was discussed on the mailing list as a useful debugging tool, currently
the log prints the DB id of the agent, which makes admins have to look
it up to know where the Command was run.
When running DatabaseUpgradeChecker as a standalone program _dao will not
be injected. Still create an instance of VersionDaoImpl in constructor
and when DatabaseUpgradeChecker is ran in the mgmt server it will be
overwritten by the injected value.
These changes are a joint effort between Edison and I to refactor some
of the code around snapshotting VM volumes and creating
templates/volumes from VM volume snapshots. In general, we were working
towards allowing PrimaryDataStoreDrivers to create snapshots on primary
storage and not requiring the snapshots to be transferred to secondary
storage.
High level changes:
-Added uuid to NfsTO, SwiftTO & S3TO to cut down on the requirement of
PrimaryDataStoreTO and ImageStoreTO which don't really serve much of a
purpose
-Initial work towards enable reverting VM volume from snapshots
-Added hypervisor commands for introducing and forgetting new hypervisor
objects (snapshots, templates & volumes)
Signed-off-by: Edison Su <sudison@gmail.com>
ACS is now comprised of a hierarchy of spring application contexts.
Each plugin can contribute configuration files to add to an existing
module or create it's own module.
Additionally, for the mgmt server, ACS custom AOP is no longer used
and instead we use Spring AOP to manage interceptors.
The managed context framework provides a simple way to add logic
to ACS at the various entry points of the system. As threads are
launched and ran listeners can be registered for onEntry or onLeave
of the managed context. This framework will be used specifically
to handle DB transaction checking and setting up the CallContext.
This framework is need to transition away from ACS custom AOP to
Spring AOP.
Various classes are using member injection to inject extensible objects.
Really those object should come from an AdapterList that is injected in.
This patch switches the code to use setter injection that will later allow
spring to inject an AdapterList or something similar to allow
extensibility.
Also fixed existing bugs for the API:
* corrected action event to be VOLUME.UPDATE (was VOLUME.ATTACH)
* all parameters to update, should be optional - fixed that. If nothing is specified, the db object will remain with its original fields
Initial patch for VXLAN support.
Fully functional, hopefully, for GuestNetwork - AdvancedZone.
Patch Note:
in cloudstack-server
- Add isolation method VXLAN
- Add VxlanGuestNetworkGuru as plugin for VXLAN isolation
- Modify NetworkServiceImpl to handle extended vNet range for VXLAN isolation
- Add VXLAN isolation option in zoneWizard UI
in cloudstack-agent (kvm)
- Add modifyvxlan.sh script that handle bridge/vxlan interface manipulation script
-- Usage is exactly same to modifyvlan.sh
- BridgeVifDriver will call modifyvxlan.sh instead of modifyvlan.sh when VXLAN is used for isolation
Database changes:
- No change in database structure.
- VXLAN isolation uses same tables that VLAN uses to store vNet allocation status.
Known Issue and/or TODO:
- Some resource still says 'VLAN' in log even if VXLAN is used
- in UI, "Network - GuestNetworks" dosen't display VNI
-- VLAN ID field displays "N/A"
- Documentation!
Signed-off-by : Toshiaki Hatano <haeena@haeena.net>
CLOUDSTACK-4457:
CLOUDSTACK-4459:
harden kvm getvolume. It's possible that one volume created on other kvm host, won't show up on another host, try more times to refresh storage pool if volume won't shown up
Conflicts:
engine/storage/integration-test/test/org/apache/cloudstack/storage/test/FakeDriverTestConfiguration.java
plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/storage/KVMStorageProcessor.java
It was implemented by extending the NFS provider. Its validation was updated so that you can pass it a URL containing the
details of a CIFS share. The code that mounts NFS shares was extended to allow it do the same for CIFS shares. Otherwise,
the secondary storage code is left unchanged.
Issue happens as there are more than one thread processing connect for a host simultaneously. The VM full sync. is not designed to work in this scenario and as a result user VMs may get stopped incorrectly.
Direct agent scan task runs at regular intervals (direct.agent.scan.interval defaulted to 90 secs) and identifies hosts that needs to be processed for connect. In a normal scenario hosts mostly get connected within that interval and there are no issues. But if due to some reason the connect process takes more time and is not completed by the time next agent scan runs. In this case, based on the db. state same hosts may get picked up again. And then there will be situations where more than one thread is processing connect for the same host.
The fix is to check if there is a thread already processing connect for a host and in this case all subsequent threads for that host will simply bail out. Also there may be a scenario where one thread already completed processing connect but another thread already got scheduled before that and will again repeat the same. This is also prevented by putting appropriate checks.
Check for the all the transition states for Maintenance. Also corrected the isMaintenance function for StoragePoolVo
Signed off by : nitin mehta<nitin.mehta@citrix.com>
Description:
Set the criterion for overriding/preserving the vmware.create.full.clone
flag so that if past version deployments have any deployments (data centers),
this flag will be set to false. Else, it will be set to true.
The earlier criterion to set this flag was based on the CS version numbers,
but that is not a good business logic to serve as a basis to set the flag.
Changes:
- In the upgrade path, for a private zone, entry needs to be added in the affinity_group_domain_map to provide access to the private zone for the domains it belongs too.
Changes:
- Implict creation of the 'ExplicitDedication' Affinity group during resource dedication
- Only one group per account or per domain will be present
- ListDedicatedResources by affinityGroup
- Deployment should consider dedicated resources associated to the group only
- Deleting affinity group should release the dedicated resouces
- Releasing the dedicated resources should remove the group associated if there are no more resources.
Conflicts:
plugins/dedicated-resources/src/org/apache/cloudstack/dedicated/DedicatedResourceManagerImpl.java
plugins/dedicated-resources/test/org/apache/cloudstack/dedicated/manager/DedicatedApiUnitTest.java
server/src/com/cloud/configuration/ConfigurationManagerImpl.java
Changes:
- 'ExcplicitDedication' type of group can be created/deleted by Root admin only
- Users can no longer create this type of affinity group
- RootAdmin can create this type of affinitygroup at domain level. Such a domain level group is available for all accounts in that domain for listing and for use during deployVM.
- The domain level affinitygroup should be visible to the users in that domain, domain admins and Root admin.
Conflicts:
server/src/com/cloud/api/query/QueryManagerImpl.java
server/src/org/apache/cloudstack/affinity/AffinityGroupServiceImpl.java
server/test/org/apache/cloudstack/affinity/AffinityApiUnitTest.java
On a fresh environment, some values in cloud.configuration table are persisted in com.cloud.server.ConfigurationServerImpl.persistDefaultValues()
A default value need to be set before com.cloud.upgrade.DatabaseUpgradeChecker
UI support for baremetal PXE server
CloudStack CLOUDSTACK-1364
UI support for baremetal DHCP server
Conflicts:
plugins/hypervisors/baremetal/src/com/cloud/baremetal/networkservice/BareMetalPingServiceImpl.java
plugins/hypervisors/baremetal/src/com/cloud/baremetal/networkservice/BaremetalKickStartServiceImpl.java
plugins/hypervisors/baremetal/src/com/cloud/baremetal/networkservice/BaremetalPxeManagerImpl.java
While upgrading to 4.2, the type of vswitch being used by each cluster is persisted to cluster_details table. This helps if user want to change the type of vswitch used in a zone or entire cloud later on but leave existing cluster continue to use old vswitches. Hence even after modifying the type of vswitch at cloud level (by modifying global configuration parameters) or modifying the type vswitch at zone level (by modifying the traffic label) would not disturb operation of existing clusters.
Signed-off-by: Sateesh Chodapuneedi <sateesh@apache.org>
Builtin template Ready Status is No even after the Status is downlaod complete. The reason was that template sync updates only the template download state but not the state. Fixing that. Ideally we should change the state through state machine only.
Signed off by : nitin mehta<nitin.mehta@citrix.com>
only when there is portable IP range added at region level.
region response will now have details if portable IP service is enabled
or not. Portable IP service for a region is turned off by default. when
adming adds a portable ip range portable ip service is enabled for the
region.
Updating the new system template URLs for the existing templates during upgrade to 4.2.
If new 4.2 system template is registered before upgrade then marking the old templates as removed during upgrade.
The time increased due to the newly added dedicated resources feature. During regular VM deployment, all dedicated resources are put in avoid list so that they are not considered for deployment.
Now the way to compute the list of dedicated resources is not optimal and performance deteriorates in an environment having lot of pods, clusters and hosts as the logic is to query db. for each suc resource.
The fix is to optimize the logic not to loop through all resources but get the list of each resource type in a single query.
Conflicts:
server/src/com/cloud/deploy/DeploymentPlanningManagerImpl.java
Introduce a global lock on template and pool id so that concurrent threads wont be inserting the same row in DB and hit MySQLIntegrityConstraintViolationException
Signed off by : nitin mehta<nitin.mehta@citrix.com>
Snapshot object is being accessed even when it is null. In case snapshot is not present in backup store the code should return after cleaning db entry.
Also noticed discrepancy in the upgraded db setup but couldn't fully verify it, so added logging in the upgrade logic for template/snapshots etc.
Track the Datacenter of previous cluster correctly while going through each cluster in the zone to see if 2 clusters are from different DC/vCenter.
Cherry picked from 4.2 commit a3450afff5
Signed-off-by: Sateesh Chodapuneedi <sateesh@apache.org>
System template upgrade is not required during 4.0 upgrade since we handle the same during 4.2 upgrade. So removing the system template update during 4.0 upgrade.
KVM.snapshot.enabled is lowercased by f025db95 to keep the configs
uniformly lower-case. But it missed the upgrade script and the
references in SnapshotManagerImpl. This commit will fix the issue in all
locations
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 0e216fa7e41bdfe0cc744006bb896c8b00138ca2)
Private templates would now get copied to only one of image storage chosen randamly as was the case earlier. Dont throw an exception for uploading volumes when there are multiple image stores, instead choose one of them randomly
Signed off by : nitin mehta<nitin.mehta@citrix.com>
Persist the download url in the db for volume download.
Signed off by : nitin mehta<nitin.mehta@citrix.com>
Conflicts:
server/src/com/cloud/storage/VolumeManagerImpl.java
Marked the system template new system template as dynamicallyScalable
- handled upgrade case
- moved "dynamicallyScalable" flag to vm_instance table from user_vm_details to support dynamic scaling of system vm
Signed off by : Nitin Mehta<nitin.mehta@citrix.com>
Changes:
- During host deletion, host entry in databse gets removed prior to the disconnect task getting processed.
- This causes the disconnect task to get NPE while trying to do the host state transition
Some existing scenarios for root and data volume combination was not working. These are
a. Local root + Shared data
b. Shared root + Local data
Enabled these scenarios as part of this fix
The request to migrate was coming till ancient data motion strategy but wasn't worked on and forwarded
to the resource. The code probably got removed by a bad merge. Bringing it back.
The issue is that while calculating the used primary storage size, the updateResourceCount
API is also calculating the disk size of virtual router VM, created for that account and
because of this, the API is returning the incorrect result.
Changes:
- Passing the avoid set generated by the first pass of deployment to the second try.
- The second try is done, when the first pass that uses a reserved plan fails to deploy on the reserved host, to search over the entire zone again
Changes:
- Locking the group and save reservation mechanism done by DPM
- Added admin operation to cleanup VM reservations
- DPM will also cleanup VM reservations on startup
Available bytes was getting stored in the used bytes property of local storage pools. As a result of this, for newly added local pools Cloudstack thinks that there is no space available and generated alerts.
User should be able to delete/archive alerts and events by selecting a time period or by
choosing the alerts and events older than a date. Added the ability to choose a time period
too.
Summary of changes in the fix
- Optimized host scan logic, now instead of iterating over each cluster host scan is done for a batch of clusters
- Made host scan task interval configurable
only on first rule is created on the IP and last rule is revoked on the
IP
Current suboptima logic of IP Assoc
- On associate IP to GuestNetwork there is an IPAssoc command sent to
corresponding network service providers of the network
- On every rule apply on IP associated with the network send IP assoc
to the network service providers
- On every rule deletion on IP associated with a network sernd IP assoc
command to the network service providers
With this fix logic of IP assoc is changed as below which eliminates
executio of unnessary and expensive IpAssocCommand resource command
- On associate IP to GuestNetwork, associate IP only to the network,
Untill any service is associated with the IP dont send IP Assoc
- On creation of first rule on the IP send IPAssoc to corresponding
network service provider. Since IP is used for a service, IPAssoc
need to be sent to correpondign service provider
- On deletion of last rule on the IP send IPAssoc to corresponding
network service provider. When last rule is deleted, IP has no
service associated with it, so send IP assoc to service provider to
remove the IP association
Filter the detail map sent over the wire into Map<String, String> before
processing underneath by storage life cycle
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Description:
When retrieving primary datastore, handle case for non-existing datastores/hosts.
Throw exception and handle the exception in datastore mgmt layer and pass onward
to create storage pool API.
involved in the GSLB
add weights to each site participating in the GSLB. Traffic will be load
balanced across the sites based on the weigths associated with each
site. If not specified weight of site is defaulted to 1.
Breaking down storage components among oss, nonoss and simulator
contexts. The default components are loaded by
OSS - applicationContext + componentContext
NonOSS - applicationContext + nonossComponentContext
Simulator - applicationContext + simulatorComponentContext
provider beans are are selectively overridden for simpler configuration.
Where possible beans are loaded by local reference.
<list merge=true> does not unfortunately work perfectly for bean merging
the providers causing a bit of bloat. Explore for later.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Min Chen
Chris Suich
Marcus Sorensen
Nitin Mehta
Alena Prokharchyk
Harikrishna Patnala
Darren Shepherd
Edison Su
Laszlo Hornyak
Hugo Trippaers
Darren Shepherd
Alex Huang
ynojima
Wei Zhou
Prachi Damle
Daan Hoogland
Kelven Yang
Sheng Yang
Donal Lafferty
Koushik Das
Sanjay Tripathi
Vijayendra Bhamidipati
bharat kumar
Wido den Hollander
Likitha Shetty
frank
Kishan Kavala
Edison Su
Sateesh Chodapuneedi
Mike Tutkowski
Murali Reddy
Prasanna Santhanam
Rajesh Battala
Chiradeep Vittal
Chip Childers
Mice Xia
Saksham Srivastava
Devdeep Singh
Anthony Xu
Koushik Das
Dave Brosius