Arctera Application Mobility Service Help

Last Published:
Product(s): InfoScale & Storage Foundation (8.0.2, 7.4.2, 7.4.1, 1.0)
Platform: Linux

Troubleshooting

Table: Common troubleshooting steps

Problem

Solution

Even after running a discovery operation, I am unable to see all my applications and clusters.

  1. Ensure that you have correctly added entries in the /etc/ansible/hosts file in the gateway nodes.

  2. Make sure that you have passwordless SSH configured between the gateway and the source cluster nodes.

  3. Ensure that all nodes in the clusters are accessible. These are the nodes that need to be added to the /etc/ansible/hosts file on the gateway mentioned above.

Migration plan fails during execution.

  1. If a migration plan fails during the target environment creation, you can cancel the plan, resolve any errors, and rerun the plan again.

  2. Resolving errors depends on the type of error you encounter. You can refer to /var/log/ansible.log and /var/log/vrtsgateway.log located at the gateway node to better understand the type of error. Some examples of errors are:

    • An issue with the cloud infrastructure: quota, unsupported EC2/EBS/VM/Data disk types, existing resources with the same name, etc.

    • Errors in migration plan definition, such as wrong path to the SSH key file, wrong subnet information, blocked or already used port number or protocol for ELB/Data disk, etc.

  3. The application switchover is the last step in the migration process. If a migration plan execution fails during the application switchover, you can choose to make some adjustments in the configuration and manually switchover the application.

  4. In addition, you can cancel the plan, but all resources get deleted, and you need to start over.

One or more of the registered datacenters that have been connected to the Application Mobility Service, now shows connection status as 'Not connected' in the portal.

If a previously connected datacenter has a connection status as 'Not connected', you can reconnect the datacenter by running the command systemctl restart gatewayd on the gateway system. This command restarts the gateway services and re-establishes the connection.

In CPS-based fencing (new or existing), the vxfen service group fails and enters the FAULTED or PARTIAL state.

Perform the following steps to bring the vxfen service group ONLINE:

  1. Check the state of the service group by running the command:

    [root@ip-XX-XXX-X-XXX vxfen]# hagrp -state vxfen

    Output:

    #Group Attribute System          Value
vxfen  State     ip-x-xx-x-xx |OFFLINE|FAULTED|

  2. Clear the FAULTED state by running the command:

    hagrp -clear vxfen

  3. Verify the state of vxfen service group is OFFLINE by running the command:

    [root@ip-xx-xxx-x-xxx vxfen]# hagrp -state

    Output:

    #Group Attribute System        Value
vxfen  State     ip-x-xx-x-xx  |OFFLINE|

  4. Bring the service group ONLINE by running the command:

    root@ip-xx-xxx-x-xxx vxfen]# hagrp -online vxfen

  5. Verify the state of vxfen service group is ONLINE by running the command:

    [root@ip-xx-xxx-x-xxx vxfen]# hagrp -state vxfen

    Output:

    #Group Attribute System        Value
vxfen  State     ip-x-xx-x-xx  |ONLINE|

During a node failure in InfoScale or a split-brain scenario, the service group (e.g., CFSMount) may fail to transition to the ONLINE state.

Perform the following steps to manually bring CFSMount service group in ONLINE state:

  1. [root@ip-XX-XXX-X-XXX vxfen]# sudo /opt/VRTSvcs/bin/hagrp -state CFSMount

    Output:

    #Group    Attribute  System        Value
CFSMount  State      ip-XX-XXX-X-X |OFFLINE|

  2. [root@ip-XX-XXX-X-XXX vxfen]# sudo /opt/VRTSvcs/bin/hagrp -online CFSMount -any

    Output:

    #Group    Attribute  System        Value
CFSMount  State      ip-XX-XXX-X-X |ONLINE|

Infoscale installation in Custom AMI configuration may fail in some circumstances due to pre-requisite packages installation issue.

Use the Retry option on the UI to continue with the deployment.

Infoscale CPS fencing configuration may fail intermittently due to timing issue.

Do not attempt the Retry option in this case. Instead, select the Cancel plan option and re-execute the plan.