Portal:Cloud VPS/Admin
Documentation of admin procedures for Cloud VPS. See Help:Cloud VPS for user-facing documentation.
Admin permissions
Performing admin procedures requires having admin permissions on Cloud VPS. There is not a single "admin" flag, but a set of interrelated permissions you can be granted. These are described in detail in the page Cloud roots and Cloud admins.
Deployments
Cloud VPS is made of two customized OpenStack deployments: eqiad1 (production) and codfw1dev (testing). Both deployments are managed by the Cloud Services team together with other WMF staff members and with the help of external technical volunteers.
Useful links
- Cloud VPS Runbooks
- Who to ask
- Packaging software
- Aptly: A guide to setting up a local debian repo for installing your own .deb packages within a labs project
- Disabling an account
- Logout all users
- Suspending instances
- Using nova-manage (OpenStack cli tools)
- DNS with Designate + PDNS
- LDAP
- Testing deployment (Cloud VPS test environment, currently hosted in codfw)
- cloudweb1003: WMCS web UIs - Horizon and Striker
All subpages
- Alerts
- Auth logging
- Backy2
- Bootstrap
- Ceph
- Cinder backups
- Cloud roots and Cloud admins
- Cloudinfra
- Cloudlb
- Cookbooks
- DB
- DNS
- DNS/Designate
- Deployment confidence checklist
- Deployments
- Devstack
- Devstack magnum
- Devstack magnum/PAWS dev devstack
- Devstack magnum/Stable wallaby
- Devstack magnum/Stable xena
- Galera
- Haproxy
- Hiera
- Horizon
- Host aggregates
- IPv6/initial deploy
- Infrastructure projects
- Instance backups
- Interest matrix
- Keepalived
- Local customization and hacks
- Magnum
- Maintenance
- Maintenance notification
- Monitoring
- Network
- Network/Tests
- Network and Policy
- Neutron LVS
- Nova-manage
- Openstack cli
- Openstack roles and policy
- Openstack source
- Openstack upgrade
- Procedures and operations
- Projects
- Projects/Potentially unmaintained
- Projects lifecycle
- Puppet testing
- RabbitMQ
- Resizing instances
- Runbooks
- Runbooks/CephClusterAllMonsDown
- Runbooks/CephClusterInError
- Runbooks/CephClusterInUnknown
- Runbooks/CephClusterInWarning
- Runbooks/CephSlowOps
- Runbooks/Check for snapshots leaked by cinder backup agent
- Runbooks/Check unit status of backup cinder volumes
- Runbooks/Check unit status of backup vms
- Runbooks/Check unit status of maintain dbusers
- Runbooks/Cloud VPS alert Puppet failure on
- Runbooks/Designate record leaks
- Runbooks/GaleraClusterOutOfSync
- Runbooks/GaleraClusterSizeMismatch
- Runbooks/GaleraDown
- Runbooks/GaleraNoRecentWrites
- Runbooks/GaleraNodeOutOfSync
- Runbooks/GaleraNotEnabled
- Runbooks/Kernel panic
- Runbooks/MetricsinfraAlertmanagerDown
- Runbooks/NodeDown
- Runbooks/NodeDownForLong
- Runbooks/NovaFullstackStaleStats
- Runbooks/NovafullstackSustainedFailures
- Runbooks/NovafullstackSustainedLeakedVMs
- Runbooks/OpenstackAPIResponse
- Runbooks/OpenstackAPIs
- Runbooks/PuppetCertificateAboutToExpire
- Runbooks/PuppetStaleCertificates
- Runbooks/RabbitmqNetworkPartition
- Runbooks/Refresh sssd caches
- Runbooks/Runbook template
- Runbooks/SystemdUnitDown
- Runbooks/SystemdUnitDownForLong
- Runbooks/TfInfraTestApplyFailed
- Runbooks/TfInfraTestDestroyFailed
- Runbooks/TooManyCloudcontrolsDown
- Runbooks/TooManyCloudgwsDown
- Runbooks/TooManyCloudnetsDown
- Runbooks/TooManyCloudvirtsDown
- SSL certificate
- Service accounts
- Skill matrix
- Sshd
- Testing deployment
- Troubleshooting
- Trove
- Users not in bastion project
- VM flavors
- VM images
- Virt capacity
- Web proxy
- notes/Magnum POC for release engineering
- notes/Meltdown Response
- notes/Monitoring
- notes/NAT loophole
- notes/NAT loophole/NFS
- notes/Neutron Migration
- notes/Neutron ideal model
- notes/Neutron ideal model/meeting-2018-11-29
- notes/Puppet Possibilities
- notes/Service predictions for cross realm situation
- notes/keystone
- tofu-infra