disappears. On the command line with nova, I can do this: nova boot --nic net-name=$MYNET --key-name $MYKEY --flavor $MYFLAVOR \ --block-device "id=$MYIMGIFD,source=image,dest=volume,size=$MYSIZE,shutdown=remove,bootindex=0" $MYNAME I did not find a way to do this with openstack server create. Is there one? Here's what I tried: --image $MYIMGID --boot-from-volume $MYSIZE works, except that there is no way to specify delete_on_termination --block-device-mapping "sda=$MYIMGID:image:$MYSIZE:true" does complain that no --image or --volume have been passed. The option does not appear to be used for bootable volumes. I have seen a BP (merged with Ussuri) that would allow a PUT call to update delete_on_termination, but I don't see it usable with the openstackclient ... https://specs.openstack.org/openstack/nova-specs/specs/ussuri/implemented/d… Anything obvious I missed? Thanks, -- Kurt Garloff <kurt(a)garloff.de> Cologne, Germany

2 2

Re: Regarding Policy.json entries for glance image update not working for a user
by Adivya Singh 06 Jul '22

06 Jul '22

Hi Brian, Please find the response > 1> i am using Xena release version 24.0.1 > > Now the scenario is line below, my customer wants to have their login > access on setting up the properties of an image to the public. now what i > did is > > 1> i created a role in openstack using the admin credential name as "user" > 2> i assigned that user to a role user. > 3> i assigned those user to those project id, which they want to access as > a user role > > Then i went to Glance container which is controller by lxc and made a > policy.yaml file as below > > root@aio1-glance-container-724aa778:/etc/glance# cat policy.yaml > > "modify_image": "role:user" > > then i went to utility container and try to set the properties of a image > using openstack command > > openstack image set --public <image id> > > and then i got this error > > HTTP 403 Forbidden: You are not authorized to complete publicize_image > action. > > Even when i am trying the upload image with this user , i get the above > error only > > export OS_ENDPOINT_TYPE=internalURL > export OS_INTERFACE=internalURL > export OS_USERNAME=adsingh > export OS_PASSWORD='adsingh' > export OS_PROJECT_NAME=adsingh > export OS_TENANT_NAME=adsingh > export OS_AUTH_TYPE=password > export OS_AUTH_URL=https://<Internal IP of horizon>:5000/v3 > export OS_NO_CACHE=1 > export OS_USER_DOMAIN_NAME=Default > export OS_PROJECT_DOMAIN_NAME=Default > export OS_REGION_NAME=RegionOne > > Regards > Adivya Singh > > > > > On Mon, Jun 13, 2022 at 6:41 PM Alan Bishop <abishop(a)redhat.com> wrote: > >> >> >> On Mon, Jun 13, 2022 at 6:00 AM Brian Rosmaita < >> rosmaita.fossdev(a)gmail.com> wrote: >> >>> On 6/13/22 8:29 AM, Adivya Singh wrote: >>> > hi Team, >>> > >>> > Any thoughts on this >>> >>> H Adivya, >>> >>> Please supply some more information, for example: >>> >>> - which openstack release you are using >>> - the full API request you are making to modify the image >>> - the full API response you receive >>> - whether the user with "role:user" is in the same project that owns the >>> image >>> - debug level log extract for this call if you have it >>> - anything else that could be relevant, for example, have you modified >>> any other policies, and if so, what values are you using now? >>> >> >> Also bear in mind that the default policy_file name is "policy.yaml" (not >> .json). You either >> need to provide a policy.yaml file, or override the policy_file setting >> if you really want to >> use policy.json. >> >> Alan >> >> cheers, >>> brian >>> >>> > >>> > Regards >>> > Adivya Singh >>> > >>> > On Sat, Jun 11, 2022 at 12:40 AM Adivya Singh <adivya1.singh(a)gmail.com >>> > <mailto:adivya1.singh@gmail.com>> wrote: >>> > >>> > Hi Team, >>> > >>> > I have a use case where I have to give a user restriction on >>> > updating the image properties as a member. >>> > >>> > I have created a policy Json file and give the modify_image rule to >>> > the particular role, but still it is not working >>> > >>> > "modify_image": "role:user", This role is created in OpenStack. >>> > >>> > but still it is failing while updating properties with a >>> > particular user assigned to a role as "access denied" and >>> > unauthorized access >>> > >>> > Regards >>> > Adivya Singh >>> > >>> >>> >>>

3 5

[largescale-sig] Next meeting: July 6th, 15utc
by Thierry Carrez 06 Jul '22

06 Jul '22

Hi everyone, The Large Scale SIG will be meeting this Wednesday in #openstack-operators on OFTC IRC, at 15UTC, before taking a break for July and most of August. You can check how that time translates locally at: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20220706T15 Feel free to add topics to the agenda: https://etherpad.openstack.org/p/large-scale-sig-meeting Regards, -- Thierry Carrez

1 1