Troubleshooting

When an error occurs, check the status of Spectrum Virtualize Family, VMware ESX servers, vCenter servers, and SANs.

Managing DiscoverArrays failure

Verify the following items when you encounter a failure during Array Manager operation:
  1. Check that the IP address is correct and in the correct format. For example, the format can resemble: 192.168.10.1:5989.
  2. Check that the user name and password are correct.
  3. Check to see whether the Spectrum Virtualize Clusters are paired.

Managing DiscoverDevices failure

Important: Relationship and consistency group (CG) status typically is consist_synchronized or consistent_copying for Global Mirror with Cycling Mode.

Managing recovery failure

For environments that are not pre-configured, ensure that the following items are checked or corrected:
  1. Recovery, after you switch from pre-configuration without a rescan of the SRA.
  2. CIM user privilege problem, administrator access required.
  3. After disaster recovery, it is necessary to wait a few minutes and then refresh the devices in the SRM Array Managers on both sites. This step is necessary to ensure that there is no error before the plan is reprotected.
  4. All volumes part of a consistency group are not included in the SRM Recovery Plan.
For environments that are pre-configured, ensure that the following items are checked or corrected:
  1. FlashCopy® was not created.
  2. Corresponding fcmap consistency group was not created.
  3. The Remotecopy target volume is not mapped to the recovery site.
  4. All volumes part of a consistency group are not included in the SRM Recovery Plan.

Managing test failure

For environments that are not preconfigured, ensure that the following are checked or corrected:
  1. CIM user privilege problem, administrator access required.
  2. Not enough space in the mdisk group.
  3. Switching from pre-configured without rescan SRA.
  4. The Mdiskgroup and volume type is not set for Spectrum Virtualize Family SRA.
  5. All volumes part of a consistency group are not included in the SRM Recovery Plan.
For environments that are preconfigured, ensure that the following items are checked or corrected:
  1. FlashCopy not created.
  2. FlashCopy target volume is not mapped to host.
  3. Corresponding FlashCopy consistency group not created.
  4. FlashCopy target volume that is mapped to the host are shown as Mounted in Hosts and Clusters/Configuration/Storage/Devices.
  5. All volumes part of a consistency group are not included in the SRM Recovery Plan.

Managing SRM procedures

  1. Refresh the devices in the SRM array managers. If there are no errors, go to step 7. Otherwise, continue to step 2.
  2. Check all the previously related remote copies in the SRA-STATUS file in the SRA installation directory. For example, sudo docker exec -it <container id> cat /srm/sra/SRA-STATUS. Retrieve <container id> from sudo docker ps command. The format of the SRA-STATUS file is shown in the following example:
    [ArrayID]|[LunID]=* [ArrayID] + "|CG;" + 
[RemoteCopyConsistencyGroupName]=*
    A sample of the content of the SRA-STATUS file can be seen in the following example: 
    000002006E814E66|2=CREATEVOLUME@4;
60:05:07:68:01:BA:05:39:98:00:00:00:00:00:00:2C 
TESTFAILOVER@4;
60:05:07:68:01:BA:05:39:98:00:00:00:00:00:00:2C
    This example shows that the standalone remote copy (target volume id:[2]) in the array [000002006E814E66] was in testfailover status. Spectrum Virtualize Family SRA created a FlashCopy for the remote copy target [2] and the FlashCopy target volume id is [4]. The UUID is 60:05:07:68:01:BA:05:39:98:00:00:00:00:00:00:2C.
  3. Copy the SRA-STATUS file from container using docker cp. For example, docker cp <container id>:/srm/sra/SRA-STATUS <local path>.
  4. Modify the SRA-STATUS file and upload it to the container. For example, docker cp <local path>/SRA-STATUS <container id>:/srm/sra/.
  5. Verify the SRA-STATUS content. For example, docker exec -it <container id> cat /srm/sra/SRA-STATUS.
  6. Delete the SRA-STATUS file if none of the entries are required. For example, docker exec -it <container id> rm /srm/sra/SRA-STATUS.
  7. Refresh the devices in the SRM array managers again to ensure that there are no errors.

Collect log files

If the problem cannot be resolved by following any of the troubleshooting tips, collect the SRM log files from /var/log/vmware/srm and contact IBM® support.