cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dave Hoffman <david.hoff...@appcore.com>
Subject Re: SystemVM attempts to use incorrect hypervisor on router creation after host tags have been applied
Date Mon, 07 Jul 2014 14:30:00 GMT
I am aware that it cannot find a VMware host that is tagged Xen.  That 
is intentional.  I have 2 Xen clusters that are able to run the cloud 
router but, Cloudstack is still trying to use vmware for the hypervisor 
to launch the template.

I am wondering if there is not logic to use all hypervisors in a cloud 
and Cloudstack assumes if you launch a vm on a certain hypervisor, the 
cloud router MUST be launched on the same hypervisor.

If that's the case, I'm guessing we may have to remove the vmware 
system template and test if that works as XenServer will then be the 
only choice.

On Thu 03 Jul 2014 11:31:36 PM CDT, Koushik Das wrote:
> As per the logs it is not able to find any host with tag=Xen under cluster 7. Check if
the host is tagged correctly
>
> 2014-07-03 04:56:12,611 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Looking for hosts in dc: 1  pod:1
 cluster:7
> 2014-07-03 04:56:12,611 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Looking for hosts having tag specified
on SvcOffering:Xen
> 2014-07-03 04:56:12,613 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Hosts with tag 'Xen' are:[]
>
> -----Original Message-----
> From: Dave Hoffman [mailto:david.hoffman@appcore.com]
> Sent: Thursday, 3 July 2014 9:49 PM
> To: users@cloudstack.apache.org
> Subject: SystemVM attempts to use incorrect hypervisor on router creation after host
tags have been applied
>
> Hello,
>
> We have a mixed environment with XenServer and Vmware.  We are trying to force the system
vm to launch on to Xen but, when a new network is created with the first vm created being
a vmware template, Cloudstack tries to deploy a vmware cloud router.
>
> If the first vm created is on XenServer, the router creates on XenServer without issue.
>
> Manually create a network does not create a router until a vm is launched so same issue
applies on first vm creation
>
> All hosts are tagged appropriately - Xen for XenServer and VMware for VMware System Offering
For Software Router has a host tag of Xen
>
> Below is the log showing the process of the failed creation.
>
>
> 2014-07-03 04:56:12,482 DEBUG
> [network.router.VirtualNetworkApplianceManagerImpl]
> (Job-Executor-116:job-31669) Allocating the domR with the hypervisor type VMware
> 2014-07-03 04:56:12,485 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Allocating entries for VM:
> VM[DomainRouter|r-35772-VM]
> 2014-07-03 04:56:12,490 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Allocating nics for VM[DomainRouter|r-35772-VM]
> 2014-07-03 04:56:12,491 DEBUG [cloud.network.NetworkManagerImpl]
> (Job-Executor-116:job-31669) Allocating nic for vm VM[DomainRouter|r-35772-VM] in network
Ntwk[633|Guest|6] with requested profile NicProfile[0-0-null-10.50.1.1-vlan://2695
> 2014-07-03 04:56:12,509 DEBUG [cloud.network.NetworkModelImpl]
> (Job-Executor-116:job-31669) Service SecurityGroup is not supported in the network id=633
> 2014-07-03 04:56:12,511 DEBUG [cloud.network.NetworkManagerImpl]
> (Job-Executor-116:job-31669) Allocating nic for vm VM[DomainRouter|r-35772-VM] in network
Ntwk[202|Control|3] with requested profile null
> 2014-07-03 04:56:12,521 DEBUG [cloud.network.NetworkManagerImpl]
> (Job-Executor-116:job-31669) Allocating nic for vm VM[DomainRouter|r-35772-VM] in network
Ntwk[200|Public|1] with requested profile NicProfile[0-0-null-205.172.1.176-vlan://1001
> 2014-07-03 04:56:12,529 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Allocaing disks for VM[DomainRouter|r-35772-VM]
> 2014-07-03 04:56:12,540 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Allocation completed for VM:
> VM[DomainRouter|r-35772-VM]
> 2014-07-03 04:56:12,557 DEBUG
> [network.router.VirtualNetworkApplianceManagerImpl]
> (Job-Executor-116:job-31669) Lock is released for network id 633 as a part of router
startup in Dest[Zone(Id)-Pod(Id)-Cluster(Id)-Host(Id)-Storage(Volume(Id|Type-->Pool(Id))]
> :
> Dest[Zone(1)-Pod(1)-Cluster(7)-Host(26)-Storage(Volume(37515|ROOT-->Pool(268))]
> 2014-07-03 04:56:12,557 DEBUG
> [network.router.VirtualNetworkApplianceManagerImpl]
> (Job-Executor-116:job-31669) Starting router VM[DomainRouter|r-35772-VM]
> 2014-07-03 04:56:12,564 DEBUG [cloud.capacity.CapacityManagerImpl]
> (Job-Executor-116:job-31669) VM state transitted from :Stopped to Starting with event:
StartRequestedvm's original host id: null new host
> id: null host id before state transition: null
> 2014-07-03 04:56:12,564 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Successfully transitioned to start state for VM[DomainRouter|r-35772-VM]
reservation id = fad64dad-8aee-47ce-bff3-e669ed246e8d
> 2014-07-03 04:56:12,564 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Successfully transitioned to start state for VM[DomainRouter|r-35772-VM]
reservation id = fad64dad-8aee-47ce-bff3-e669ed246e8d
> 2014-07-03 04:56:12,574 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Trying to deploy VM, vm has dcId: 1 and
> podId: null
> 2014-07-03 04:56:12,575 DEBUG [cloud.vm.VirtualMachineManagerImpl]
> (Job-Executor-116:job-31669) Deploy avoids pods: null, clusters: null,
> hosts: null
> 2014-07-03 04:56:12,579 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) DeploymentPlanner allocation algorithm:
> userconcentratedpod_random
> 2014-07-03 04:56:12,579 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Trying to allocate a host and storage pools from dc:1, pod:null,cluster:null,
requested cpu: 500, requested ram:
> 134217728
> 2014-07-03 04:56:12,579 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Trying to allocate a host and storage pools from dc:1, pod:null,cluster:null,
requested cpu: 500, requested ram:
> 134217728
> 2014-07-03 04:56:12,579 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Is ROOT volume READY (pool already
> allocated)?: No
> 2014-07-03 04:56:12,579 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Searching all possible resources under this
> Zone: 1
> 2014-07-03 04:56:12,583 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Listing clusters in order of aggregate capacity, that have
(atleast one host with) enough CPU and RAM capacity under this Zone: 1
> 2014-07-03 04:56:12,586 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) CPUOverprovisioningFactor considered: 1.5
> 2014-07-03 04:56:12,590 DEBUG [cloud.deploy.UserConcentratedPodPlanner]
> (Job-Executor-116:job-31669) Applying UserConcentratedPod heuristic for
> account: 1
> 2014-07-03 04:56:12,593 DEBUG [cloud.deploy.UserConcentratedPodPlanner]
> (Job-Executor-116:job-31669) Reordering cluster list as per pods ordered by user concentration
> 2014-07-03 04:56:12,608 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Cluster: 4 has HyperVisorType that does not match the VM,
skipping this cluster
> 2014-07-03 04:56:12,610 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Checking resources in Cluster: 7 under Pod: 1
> 2014-07-03 04:56:12,611 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Looking for hosts in dc: 1  pod:1
 cluster:7
> 2014-07-03 04:56:12,611 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Looking for hosts having tag specified
on SvcOffering:Xen
> 2014-07-03 04:56:12,613 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Hosts with tag 'Xen' are:[]
> 2014-07-03 04:56:12,614 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) FirstFitAllocator has 0 hosts to
check for allocation: []
> 2014-07-03 04:56:12,617 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Found 0 hosts for allocation after
prioritization: []
> 2014-07-03 04:56:12,618 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Looking for speed=500Mhz, Ram=128
> 2014-07-03 04:56:12,618 DEBUG [allocator.impl.FirstFitAllocator]
> (Job-Executor-116:job-31669 FirstFitRoutingAllocator) Host Allocator returning 0 suitable
hosts
> 2014-07-03 04:56:12,618 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) No suitable hosts found
> 2014-07-03 04:56:12,618 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) No suitable hosts found under this Cluster: 7
> 2014-07-03 04:56:12,618 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) No suitable hosts found under this Cluster: 7
> 2014-07-03 04:56:12,620 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Cluster: 1 has HyperVisorType that does not match the VM,
skipping this cluster
> 2014-07-03 04:56:12,620 DEBUG [cloud.deploy.FirstFitPlanner]
> (Job-Executor-116:job-31669) Could not find suitable Deployment Destination for this
VM under any clusters, returning.
> 2014-07-03 04:56:12,638 DEBUG [cloud.capacity.CapacityManagerImpl]
> (Job-Executor-116:job-31669) VM state transitted from :Starting to Stopped with event:
OperationFailedvm's original host id: null new host
> id: null host id before state transition: null
>
> --
> Dave Hoffman
> Systems Administrator | Appcore
>
> Office +1.800.735.7104
> Direct +1.515.612.7808
> dave.hoffman@appcore.com | www.appcore.com

--
Dave Hoffman
Systems Administrator | Appcore

Office +1.800.735.7104
Direct +1.515.612.7808
dave.hoffman@appcore.com | www.appcore.com


Mime
View raw message