Last Modified: Jun 03, 2021
See more info
Known Affected Versions:
126.96.36.199, 16.0.1, 15.1.3, 15.1.2, 188.8.131.52, 184.108.40.206, 14.1.4, 220.127.116.11, 13.1.4, 18.104.22.168, 12.1.6, 22.214.171.124, 126.96.36.199
Opened: Apr 28, 2021
An ephemeral pool member may not created if the FQDN name resolves to a new IP address that matches an existing statically-configured node. When this occurs, a message like the following appears in the LTM log: err mcpd: 01070734:3: Configuration error: node (/Common/_auto_10.10.120.12) not found. Note that the node name in the message is the expected name of an ephemeral node created for this address, not the actual name of the statically-configured node with that IP address.
An ephemeral pool member is not created for the IP address newly included in the DNS query result. This results in traffic not being load-balanced to all of the expected pool members.
This may occur if: -- The FQDN node and pool member are created with the "autopopulate enabled" option. -- The FQDN name resolves to more than one IP address. -- One of these IP addresses was not included in the previous DNS query result. -- There is a statically-configured node with the same IP address.
Use one of the following methods to prevent this issue from occurring: -- Avoid creating statically-configured nodes using the same IP addresses returned by resolution of configured node/pool member FQDN names. -- Configure the FQDN pool member with "autopopulate disabled" (default), which creates only a single ephemeral pool member. Perform this sequence of actions to recover from an occurrence of this issue: 1. Remove any pool members referencing the conflicting IP address(es) from their respective pool(s). 2. Delete the statically-configured node(s) using the conflicting IP address(es). 3. Add any pool members referencing the conflicting IP address(es) back to their respective pool(s).