venkata swamybabu budumuru created CLOUDSTACK-4327:
------------------------------------------------------
Summary: [Storage Maintenance] SSVM, CPVM and routerVMs are
running even after storage entered into maintenance.
Key: CLOUDSTACK-4327
URL: https://issues.apache.org/jira/browse/CLOUDSTACK-4327
Project: CloudStack
Issue Type: Bug
Security Level: Public (Anyone can view this level - this is the default.)
Components: Storage Controller
Affects Versions: 4.2.0
Environment: commit id # 8df22d1818c120716bea5fce39854da38f61055b
Reporter: venkata swamybabu budumuru
Fix For: 4.2.0
Step to reproduce :
1. Have latest CloudStack setup with at least 1 advanced zone.
2. Above setup was created using Marvin framework using APIs
3. During the creation of zone, I have added 2 cluster wide primary storages
- PS0
- PS1
mysql> select * from storage_pool where id<3\G
*************************** 1. row ***************************
id: 1
name: PS0
uuid: 5458182e-bfcb-351c-97ed-e7223bca2b8e
pool_type: NetworkFilesystem
port: 2049
data_center_id: 1
pod_id: 1
cluster_id: 1
used_bytes: 4218878263296
capacity_bytes: 5902284816384
host_address: 10.147.28.7
user_info: NULL
path: /export/home/swamy/primary.campo.kvm.1.zone
created: 2013-08-14 07:10:01
removed: NULL
update_time: NULL
status: Maintenance
storage_provider_name: DefaultPrimary
scope: CLUSTER
hypervisor: NULL
managed: 0
capacity_iops: NULL
*************************** 2. row ***************************
id: 2
name: PS1
uuid: 94634fe1-55f7-3fa8-aad9-5adc25246072
pool_type: NetworkFilesystem
port: 2049
data_center_id: 1
pod_id: 1
cluster_id: 1
used_bytes: 4217960071168
capacity_bytes: 5902284816384
host_address: 10.147.28.7
user_info: NULL
path: /export/home/swamy/primary.campo.kvm.2.zone
created: 2013-08-14 07:10:02
removed: NULL
update_time: NULL
status: Maintenance
storage_provider_name: DefaultPrimary
scope: CLUSTER
hypervisor: NULL
managed: 0
capacity_iops: NULL
2 rows in set (0.00 sec)
Observations:
(i) SSVM and CPVM volumes got created on pool_id=1
4. Zone got setup without any issues.
5. Added following zone wide primary storages
- test1
- test2
mysql> select * from storage_pool where id>7\G
*************************** 1. row ***************************
id: 8
name: test1
uuid: 4e612995-3cb1-344e-ba19-3992e3d37d3f
pool_type: NetworkFilesystem
port: 2049
data_center_id: 1
pod_id: NULL
cluster_id: NULL
used_bytes: 4214658203648
capacity_bytes: 5902284816384
host_address: 10.147.28.7
user_info: NULL
path: /export/home/swamy/test1
created: 2013-08-14 09:49:56
removed: NULL
update_time: NULL
status: Up
storage_provider_name: DefaultPrimary
scope: ZONE
hypervisor: KVM
managed: 0
capacity_iops: NULL
*************************** 2. row ***************************
id: 9
name: test2
uuid: 43a95e23-1ad6-30a9-9903-f68231dacec5
pool_type: NetworkFilesystem
port: 2049
data_center_id: 1
pod_id: NULL
cluster_id: NULL
used_bytes: 4214658793472
capacity_bytes: 5902284816384
host_address: 10.147.28.7
user_info: NULL
path: /export/home/swamy/test2
created: 2013-08-14 09:50:12
removed: NULL
update_time: NULL
status: Up
storage_provider_name: DefaultPrimary
scope: ZONE
hypervisor: KVM
managed: 0
capacity_iops: NULL
6. Have created a non-ROOT domain user and deployed VMs
7. Create 5 volumes as above users (volume ids : 23,24,25,26,27)
8. Tried to attach 23 & 24 volumes to the above deployed VM
Observations :
(ii) user VMs came up on pool_id=1 and router VMs came up on pool_id=2
(iii) both the DATADISKS (23 & 24) got attached but, they got allocated on
pool_id=1. It never picked zone wide storages and it looks like the only way
for this to happen is through storageTags.
9. Now place the storages (PS0, PS1) in maintenance mode.
Observations:
===========
a. Both PS0 and PS1 went into maintenance mode successfully but, SSVM, CPVM and
routerVMs are still running. Only User VMs went into stopped state.
b. Since my network.gc.* is set to 60 seconds then it went and stopped my
router VMs.
c. Job-63 and job-64 are the jobs for prepareMaintenanceMode commands.
Expected Result:
============
- When Storage Says that it is in maintenance mode then it should not have any
VMs in running state.
Attaching all the required logs along with db dump to the bug.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
