New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix quorum related issues and a race condition #56

Merged
merged 12 commits into from Jun 8, 2018

Conversation

Projects
None yet
2 participants
@payes
Copy link
Member

payes commented Jun 8, 2018

1)Reply with success when IOs are written to atleast 51% of replicas.
Add delays in through signals in read/write code to simulate the above case.
2)Attempt replica close before trying to connect to controller.
Replica can only connect to controller in closed state.
3)Add delays through signals in replica addition path to simulate race condition.

payes added some commits May 30, 2018

Add test case for 3 replicas and reduce R/W timeout and retry values
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Fix quorum related issues and a race condition
1)Reply with success when IOs are written to atleast 51% of replicas.
  Add delays in through signals in read/write code to simulate the above case.
2)Attempt replica close before trying to connect to controller.
  Replica can only connect to controller in closed state.
3)Add delays through signals in replica addition path to simulate race condition.

Signed-off-by: Payes <payes.anand@cloudbyte.com>
Revert i/o timeout related changes
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Add description to the changes made
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Add description of the test cases in the ci script
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Correct a typo
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Address review comments
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Update jiva with latest gotgt
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Resolve golint issues
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Run make with sudo
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Resolve sudo related issues during build
Signed-off-by: Payes <payes.anand@cloudbyte.com>
Update ci test to delay IOs are the right place
Signed-off-by: Payes <payes.anand@cloudbyte.com>

@vishnuitta vishnuitta self-requested a review Jun 8, 2018

@vishnuitta
Copy link
Member

vishnuitta left a comment

changes are good

@vishnuitta vishnuitta merged commit 59b3351 into openebs:master Jun 8, 2018

1 of 2 checks passed

continuous-integration/travis-ci/pr The Travis CI build is in progress
Details
DCO All commits have a DCO sign-off from the author
Details

muratkars added a commit to muratkars/trusted-charts that referenced this pull request Jul 20, 2018

Update OpenEBS to 0.6.0
What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

muratkars added a commit to muratkars/trusted-charts that referenced this pull request Jul 20, 2018

Update OpenEBS to 0.6.0
What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

@muratkars muratkars referenced this pull request Jul 20, 2018

Merged

Update OpenEBS to 0.6.0 #122

baldwinSPC added a commit to StackPointCloud/trusted-charts that referenced this pull request Aug 2, 2018

Update OpenEBS to 0.6.0 (#122)
* Update OpenEBS to 0.6.0

What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

* Update OpenEBS to 0.6.0

What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

baldwinSPC added a commit to StackPointCloud/trusted-charts that referenced this pull request Aug 2, 2018

Moving to staging. (#129)
* Update OpenEBS to 0.6.0 (#122)

* Update OpenEBS to 0.6.0

What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

* Update OpenEBS to 0.6.0

What's new in this release:
- Integrate the Volume Snapshots capabilities with Kubernetes Snapshot controller
- Enhance maya-apiserver to use CAS Templates for orchestrating new Storage Engines
- Enhance mayactl to provide additional details about volume such as - replica status, node details where replicas are running.
- Enhance maya-apiserver to schedule Replica Pods on specific nodes using nodeSelector
- Enhance provisioner and maya-apiserver to allow specifying cross AZ scheduling of Replica Pods.
- Support for deploying OpenEBS via Kubernetes stable Helm Charts
- openebs-operator.yaml is modified to run OpenEBS pods in its own namespace openebs
- Enhance e2e tests to simulate chaos at different layers like - CPU, RAM, Disk, Network, and Node

Major Issues Fixed
- Fixed an issue where intermittent connectivity errors between controller and replica caused iSCSI initiator to mark the volume as read-only. openebs/gotgt#15
- Fixed an issue where intermittent connectivity errors were causing the controller to silently drop the replicas and mark the Volumes as read-only. The replicas dropped in this way were not getting re-added to the Controller. openebs/jiva#45
- Fixed an issue where volume would be marked as read-only if one of the three replicas returned an error to IO. openebs/jiva#56
- Fixed an issue where replica fails to register back with the controller if the attempt to register occurred before the controller cleared the replica's previous state. openebs/jiva#56
- Fixed an issue where a volume with a single replica would get stuck in the read-only state once the replica was restarted. openebs/jiva#45

* Harbor helm chart (#128)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment