.. This work is licensed under a Creative Commons Attribution 3.0 Unported License. http://creativecommons.org/licenses/by/3.0/legalcode ============================== Detach and attach boot volumes ============================== https://blueprints.launchpad.net/nova/+spec/detach-boot-volume It is sometimes useful for a cloud user to be able to detach and attach the boot volume of an instance when the instance is not running. Currently nova does not allow this at all and some operations assume it does not happen. This spec proposes allowing the detach and attach of boot volumes when an instance is powered off or shelved and adding safeguards to ensure it is safe. Problem description =================== There is an implicit assumption in the nova code that an instance always has a cinder boot volume attached or an ephemeral boot disk. Nova allows cinder volumes to be detached and attached at any time, but the detach operation is limited to exclude boot volumes to preserve the above assumption[1]. This limitation means it is not possible to change the boot volume attached to an instance except by deleting the instance and creating a new one. However, it is safe to change boot volume attachments when an instance is not running, so preventing this altogether is unnecessarily limiting. There are use cases that require a boot volume to be detached when an instance is not running, so we propose relaxing the inherent assumption to say that a boot volume attachment can be changed when an instance is powered off or shelved. To ensure safety we can prevent it being started or unshelved without a boot volume. Use Cases --------- The first use case is based on a disaster recovery scenario. In this scenario a system of VMs attached to a network and using persistent volumes at site A is executing an online application. To provide a remote failure recovery capability the data on the persistent volumes is being replicated to volumes at remote site B. The persistent volumes include boot volumes. The use case is the following: As a cloud user I want to be able to failover my application to a remote site with minimal down time and the assurance that the remote site is able to take over. The ability to detach and attach boot volumes is required for this use case as implemented by the following failover from site A to site B: 1. Build the virtual infrastructure in advance at site B and check that the new infrastructure is complete, correctly configured and operable. Then shelve the instances and detach the disks. This infrastructure is now ready to take over when supplied with replica disks. 2. Set up continuous replication of disks from site A to site B 3. The failover procedure: stop replication to site B; attach replica disks to the shelved instances; unshelve the instances. The outline above shows that the virtual infrastructure at site B is built in advance and is kept in a dormant state. The volumes are detached and kept up to date as replicas of the volumes at site A, to be swapped back in later. This satisfies the requirements of the use case: Firstly, the build of the infrastructure, including instances that will receive replica volumes, can be done and checked to be correct before performing the failover. This gives a higher level of assurance that the switchover will be successful. Secondly, by removing the virtual infrastructure build from the critical path of the failover, the down time caused by the failover is minimised. A bug registered against nova describes further use cases (see [2]). An example is the following: As a user I want to run a VM with a windows instance. I will take snapshots of the boot volume from time to time. I may want to revert to a snapshot. If I delete my instance and recreate it from the snapshot I will incur additional costs from licensing and may invalidate my license. Proposed change =============== This change assumes that only cinder volumes can be dynamically changed in this way. We will not support detaching ephemeral disks. Volume backed instances are always offloaded after a period of time[3] when shelved, so the instance will not be on a host. As a result the implementation will be to change the recorded block device mapping and register the attachment/detachment with cinder. The usual detach volume API call will be used to detach the boot volume. The guard on this call will be changed to allow the detach if the instance is powered off or shelved_offloaded. When a boot volume is detached, we will set the root block device mapping(boot_index=0) with ``volume_id=None``, meaning that it's not attached to any volume. A new microversion will be added to the attach volume API. A new ``is_root`` parameter will be allowed for requests with the new microversion or greater, indicating that the user is trying to attach a root volume. The attachment with this parameter will only be allowed if the instance is powered off or shelved_offloaded and it has a "no volume" block device mapping for the root device. There are some specific considerations for detaching/attaching a root volume instances, here is what will happen: - Detach: Delete the volume attachment referenced via the BDM.attachment_id field and null out the BDM.attachment_id and BDM.volume_id fields (save those changes to the DB). At that point the old root volume is made 'available' again. For shelved instances, this will be handled by API service. For, stopped instance, this will be handled by nova-compute service, thus the compute service version will be bumped in order to represent that detach root volume for deleted instance is supported. - Attach: Find the root BDM via BlockDeviceMappingList.root_bdm(); at this point the root BDM has a null attachment_id and volume_id. Create a volume attachment record for the new root volume [4] and then update the BDM's attachment_id and volume_id fields and save those to the DB. The normal error conditions around volume attach will be in play, i.e. you can't attach a volume that is already in-use unless it's a multiattach volume. The start and unshelve operation will be guarded with a check for the "no volume" block device mapping. An instance will not be allowed to start or unshelve when its boot volume has been detached unless another has been attached in its place. For dettach/attach volume for powered off instances, This would involve affecting the connection to the hypervisor on the compute node, thus will depend on the ability of compute drivers. We are now aware of that ``libvirt``, ``vmaware`` and ``xen`` driver will be capable of doing this. The feature support matrix will be appropriately updated for this feature. There is a race condition identified in this bug [5] between volume operations and instance state changes. The same race condition will exist between the boot volume detach and the unshelve operations until that bug is fixed. That bug will be addressed by spec [6]. Alternatives ------------ One alternative is simply not to allow a boot volume to be detached. This implies that root devices can only be changed by deleting and recreating an instance. Currently many devices on an instance can be added and removed dynamically. Another alternative is to be more general by allowing any type of boot device to be removed and any type added. This would include images on local ephemeral disks, snapshots and volumes. Because this goes beyond the existing volume API this generalization would suggest the need for a new API. This is not needed to satisfy the use cases provided so we propose restricting this behavior to the existing APIs. Another alternative is to only allow boot volumes to be swapped in a single operation. This retains the assumption that an instance always has a volume (except during the operation) but removes some flexibility. In the disaster recovery use case an instance could be shelved and its boot volume detached. If the instance must have a volume at all times this will require a second volume (besides the replica) for each instance that is not being used. This is wasteful of resources. Data model impact ----------------- None REST API impact --------------- Add a new microversion for attach volume REST API to allow passing ``is_root`` as a parameter. An attempt to detach a boot volume currently always returns the error: "Can't detach root device volume (HTTP: 403)" This will change in the case of an instance being in stopped or shelved_offloaded state to allow the detach. An attempt to start or unshelve an instance that has a missing boot volume because it has been detached will return an error: "Can't unshelve instance without a root device volume (HTTP: 403)" These error changes will also require an API micro version increment. Security impact --------------- None. Notifications impact -------------------- None. Other end user impact --------------------- The python-novaclient and python-openstackclient will be updated to support the new capability. Performance Impact ------------------ None. Other deployer impact --------------------- None. Developer impact ---------------- None. Upgrade impact -------------- Compute service version will be bumped to represent that the feature for detach_volume flow for deleted instance is supported. Implementation ============== Assignee(s) ----------- Primary assignee: Zhenyu Zheng Work Items ---------- This spec will build on the ground work of [7]. The following changes are part of this spec. - Add "no volume" block device mapping utility methods to indicate a boot device has been removed. These will create the "no volume" block device mapping setting the ``volume_id`` field to ``None`` and inspect the mapping for a volume that is not present. - Extend methods to detach volumes for stopped and shelved_offloaded instances to deal with boot volume and "no volume" block device mapping. Add a new microversion to attach volume API to indicate that the specified volume is a root volume. - Add guard in API for "no volume" mapping before start and unshelving an instance. - Change conditional guard on compute api to allow detach of boot device when instance is stopped or shelved_offloaded. Dependencies ============ This spec extends the volume operations enabled by [7]. There is a parallel (but not dependant) spec [6] that addresses bug [5]. That spec is not required for this one, but it is worth noting that this feature will benefit from the general bug fix dealt with there. Testing ======= All the existing volume operations have both unit tests and system tests. The changes described here can be covered in nova by unit tests. We will also add system tests to tempest after the changes are made to ensure coverage of the new use cases for the detach and attach operations. Documentation Impact ==================== Document when a root device volume can be detached and attached. Feature support matrix will be updated about this capability. References ========== [1] Check for root volume when doing detach https://github.com/openstack/nova/blob/aa9f9448c9cf77bb1e55aa0cde5e7f9c4e0157c4/nova/api/openstack/compute/volumes.py#L434 [2] Add capability to detach root device volume of an instance, when in shutoff state. https://bugs.launchpad.net/nova/+bug/1396965 [3] shelved_offload_time config option https://docs.openstack.org/nova/latest/configuration/config.html#DEFAULT.shelved_poll_interval [4] Cinder attachment create https://github.com/openstack/nova/blob/85b36cd2f82ccd740057c1bee08fc722209604ab/nova/volume/cinder.py#L710 [5] Volume operations should set task state. https://bugs.launchpad.net/nova/+bug/1275144 [6] https://blueprints.launchpad.net/nova/+spec/avoid-parallel-conflicting-api-operations [7] Spec for volume-ops-when-shelved (Completed in Mitaka) https://blueprints.launchpad.net/nova/+spec/volume-ops-when-shelved History ======= .. list-table:: Revisions :header-rows: 1 * - Release Name - Description * - Mitaka - Introduced * - Newton - Re-proposed. * - Ocata - Re-proposed. * - Stein - Re-proposed.