Image repository problems

Use this information if an image repository for AIX®, IBM® i, or Linux on IBM Flex System Power Systems™ was previously configured, but no longer appears on the IBM Flex System Manager.

Table 1. Image repository problems and corrective actions
Problem description Corrective action
An image repository that was created previously does not appear on the IBM Flex System Manager. To recover the image repository, collect inventory in the IBM Flex System Manager management software web interface for the following managed endpoints in the following order.
Important: You must complete each inventory action in the following list without errors, in order, before you proceed to the next inventory action.
Note: For more information about how to collect inventory for a managed endpoint, see Collecting inventory.
  1. Storage array (for example, the Flex System V7000) managed endpoint
  2. Storage Area Network (SAN) switch managed endpoint
  3. Storage farm managed endpoint
  4. Every compute node managed endpoint that hosts a VIOS logical partition (LPAR)
  5. Every virtual server managed endpoint that represents a VIOS LPAR
  6. Every operating system managed endpoint that represents a VIOS installation

If the image repository does not appear on the IBM Flex System Manager after you collect inventory for the preceding managed endpoints, you can use the smcli dumpstcfg command to investigate the problem.

To run the smcli dumpstcfg command, complete the following steps:
  1. Open the management software command-line interface.
  2. Log in with the USERID account credentials.
  3. Enter command smcli dumpstcfg -v
Information about the SAN configuration is displayed.
After you restored an management software image and collected inventory on the storage farm managed endpoint without errors, you experience storage fabric errors and the management software incorrectly displays multiple endpoints for image repositories and virtual appliances. To solve the problems, complete the following steps:
  1. From the management software web interface, click the Chassis Manager tab and make sure that the chassis status is Managed.
  2. Click General Actions > Resource Explorer; then, after the Resource Explorer page opens, click All Systems. Make sure that all of the discovered endpoints are displayed with an Access status of OK (and not the status Not Trusted).
  3. Verify your network configuration from the SMIA Configuration Tool:
    1. From the Home page, click the Applications tab; then, make sure that the SMIA Configuration Tool is running.
      Note: If you have external versions of the SMIA Configuration Tool, make sure that those are also running and can be pinged from the management node.
    2. From the Applications tab, click Launch administration console and log in to the SMIA Configuration Tool.
    3. In the SMIA Configuration Tool, click the CIMOM tab and make sure that the CIMOM settings are correct.
    4. Click the Home tab; then, click the Fabric Discovery link. Make sure that all of the required network switches are discovered, with a green check mark that indicates proper SNMP trap communication.
  4. Use the management software command-line interface (CLI) to check data sources, remove an Fabric Data Source object, rediscover the object, and ping the operating system endpoints that host image repositories:
    1. From the management software CLI, use the lsdatasource command to verify that all of the expected data sources are present, and an OID is listed for each network device.
    2. Use the rmdatasource to remove the Fabric Data Source object, as shown in the following example:
      smcli rmdatasource -c fabric -i <IPv6>
      where the IPv6 is the address for Fabric Data Source 3.
    3. Use the mkdatasource to rediscover the Fabric Data Source object, as shown in the following example:
      smcli mkdatasource -c fabric -t https <IPv6> -p 25989 -u USERID -w Passw0rd -n /interop
      where the IPv6 is the address for Fabric Data Source 3, and USERID and Passw0rd are the management software administrator user name and password.
    4. Use the pingsys command to ping the each of the operating system endpoints that host image repositories.
  5. Use the management software web interface to view operating-system endpoint access states and collect inventory:
    1. Open the Resource Explorer page.
    2. Click All Systems; then, make sure that all of the operating system endpoints that host image repositories are displayed with an Access status of OK.
    3. Select the operating system and farm endpoints from the Resource Explorer table; then, click Actions > Inventory > Collect Inventory.
  6. After inventory has been collected, check the image repositories and virtual appliances in VMControl:
    1. From the VMControl summary page, click the Virtual Appliances tab.
    2. Under Where to deploy:, click the <number> Image repositories link, where <number> is the number of image repositories in your environment. The Image Repositories page opens.
    3. Select each image repository; then, click Actions > Related Resources > Server. Make sure that the server endpoint that hosts each image repository is displayed with an Access status of OK.
    4. With each image repository still selected, click Actions > Related Resources > Server. Make sure that the server endpoint that hosts each image repository is displayed with an Access status of OK.
      If you find that the server access for an image repository is OK, but the image container is missing, then an image repository endpoint remains, even though it was deleted after the management software backup was created. The errant image repository endpoint must be deleted manually.
      Attention: Make sure that you do not remove the image repository that still has one or more virtual appliances.
    5. Make sure that each virtual appliance is associated with its captured disk in the storage device.
      Note: You can verify the Captured SCS Virtual Appliances for the storage device from the management software web interface. For example, if your storage device is an IBM Flex System V7000 Storage Node, click Launch IBM Flex System V7000, log in to the V7000 Storage Node management GUI, and click Volumes. For other storage devices, go to the Plug-ins tab and click Systems and Volumes under Storage Management.
    6. From the Virtual Appliances tab, under What to deploy:, click the <number> Virtual appliances link, where <number> is the number of virtual appliances in your environment. The Virtual Appliances page opens.
    7. Select each virtual appliance; then, click Actions > Related Resources > Software Image.
    8. Compare the virtual appliance software images with the volumes listed for the storage device. Any virtual appliance that does not have an associated software image must be removed manually; the virtual appliance might have been deleted after the management software backup was created, but still exists in the management software database. After you remove the errant virtual appliance endpoint, the management software reflects the current state of your environment.
Parent topic: Troubleshooting the IBM Flex System Manager management software and hardware