Q. Why do I see an error “Failed to reach the container” on the application GUI?
The application back end is running as a Docker container on ACI’s APIC server.
For legacy apps, the health-thread APIC checks the health of Gluster-FS (APIC filesystem). If it passes, it checks to ensure the app’s Docker container is able to access it’s data folder under Gluster-FS. If it is not able, it restarts the application container.
There are a few other reasons why the ACI app framework might restart an app’s backend container. In these cases, the application GUI will show the error “Failed to reach the container.”
After the container restarts, a new container runs the application’s back end. The application does a stateful restart and any data available before the restart should be available when the new container is launched. As a result, even though the application might be momentarily unavailable during the restart and show the error, it should recover gracefully.
- If an APIC cluster size changes and the APIC cluster node - Hosting Application container - reboots, you can expect up to three minutes of downtime for the application. It takes three minutes for APIC cluster to bring up a new container on the other currently available APIC nodes. The same thing happens when the APIC node undergoes Commission/Decommission.
- If all APIC cluster nodes are rebooted accidentlly at the same time, it may take up to twenty minutes for the application to be up.
Q. In the app, why do I see the error “BIG-IP session timed out. Please log in again.”?
Every BIG-IP session on the app has an operation-idle time out of 10 minutes. If you do not carry out any operations on a BIG-IP session of the application for 10 minutes, you will see the above error. This timeout check is triggered only on tab switch, or on left-hand menu item click for logged in BIG-IP devices.
Q. In the app, why do I see the error “ERROR : Request failed due to server side error” on APIC?
If App UI is accessed from 2 parallel browser tabs with certificate warnings enabled from only one of the tabs, it may generate this error: “Error: Request failed due to server side error”
Workaround: Login to APIC again
Q. Why do VLANs from the F5 ACI ServiceCenter application visibility table vanish if I destroy and re-create service graph template of my VIRTUAL Logical Device on Cisco APIC?
For virtual ADC logical devices, if you did the following steps
- Take snapshot
- Delete service graph template
- Revert to snapshot config
The VLAN encap values associated with logical interfaces of the LDEV change and do not remain the same. The application detects this change and shows a warning on the L2-L3 stitching LDEV info page that displays VLANs. You can click the warning to update the VLAN tag.
After a VLAN tag is updated on BIG-IP, the visibility vlan table will start showing the VLANs again.
Q. Why do I get an error for VLAN/self IP delete operation from the App?
This is a known issue for BIG-IP v 12.x. If a pool with nodes is associated with a self IP of the same subnet, BIG-IP doesn’t allow user to delete that self IP. As a result, the VLAN delete operation also fails with the error.
- Delete the corresponding pool member from BIG-IP.
- Perform the VLAN/self IP delete from App.
- Recreate the pool member on BIG-IP.
Q. When I try to stitch a VLAN tag, why do I see “VLAN not available for stitching”? Not able to configure the VLAN.
For a single BIG-IP device, after a VLAN tag is stitched for a particular logical device (say LDEV1), the same VLAN tag is not available for stitching again for a different Logical device (say LDEV2). This is because the VLAN tag is already present on the BIG-IP device and re-creating it for a different logical device is not allowed. In order to proceed with stitching, delete the original VLAN from the stitched LDEV, which is mentioned in this VLAN card’s info message.
For a different BIG-IP login, this stitched VLAN tag will still be available for configuration.
Q. Why don’t I see the pre-existing BIG-IP VLANs and self IPs that have a different naming convention than the application?
The application does not support pre-existing VLANs that have a different naming convention than the app. It is able to detect VLANs that have been created and managed from the application only.
Although, after uninstalling and reinstalling the application, if the app database is lost, the application will be able to detect the previously created application VLANs by reading BIG-IP information and show them as Out-of-sync VLANs. The users will also be able to sync them to the application to rebuild App Database.
The application displays APIC VLAN tags for a particular Logical Device Cluster on the L2-L3 stitching page. If there is an out-of-band VLAN with different naming convention but same VLAN tag on the BIG-IP device, the application detects it and shows it in the Out-of-sync information too. But the only action available for such a VLAN or self IP will be deletion of that object from BIG-IP. It cannot sync to application, since it has a different naming convention. The application also does not detect out-of-band information for any of the other VLAN tags that are not a part of APIC VLAN list.
L4-L7 Application Services¶
Q. Why is my L4-L7 Configuration tab disabled?
For the L4-L7 configuration tab to work correctly, f5-appsvcs RPM version 3.7.0 or later is required. Installation steps are available here: https://clouddocs.f5.com/products/extensions/f5-appsvcs-extension/latest/userguide/installation.html#installation
Q. Why is there a warning about “f5-appsvcs package” installation when I log in to my BIG-IP device?
Q. I deleted an application services declaration from the F5 ACI ServiceCenter application. Why do I still see partitions in the declaration?
If your AS3 declaration contains “optimisticLockKey” mentioned explicitly, the AS3 configuration may not be deleted completely, even after multiple attempts from the application UI. However, the configuration gets removed from the BIG-IP device.
Workaround: Upload one more AS3 sample declaration to the app and then perform a Delete all operation. (Use View AS3 Declaration and click Delete.)
Q. For L4-L7 App Service tab, why does the partition get deleted when I delete the last application belonging to that partition?
If there is a single application in a particular partition, and if that application is deleted through the application, the partition that has no other applications under it will be deleted from the BIG-IP device. This is standard F5 BIG-IP behavior. You will be warned about this in the delete confirmation prompt.
Q. What is the best way to delete LDEV from APIC?
Do not delete Logical devices from APIC directly. Instead, as a first step, delete self IPs, VLANs and routes from the BIG-IP device by using the application. When you are done, you can delete the Logical Device from APIC. This ensures there are no stale self IP, VLAN, and route entries on BIG
Q. What browsers are supported?
The app has been tested with IE11, Mozilla FireFox 56 and Google Chrome v72.
Q. What scale numbers were tested with the app?
|Number of BIG-IPs||60|
|Per BIG-IP paritions||100|
|Per BIG-IP Virtual IPs||100|
|APIC logical devices||60|
|Per BIG-IP nodes members||4|
|Concurrent app operations||4 BIG-IPs|