Create Instance from backup on another Zone (DRaaS use case) #11560
Conversation
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #11560 +/- ##
============================================
+ Coverage 17.36% 17.43% +0.07%
- Complexity 15237 15340 +103
============================================
Files 5888 5890 +2
Lines 525741 526306 +565
Branches 64164 64257 +93
============================================
+ Hits 91274 91782 +508
- Misses 424167 424178 +11
- Partials 10300 10346 +46
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
…ed UpdateBackupRepositoryCmd api.
|
This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch. |
abh1sar
changed the title
| } else { | ||
| clusterId = getClusterIdFromRootVolume(vm); |
There was a problem hiding this comment.
fyi - hostId won't be set for create Instance from Backup. But we can get the clusterId from the root volume which is already created.
| mount += " -o " + mountOptions; | ||
| } | ||
|
|
||
| int exitValue = Script.runSimpleBashScriptForExitValue(mount, mountTimeout, false); |
There was a problem hiding this comment.
fiy - runSimpleBashScript doesn't return any exception. just a return value. So no need to keep it inside try-catch. So, checking the return value instead.
| } catch (Exception e) { | ||
| throw new CloudRuntimeException(String.format("Failed to unmount backup directory: %s", backupDirectory), e); | ||
| String umountCmd = String.format(UMOUNT_COMMAND, backupDirectory); | ||
| int exitValue = Script.runSimpleBashScriptForExitValue(umountCmd); |
There was a problem hiding this comment.
fyi - same here. runSimpleBashScript doesn't throw any exception.
| String error_msg = String.format("Failed to create Instance [%s] from backup [%s] due to: %s.", vm.getInstanceName(), backupDetailsInMessage, result.second()); | ||
| logger.error(error_msg); | ||
| processRestoreBackupToVMFailure(vm, backup, eventId); | ||
| throw new CloudRuntimeException(error_msg); |
There was a problem hiding this comment.
fyi - this is done to pass on the error passed by the provider.
There was a problem hiding this comment.
overall looking good, can you please clarify me with few questions I've raised @abh1sar thank you
...java/org/apache/cloudstack/api/command/user/backup/repository/UpdateBackupRepositoryCmd.java
Outdated
Show resolved
Hide resolved
engine/orchestration/src/main/java/com/cloud/vm/VirtualMachineManagerImpl.java
Outdated
Show resolved
Hide resolved
plugins/backup/nas/src/main/java/org/apache/cloudstack/backup/NASBackupProvider.java
Outdated
Show resolved
Hide resolved
|
@abh1sar also can you please check the failures in simulator CI run |
|
@blueorangutan package |
|
@abh1sar a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ el10 ✔️ debian ✔️ suse15. SL-JID 15024 |
|
[SF] Trillian test result (tid-14345)
|
|
@blueorangutan package |
|
@borisstoyanov a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ el10 ✖️ debian ✔️ suse15. SL-JID 15034 |
boring-cyborg
bot
added
component:integration-test
component:marvin
Python
|
@blueorangutan package |
|
@abh1sar a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ el10 ✔️ debian ✔️ suse15. SL-JID 15037 |
|
@blueorangutan help |
|
@abh1sar [SL] I understand these words: "help", "hello", "thanks", "package", "test" Blessed contributors for kicking Trillian test jobs: ['rohityadavcloud', 'shwstppr', 'vishesh92', 'Pearl1594', 'harikrishna-patnala', 'nvazquez', 'DaanHoogland', 'weizhouapache', 'borisstoyanov', 'vladimirpetrov', 'kiranchavala', 'andrijapanicsb', 'NuxRo', 'rajujith', 'alexandremattioli', 'sureshanaparti', 'abh1sar', 'sudo87', 'rosi-shapeblue'] |
|
@blueorangutan test keepEnv |
|
@abh1sar a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests |
|
[SF] Trillian test result (tid-14361)
|
Description
In continuation of #10140 , this PR adds the functionality that a new Instance can be created on a different Zone from which the backup was taken.
Doc PR: apache/cloudstack-documentation#556
DRaaS Enabledshould be checked while creating the backup repository if it is to be used for creating instance in other zones.(The operator should ensure that the backup repository is accessible in other the Zones as well.)
Other changes done:
Throw relevant errors to the end user. i.e, if backup repository is not accessible or the backup files are not present or corrupted.
Added a timeout to mount operation so that restore fails early if backup repository is not accessible. Current behaviour is that mount will block indefinitely until libvirt times out the command after 1 hour.
Changes to decrease the RTO: Earlier, the new Instance had to be started (to create volumes), stopped, restored and started again. The start-stop cycle takes a long time. The whole process took around 200s on my local env with 15 GB of volume size.
I have added a new param VirtualMachineProfile.Param.ReturnAfterVolumePrepare which creates the volume but doesn't start the VM in orchestrateStart. So one Stop and one Start is avoided during the Instance creation. The whole process finished within 60s with this change.
Types of changes
Feature/Enhancement Scale or Bug Severity
Feature/Enhancement Scale
Bug Severity
Screenshots (if appropriate):
How Has This Been Tested?
How did you try to break this feature and the system with this change?