A flaw was found in the Netfilter subsystem in the Linux kernel. This issue occurs in the nft_set_rbtree, where new elements in this transaction might expire before the transaction ends. Skip sync GC for such elements, otherwise a commit path might walk over an already released object. Once the transaction is finished, async GC will collect expired elements.
1. This flaw can be mitigated by preventing the affected netfilter (nf_tables) kernel module from being loaded. For instructions on how to blacklist a kernel module, please see <https://access.redhat.com/solutions/41278>.
2. If the module cannot be disabled, on non-containerized deployments of Red Hat Enterprise Linux, the mitigation is to disable user namespaces:
# echo "user.max_user_namespaces=0" > /etc/sysctl.d/userns.conf
# sysctl -p /etc/sysctl.d/userns.conf
On containerized deployments such as Red Hat OpenShift Container Platform, do not use the second mitigation (disabling user namespaces) as the functionality is needed to be enabled. The first mitigation (blacklisting nf_tables) is still viable for containerized deployments, providing the environment is not using netfilter.