Increase VDFS project quotas - Githubissues

lsst-uk / somerville-operations

User issue reporting and tracking for the Somerville Cloud

0 stars 0 forks source link

Increase VDFS project quotas #179

Closed astrodb closed 4 months ago

astrodb commented 4 months ago

With the new hardware money now confirmed, VDFS will need small increases to their quotas. Please update the VDFS project quotas to:

vCPUs: 24 RAM: 96GB Volumes: 50 Volume Storage: 500TB Floating IPs: 4

GregBlow commented 4 months ago

current:

(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota show vdfs
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Field                | Value                                                                                                                                                                                      |
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| backup-gigabytes     | -1                                                                                                                                                                                         |
| backups              | -1                                                                                                                                                                                         |
| cores                | 12                                                                                                                                                                                         |
| fixed-ips            | -1                                                                                                                                                                                         |
| floating-ips         | 2                                                                                                                                                                                          |
| gigabytes            | 368640                                                                                                                                                                                     |
| gigabytes_ceph-hdd   | -1                                                                                                                                                                                         |
| gigabytes_ceph-ssd   | 1000                                                                                                                                                                                       |
| gigabytes_nvme       | -1                                                                                                                                                                                         |
| groups               | 10                                                                                                                                                                                         |
| injected-file-size   | -1                                                                                                                                                                                         |
| injected-files       | -1                                                                                                                                                                                         |
| injected-path-size   | 255                                                                                                                                                                                        |
| instances            | 10                                                                                                                                                                                         |
| key-pairs            | -1                                                                                                                                                                                         |
| location             | Munch({'cloud': '', 'region_name': 'RegionOne', 'zone': None, 'project': Munch({'id': 'ff3e2de6a0b844d581bcd4335c18d2a4', 'name': 'admin', 'domain_id': None, 'domain_name': 'Default'})}) |
| networks             | 100                                                                                                                                                                                        |
| per-volume-gigabytes | -1                                                                                                                                                                                         |
| ports                | 500
   |
| project              | 5d1247c0315c40cfa13f71ac2fee0764
   |
| project_name         | VDFS
   |
| properties           | 128
   |
| ram                  | 24576
   |
| rbac_policies        | 10
   |
| routers              | 10
   |
| secgroup-rules       | 100
   |
| secgroups            | 10
   |
| server-group-members | 10
   |
| server-groups        | 10
   |
| snapshots            | 20
   |
| snapshots_ceph-hdd   | -1
   |
| snapshots_ceph-ssd   | -1
   |
| snapshots_nvme       | -1
   |
| subnet_pools         | -1
   |
| subnets              | 100
   |
| volumes              | 10
   |
| volumes_ceph-hdd     | -1
   |
| volumes_ceph-ssd     | -1
   |
| volumes_nvme         | -1
   |
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

GregBlow commented 4 months ago

https://git.ecdf.ed.ac.uk/lsstuk-somerville/openstack-config/-/merge_requests/111

GregBlow commented 4 months ago

currently automatic deployment blocked by ansible failing:

Greg Blow 41 minutes ago trying to apply quota changes using ansible and getting: TASK [stackhpc.openstack.os_projects : List OpenStack domains] ** fatal: [localhost]: FAILED! => {"changed": false, "cmd": ". /home/stack/kayobe/src/openstack-config/ansible/somerville-config-venv/bin/activate && openstack --os-project-domain-id='' --os-user-domain-id='' --os-project-domain-name='Default' --os-user-domain-name='Default' --os-project-id='' --os-project-name='admin' --os-username='admin' --os-password='' --os-auth-url='http://10.19.3.200:5000/' --os-cacert='' --os-interface=admin domain list -f json -c Name -c ID\n", "delta": "0:00:01.439531", "end": "2024-07-19 13:06:33.656971", "msg": "non-zero return code", "rc": 1, "start": "2024-07-19 13:06:32.217440", "stderr": "admin endpoint for identity service in RegionOne region not found", "stderr_lines": ["admin endpoint for identity service in RegionOne region not found"], "stdout": "", "stdout_lines": []} (edited) 13 replies

Mark Holliman 40 minutes ago login credential file is out of date?

Mark Holliman 37 minutes ago check you're in the right project?

Greg Blow 37 minutes ago I don't know. admin endpoint for identity service in Regionone region not found doesn't sound like a credentials error.

Mark Holliman 34 minutes ago Google search shows a lot of things all related to misconfigured env settings, or the service being offline

Mark Holliman 32 minutes ago Just checked, and keystone is listening on 10.19.3.200:5000, visible from admin node

Greg Blow 31 minutes ago references the venv at /home/stack/kayobe/src/openstack-config/ansible/somerville-config-venv/bin/activate

Greg Blow 30 minutes ago git repo there has image.png

image.png

Greg Blow 17 minutes ago changing os-interface from admin to internal causes the failed command to return properly, rather than erroring out.

Greg Blow 16 minutes ago --os-interface Select an interface type. Valid interface types: [admin, public, internal]. default=public, (Env: OS_INTERFACE)

Greg Blow 13 minutes ago (openstack-config) [stack@sv-admin-0 openstack-config]$ openstack endpoint list | grep keystone | 12a633c81bbd412692f80c4600cb3008 | RegionOne | keystone | identity | True | public | https://somerville.ed.ac.uk:5000/ | | ab49d5060d4746e8b8ce5af5cd0902d0 | RegionOne | keystone | identity | True | internal | http://10.19.3.200:5000/

Greg Blow 12 minutes ago https://bugs.launchpad.net/ceilometer/+bug/1981207

LaunchpadLaunchpad Bug #1981207 “Ceilometer depends on deprecated admin endpoint of...” : Bugs : Ceilometer Error occurs on the Devstack (Release Yoga) using the ceilometer plugin: --- local.conf --- ... enable_plugin ceilometer https://opendev.org/openstack/ceilometer stable/yoga ...

After investigating the ceilometer notification-agent log, keystoneauth1 does not found the admin entity

Show more

Greg Blow 11 minutes ago I think it may be a problem of trying to use the keystone endpoint with the admin interface.

Greg Blow 11 minutes ago (listing images with the admin interface works)

GregBlow commented 4 months ago

(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota set --volumes=50 vdfs
(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota set --floating-ips=4 vdfs
(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota set --cores=24 vdfs
(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota set --gigabytes=512000 vdfs
(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota set --ram=98304 vdfs

GregBlow commented 4 months ago

(openstack-config) [stack@sv-admin-0 openstack-config]$ openstack quota show vdfs
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Field                | Value
   |
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| backup-gigabytes     | -1
   |
| backups              | -1
   |
| cores                | 24
   |
| fixed-ips            | -1
   |
| floating-ips         | 4
   |
| gigabytes            | 512000
   |
| gigabytes_ceph-hdd   | -1
   |
| gigabytes_ceph-ssd   | 1000
   |
| gigabytes_nvme       | -1
   |
| groups               | 10
   |
| injected-file-size   | -1
   |
| injected-files       | -1
   |
| injected-path-size   | 255
   |
| instances            | 10
   |
| key-pairs            | -1
   |
| location             | Munch({'cloud': '', 'region_name': 'RegionOne', 'zone': None, 'project': Munch({'id': 'ff3e2de6a0b844d581bcd4335c18d2a4', 'name': 'admin', 'domain_id': None, 'domain_name': 'Default'})}) |
| networks             | 100
   |
| per-volume-gigabytes | -1
   |
| ports                | 500
   |
| project              | 5d1247c0315c40cfa13f71ac2fee0764
   |
| project_name         | VDFS
   |
| properties           | 128
   |
| ram                  | 98304
   |
| rbac_policies        | 10
   |
| routers              | 10
   |
| secgroup-rules       | 100
   |
| secgroups            | 10
   |
| server-group-members | 10
   |
| server-groups        | 10
   |
| snapshots            | 20
   |
| snapshots_ceph-hdd   | -1
   |
| snapshots_ceph-ssd   | -1
   |
| snapshots_nvme       | -1
   |
| subnet_pools         | -1
   |
| subnets              | 100
   |
| volumes              | 50
   |
| volumes_ceph-hdd     | -1
   |
| volumes_ceph-ssd     | -1
   |
| volumes_nvme         | -1
   |
+----------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

GregBlow commented 4 months ago

Changes made. Please verify and close?

astrodb commented 4 months ago

Looks good. Might want to open an issue on the broken Kayobe config deployment to track it.