Reviewed-by: Sheng Yang
Changes:
- Do not check if allocation_state is 'Enabled' in planner if the caller is Root Admin.
- This should let Root Admin create a VM in a disabled Zone.
Reviewed-By: Alex
Changes:
- Reuse the same storagepool where the Volume is ready on each retry of VM deployment until the cluster where the volume is has capacity
- After the cluster is out of capacity, we look in other clusters and find a new storagepool.
- At this point if the volume is recreatable on the new storagepool, depoyment will succeed provided everytyhing else goes through
- But if the volume is not recreatable and its cluster is out of capacity, we will still fail to deploy the VM
Changes:
- Once the HostAllocators have listed suitable hosts, planner should not reshuffle the list since that would lose the prioritization applied by the HostAllocators.
- E.g: HostAllocators chooses that host first which matches the guest OS category. If planner shuffles the list, that preferrence is lost.
Changes:
- Added a two new deployment planners 'UserDispersingPlanner' and 'UserConcentratedPodPlanner' to the DeploymentPlanners
- Planner can be chosen by setting the global config variable 'vm.allocation.algorithm' to either of the following values:
('random', 'firstfit', 'userdispersing', 'userconcentratedpod')
- By default, the value is 'random'. When the value is 'random', FirstFitPlanner is invoked as before that shuffles the resource lists.
- Now Admin can choose whether the deployment heuristic should be applied starting at cluster or pod level. This can be done by using the
global config variable 'apply.allocation.algorithm.to.pods' which is false by default. Thus by default as earlier, planner starts at clusters directly.
'UserConcentratedPodPlanner' changes:
- Earlier to 3.0, FirstFitPlanner used to reorder the clusters in case this heuristic was chosen.
- Now this is done by a separate planner and is applied only when 'vm.allocation.algorithm' is set to this planner
- It reorders the capacity based clusters/pods such that those pods having more number of Running Vms for the given account are tried first.
- Note that this userconcentration is applied only to pods and clusters. Not to hosts or storagepools within a cluster.
'UserDispersingPlanner' changes:
- 'UserDispersingPlanner' reorders the capacity ordered pods and clusters based on number of 'Running' VMs for the given account in ascending order. Aim is to choose thodes pods/clusters first which have less number of Running VMs for the given account
- Admin can provide weights to capacity and user dispersion so that both parameters get considered in reordering the pods/clusters. This can be done by setting
the global config parameter 'vm.user.dispersion.weight'. Default value is 1. Thus if this planner is chosen, by default, ordering will be done only by number of Running Vms, unless the weight is changed.
- HostAlllocators and StoragePoolAllocators also reorder the hosts and pools by ascending order of number of Running VMS/ Ready Volumes respectively for the given account. Thus try to choose that host or pool within a cluster with less number of VMs for the account.
Changes:
- We were ordering clusters based on capacity of the first-fit host found in each cluster. Due to this, there were cases where we deployed VMs to one cluster instead of balancing off within clusters.
- Now we order the list of clusters by aggregate capacity and choose the ones that have enough capacity for the required VM in this order.
- This should balance the load between clusters instead of bombarding one.
Conflicts:
server/src/com/cloud/capacity/dao/CapacityDao.java
server/src/com/cloud/capacity/dao/CapacityDaoImpl.java
Changes:
To make sure migration does not attempt to pick a host that has running VMs more than the max guest VM's limit:
- Changed manual migration to call host allocators to return a list of hosts suitable for migration. Host allocators check for the max guest VM limit.
- Earlier we returned hosts with enough capacity but now Host Allocators make other checks along with capacity. So the list of hosts returned are hosts that have enough capacity AND satisfy all other conditions like host tags, max guests limit etc. Or in other words Allocators dont return the hosts that dont satisfy all conditions even if they have capacity.
-Therefore, now we mark the list of hosts returned for manual migration as 'suitable' hosts instead of 'hasenoughCapacity' in the HostResponse.
- HA migration already calls allocators, so no change is needed there.
Bug 11186- Cannot restart existing VM if the cluster is disabled after the
VM has been created
Changes:
- We should not check for the cluster 'allocation_state' while starting an
existing VM provided it has storage already allocated. But if volumes are
deleted and new storage needs to be allocated, then we will not allow the
VM
start.
- However we should still prohibit adding new VMs in that cluster.
Bug 11186- Cannot restart existing VM if the cluster is disabled after the
VM has been created
Changes:
- We should not check for the cluster 'allocation_state' while starting an
existing VM provided it has storage already allocated. But if volumes are
deleted and new storage needs to be allocated, then we will not allow the VM
start.
- However we should still prohibit adding new VMs in that cluster.
This patch enable redundant virtual routers.
1. To enable this feature, db need to be updated using follow SQL by now(we
would get a UI way later):
UPDATE network_offerings SET redundant_router=1 WHERE guest_type="Virtual" AND
system_only=0;
2. System would try to start up two routers at different hosts. But if there is
only one host in the zone, system would start up two routers on it.
3. The failover part is using keepalived, and connection tracking part is using
conntrackd. There would be one master router and one backup router. The status
of router(master or backup) can be query from the database table domain_router
now. Management server would update the status every 30s by default.
4. The routers for the same zone would use same external NIC(same ip and mac).
The script used for fail-over would ensure only one external NIC present in the
network at any time.
5. Currently management server don't got the ability to stop one of router is
both of them reported as master. The feature is in the todo list.
After two routers start up, disconnect anyone of them, the guest network
shouldn't be affected, and established connection(http, ssh, etc.) should still
works. The fail-over on gateway part should be 3~4 seconds.
Currently the patch works with KVM. Would deal with vmware and XenServer soon.
Changes:
- When the ROOT volume of a VM is found to be READY, changed planner to reuse the pool for every volume(root or data) that is READY and that has a pool not in maintenance and not in avoid state
- If ROOT volume is not ready, we dont care about the DATA disk. Both would get re-allocated.
- When a pool is reused for a ready volume, Planner does not call storagepool allocators. And such volumes are not assigned a pool in the deployment destination returned by the planner. Accordingly StorageManager :: prepare method wont recreate these volumes since they are not mentioned in the destination.
Changes:
- Changed host allocators/planner to use cpu.overprovisioning.factor
- Removed following: while adding a new host, we were setting the total_cpu in op_host_capacity to be actual_cpu * cpu.overprovisioning.factor. Now we set it to actual_cpu.
- ListCapacities response now calculates the total CPU as actual * cpu.overprovisioning.factor (This change does not add anything new - listCapacities was pulling total CPU from op_host_capacity DB earlier which had the cpu.overprovisioning.factor applied already. Now we need to apply it over the DB entry.)
- HostResponse has a new field: 'cpuWithOverprovisioning' that returns the cpu after applying the cpu.overprovisioning.factor
- Db Upgrade 222 to 224 now updates the total_cpu in op_host_capacity to be the actual_cpu for each Routing host.
Changes:
While starting a System VM:
- We check, incase the ROOT volume is READY, if the templateID of the volume matches the SystemVM's template.
- If it does not match, we update the volumes' templateId and ask deployment planner to reassign a pool to this volume even if it is READY.
In general:
- If a root volume is READY, we remove its entry from the deploydestination before calling storagemanager :: prepare()
- StorageManager creates a volume if a pool is assigned to it in deploydestination passed to it.
- If a volume has no pool assigned to it in deploydestination, it means the volume is ready and has a pool already allocated to it.
Changes:
- Planner must reassign the storage pool if the template id for system vms has changed. StorageManager must then recreate the volume if the volume has been
reassigned. This is needed to do automatic update of the system template.
- Added a new flag 'allocation_state' to zone,pod,cluster and host
- The possible values for this flag are 'Enabled' or 'Disabled'
- When a new zone,pod,cluster or host is added, allocation_state is 'Disabled' by default.
- For existing zone,pod,cluster or host, the state is 'Enabled'.
- All Add/Update/List commands for each of zone,pod,cluster or host can now take a new parameter 'allocationstate'
- If 'allocation_state' is 'Disabled', Allocators skip that zone or pod or cluster or pod.
- For a root admin, ListZones lists all zones including the 'Disabled' zones. But for any other user, the 'Disabled' zones are not included in the response.
- For any usecase that creates/deploys/adds/registers a resource and takes in zone as parameter, now we check if the Zone is 'Disabled'. If yes then the operation cannot be performed by a user other than root-admin. Add volume, snapshot, templates are examples of this usecase.
- To enable the root admin to test a particular pod/cluster/host, deployVM command takes in 'host_id' parameter that can be passed in only by root admin.
If this parameter is passed in by the admin, allocators do not search for hosts and use that host only. StoragePools are searched in the cluster of that host.
If VM cannot be deployed to that host, allocators and deployVM fails without retrying
However in case of createVolume from Snapshot, there is no VM associated. So VM passed is null and this can cause a NPE.
Allocators hardly use the VM instance. LocalStoragePoolAllocator was mainly using it for checking if host has capacity. But it need not do this check, since that is done by HostAllocators anyway.
So removing the use of VM in StoragePoolAllocators.
- After applying the User concentrated Pod heuristic, order of clusters within a pod (based on capacity) should remain intact
- Only change should be clusters of which pod to be considered first
Bug 7723 - merge or re-write host tagging into master / 2.2
Bug 7627 - Need more logging for Allocators
Bug 8317 - Add better resource allocation failure messages
Changes for Deployment Planner to use host and storagePool allocators to find deployment destination.
Also has the changes for host tag feature.
Improved the logging for allocators.
