Commit Graph

86 Commits

Author SHA1 Message Date
Rajani Karuturi 85bed5fa2b COVERITY: Fixed issues reported by coverity NPEs, unwritten field access and self assignment
Signed-off-by: Sebastien Goasguen <runseb@gmail.com>
2014-07-02 16:07:06 +02:00
Harikrishna Patnala 15bf144f1a CLOUDSTACK-6211: Xenserver - HA - SSVM fails to start due to running out of management Ip ranges when testing host down scenarios
Signed-off-by: Koushik Das <koushik@apache.org>
2014-03-13 11:02:02 +05:30
Alena Prokharchyk eb61f8146d CLOUDSTACK-5872: use List DS for storing NicProfiles as public network can have more than one nic 2014-03-07 01:07:59 +07:00
Koushik Das c6a8659ac2 CLOUDSTACK-6124: During MS maintenance unfinished work items are not cleaned up resulting in them getting repeated for every subsequent maintenance
Updating the op_it_work table entry appropriately in db once the unfinished work item is completed.
2014-02-21 16:16:52 +05:30
Koushik Das 775fa0f0d0 CLOUDSTACK-5989: Trying to start a vm while 'vm snapshot' is in progress results in inconsistency
The fix is to fail the start operation if a vm snapshot is in progress
2014-01-30 18:10:56 +05:30
Kelven Yang 29e44b1b67 CLOUDSTACK-5928: Use legacy vmsync logic by default 2014-01-22 14:09:22 -08:00
Kelven Yang 50f311ed94 CLOUDSTACK-5928: disable host delta sync when new VM sync is enabled 2014-01-22 11:17:25 -08:00
Mike Tutkowski 0679af3434 CLOUDSTACK-5873: [Automation] Failed to attach volume to VM, if the vm is created with option startvm=false 2014-01-22 11:08:57 -07:00
Kelven Yang 8db0d83d1a CLOUDSTACK-5358: Bring back concurrency control in sync-queue management 2014-01-20 16:53:33 -08:00
Mike Tutkowski 0e053668b8 CLOUDSTACK-5873: [Automation] Failed to attach volume to VM, if the vm is created with option startvm=false 2014-01-18 12:36:53 -07:00
edison 7c2449dabd CLOUDSTACK-5432:sync vm and storage commands on mgt server for kvm 2014-01-17 15:23:43 -08:00
Anthony Xu b52e8028e9 only send stop command when agent reports VM running and CS thinks it is stopped. 2014-01-17 11:48:27 -08:00
Kelven Yang be0d688511 CLOUDSTACK-5731: Use general instance type to categorize VM work jobs to correctly serialize VM operations 2014-01-17 10:42:04 -08:00
Mike Tutkowski e2364cfca6 CLOUDSTACK-5873: [Automation] Failed to attach volume to VM, if the vm is created with option startvm=false 2014-01-16 12:08:05 -07:00
Kelven Yang 7a8c0e3ae0 CLOUDSTACK-5696: Fix sync issue with out-of-band changes 2014-01-14 17:33:15 -08:00
Harikrishna Patnala c3e6444adf CLOUDSTACK-5827: [Automation] Destroy VM failed, while deleting account
Signed-off-by: Koushik Das <koushik@apache.org>
2014-01-08 18:29:58 +05:30
Kelven Yang c75b0044ef CLOUDSTACK-5765: cleanup internal serialization and exception propagation issues 2014-01-07 17:12:06 -08:00
Koushik Das 2b2d5998d7 CLOUDSTACK-5002: unable to destroy vm ;VM destroy failed in Stop i-2-59-VM Command due to You gave an invalid object reference. The object may have recently been deleted.
This is happening as concurrent operations are happening on the same VM. Earlier this was not seen as all vm operations were synchronized at agent layer. By making execute.in.sequence
global config to false this restriction is no longer there. In the latest code operations to a single vm are synchronized by maintaining a job queue. In some scenarios the destroy vm operation
was not going through this job queue mechanism and so was resulting in failures due to simultaneous operations.
2014-01-06 18:19:06 +05:30
Kelven Yang 22a1b8c3fc CLOUDSTACK-5672: Fix VM work job serialization issues in Add/Remove nic 2014-01-03 17:20:24 -08:00
edison db25228b23 CLOUDSTACK-5329: fix NPE in storage migration for zone-wide primary storage, and also the serization error. 2014-01-03 16:44:21 -08:00
Kelven Yang ce7bc8fd90 CLOUDSTACK-5725: put origin flow context id into log4j context prefix to link jobs with the orchestration work flow in logging 2014-01-03 15:47:03 -08:00
Kelven Yang 9cad092054 CLOUDSTACK-5767: Remove the logic of using underlying job related fields for volume specific logic. 2014-01-03 13:10:32 -08:00
Harikrishna Patnala 6ce46e94f5 CLOUDSTACK-5730: [Automation]DirectAgentAttache's send command fails due to ArrayIndexOutOfBoundsException; Unable to cleanup hypervisor's datastructures as part of Delete Account Job
Signed-off-by: Koushik Das <koushik@apache.org>
2014-01-03 15:42:31 +05:30
Kelven Yang a03c8fe668 Fix a typo caused problem in attach-volume process 2013-12-31 15:13:10 -08:00
Kelven Yang 3094160718 CLOUDSTACK-669: Finalize VM work dispatching mechanism to avoid big switch statement 2013-12-31 13:32:28 -08:00
Mike Tutkowski e65fae412c CLOUDSTACK-5662: XenServer can't discover iSCSI targets with different credentials 2013-12-28 21:43:28 -07:00
Mike Tutkowski 8eb430f602 CLOUDSTACK-5662: XenServer can't discover iSCSI targets with different credentials 2013-12-28 00:12:11 -07:00
Kelven Yang dc303fda6e CLOUDSTACK-669: Convert volume snapshot orchestration flow to make it be serialized with other VM operations 2013-12-27 16:40:50 -08:00
Devdeep Singh 079323f591 CLOUDSTACK-5420: The agent manager wasn't transitioning the host to maintenance
mode if their are no vms running on the host. Made the change to do so.
2013-12-26 11:14:01 +05:30
Kelven Yang 497feab841 CLOUDSTACK-669: covert VMsnapshot orchestration flows to make them be serialized with other VM operations 2013-12-20 16:45:27 -08:00
Nitin Mehta dc0f8ebdf1 CLOUDSTACK-4941:
During HA call different planners which skip some heurestics
2013-12-17 18:32:46 -08:00
Kelven Yang ed2125ec34 CLOUDSTACK-669: convert volume attach/detach flows to make them be serialized with other VM operations 2013-12-17 17:11:05 -08:00
Prachi Damle 35afd012f5 CLOUDSTACK-4852: Since upgrade to 4.2 only users at the zone-attached domain level can manipulate VMs
Changes:
- The vmprofile owner passed in to the planner should be the VM's account and not the caller
- Do not do the access check for Root Admin
2013-12-17 11:23:21 -08:00
Bharat Kumar 9118bbf934 CLOUDSTACK-5495 Unable to stop VM. adding the changes to make the global config execute.in.sequence.hypervisor.commands configurable for stopvm command.
Signed-off-by: Koushik Das <koushik@apache.org>
2013-12-17 17:43:19 +05:30
Kelven Yang 96678bbbfb CLOUDSTACK-669: refactor VM work job dispatcher to allow volume/snapshot manager to participate serialized job handling 2013-12-13 17:26:13 -08:00
Kelven Yang 619f5381d5 CLOUDSTACK-669: put user vm work under new vmsync model 2013-12-11 16:19:57 -08:00
Kelven Yang d931718ab2 CLOUDSTACK-669: put system vm work under new vmsync model 2013-12-06 15:48:10 -08:00
Prachi Damle 13d2caa6b5 CLOUDSTACK-5401: VM migration during host maintenance fails if pool.storage.capacity.disablethreshold is lowered
Changes:
- During Vm migration while finding a new host within the cluster, we need to set the storagepool Id to the deployment plan too.
- This will indicate the planner that the volumes are ready and no need to find new pool
- This in turn will prevent the threshold check done during the pool allocation. This step is not needed since there is no need to allocate pools newly.
- Thus the migration wont fail because th threshold check fails.
2013-12-06 13:18:40 -08:00
Anthony Xu 0847ccfe97 XS VMmetadata platform contains virtual platform related info, CS should save platform info in DB to migrate the VM to other Cluster
Conflicts:
	engine/orchestration/src/com/cloud/vm/VirtualMachineManagerImpl.java
	plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
	server/src/com/cloud/template/TemplateManagerImpl.java
2013-12-05 01:17:56 -08:00
Alena Prokharchyk b315ca1895 CLOUDSTACK-5261: support for alert publishing via Root admin web api
listAlerts: introduced new parameter "name" to the alertResponse

Added new Admin API - generateAlert. Available to ROOT admin only

listAlerts: implemented search by alert name
2013-12-03 17:08:23 -08:00
Bharat Kumar 089c7bc9f8 CLOUDSTACK-5161 use a map to specify the custom parameters instead of having one parameter each 2013-12-03 16:53:58 +05:30
Bharat Kumar 72e0354661 CLOUDSTACK-5161 enable scaling and upgrading of a vm using custom offering
Signed-off-by: Koushik Das <koushik@apache.org>
2013-11-29 18:32:16 +05:30
Alena Prokharchyk 592863f1ef Removed unused agent lb timer
Conflicts:
	engine/orchestration/src/com/cloud/agent/manager/ClusteredAgentManagerImpl.java
2013-11-21 14:34:42 -08:00
Kelven Yang 5a44de282b CLOUDSTACK-669: side-by-side VM sync management at manager level 2013-11-20 18:22:38 -08:00
Laszlo Hornyak 6f3688d13d Fill the creationMonitors based on priority
If not priotity, append to the end of the list.

Signed-off-by: Laszlo Hornyak <laszlo.hornyak@gmail.com>
2013-11-08 21:51:04 +01:00
Bharat Kumar 7095ea2b5e CLOUDSTACK-4738 Dynamic compute offering.
Signed-off-by: Koushik Das <koushik@apache.org>
2013-11-07 12:41:20 +05:30
Kelven Yang 191dc80fa4 VM power state for job oritented sync processing 2013-11-05 19:00:37 -08:00
Koushik Das 269a4ef11e CLOUDSTACK-4855: Throttle based on the # of outstanding requests to the directly managed HV host (direct agents)
Cloudstack sends requests to directly managed HV hosts (direct agents) using the direct agent thread pool. The size of the pool is determined by global config direct.agent.pool.size defaulted to 500.

Currently there is no restriction on the number of threads a direct agent can use from this shared thread pool to send requests to the host. This is fine as long as the host is responding to requests
in a reasonable amount of time. But if there is a considerable delay in getting response, the thread remain blocked for that much time. As more commands are send to the slow host threads keep getting
blocked. This can eventually lead to a situation where requests to healthy hosts cannot be processed as there are not enough free threads.

The problem being addressed here is to localize the impact of few bad hosts, so that entire management server is not affected.

One such way is to throttle based on the # of outstanding requests on per host basis. The outstanding requests to a host will be a % of direct agent pool size. This is configurable based on
direct.agent.thread.cap. The default value is 0.1 or 10%, a value of 1 would mean the old behavior where there is no upper cap. This will ensure that the impacted host will be bound by a upper cap on the number of threads it can use to process requests and not the entire pool.
2013-11-04 14:52:26 +05:30
Alena Prokharchyk 591dcd1d1c ResourceDetails: added "display" field to a bunch of VOs - determines whether the detail should be returned to the regular user 2013-10-29 09:18:11 -07:00
Mike Tutkowski 858ce76665 Extend support of CloudStack-managed storage to KVM 2013-10-27 22:30:03 -06:00