Description of problem: After upgrading from fedora 27, I've been having intermittent issue with ipv6 traffic; pinging the gateway somehow made the connection work, however shortly after stopping, I would start getting 100% packet loss pinging an external ipv6 machine. After reading: https://serverfault.com/questions/477471/ipv6-only-works-after-pinging-the-default-gateway?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa seemed to suggest it might be a firewall and icmp6 related issue. I simply turned firewalld off and ipv6 traffic started working as expected. Version-Release number of selected component (if applicable): firewalld version: 0.5.2 How reproducible: attempt to ping, connect or use an ipv6 host outside of the local network and wait for packet loss to occur. Steps to Reproduce: 1. have ipv6 capable network 2. ping ipv6 host outside of local network; packet loss might take up to a minute or so to surface. Actual results: 100% packet loss shortly after stopping pinging the gateway. Expected results: little to no packet loss that does not depend on constantly pinging the default gateway. Additional info: I'm assuming there is a configurable way to fix this but I'm not that familiar with firewalld.
Can you post your firewalld configuration for zone of the outgoing interface? # firewall-cmd --get-zone-of-interface=<interface> # firewall-cmd --zone=<zone> --list-all
firewall-cmd --get-zone-of-interface=wlp58s0 FedoraWorkstation firewall-cmd --zone=FedoraWorkstation --list-all FedoraWorkstation (active) target: default icmp-block-inversion: no interfaces: wlp58s0 sources: services: dhcpv6-client ssh samba-client mdns ports: 1025-65535/udp 1025-65535/tcp protocols: masquerade: no forward-ports: source-ports: icmp-blocks: rich rules:
See also bug 1575431, bug 1591867. They are likely duplicated.
*** This bug has been marked as a duplicate of bug 1575431 ***