Dear OVS Community,
I have encountered an issue where some DNS requests are timing out, and it appears to be related to a failure to “put” flows. Specifically, I’m seeing the following errors in the logs: *2024-09-20T03:12:49.415Z|00392|dpif(handler2)|DBG|system@ovs-system: failed to put[create] (File exists) ufid:31e07114-bc99-4cb1-88c0-69436ef2f4ad recirc_id(0x3),dp_hash(0/0),skb_priority(0/0),in_port(2),skb_mark(0/0),ct_state(0x21/0),ct_zone(0x57/0),ct_mark(0/0),ct_label(0/0),ct_tuple4(src=163.44.196.189/0.0.0.0,dst=216.86.161.8/0.0.0.0,proto=6/0,tp_src=33260/0,tp_dst=22/0 <http://163.44.196.189/0.0.0.0,dst=216.86.161.8/0.0.0.0,proto=6/0,tp_src=33260/0,tp_dst=22/0>),eth(src=9c:05:91:c1:93:fd/00:00:00:00:00:00,dst=7e:69:17:8b:54:3e/00:00:00:00:00:00),eth_type(0x0800),ipv4(src=163.44.196.189/0.0.0.0,dst=216.86.161.8,proto=6/0,tos=0/0,ttl=46/0,frag=no <http://163.44.196.189/0.0.0.0,dst=216.86.161.8,proto=6/0,tos=0/0,ttl=46/0,frag=no>),tcp(src=33260/0,dst=22/0),tcp_flags(0/0), actions:ct(commit,zone=87,nat(dst=172.27.45.101)),recirc(0x190)* *--* *2024-09-20T03:12:53.429Z|00400|poll_loop(handler2)|DBG|wakeup due to [POLLIN] on fd 40 (NETLINK_GENERIC<->NETLINK_GENERIC) at ../lib/dpif-netlink.c:3200 (0% CPU usage)* *2024-09-20T03:12:53.429Z|00401|dpif_netlink(handler2)|DBG|added flow* *2024-09-20T03:12:53.429Z|00402|poll_loop(handler2)|DBG|wakeup due to 0-ms timeout at ../ofproto/ofproto-dpif-upcall.c:818 (0% CPU usage)* *2024-09-20T03:12:53.429Z|00354|poll_loop(handler5)|DBG|wakeup due to [POLLIN] on fd 42 (NETLINK_GENERIC<->NETLINK_GENERIC) at ../lib/dpif-netlink.c:3200* *2024-09-20T03:12:53.429Z|00355|dpif(handler5)|DBG|system@ovs-system: failed to put[create] (File exists) ufid:3c04e790-a66d-4969-b70b-0c477ba7ad64 recirc_id(0x4),dp_hash(0/0),skb_priority(0/0),in_port(2),skb_mark(0/0),ct_state(0x2a/0x3f),ct_zone(0x1/0),ct_mark(0/0x1),ct_label(0/0),ct_tuple4(src=172.27.45.101/0.0.0.0,dst=8.8.8.8/0.0.0.0,proto=17/0,tp_src=38447/0,tp_dst=53/0 <http://172.27.45.101/0.0.0.0,dst=8.8.8.8/0.0.0.0,proto=17/0,tp_src=38447/0,tp_dst=53/0>),eth(src=1a:69:bb:b2:f9:88/00:00:00:00:00:00,dst=7e:69:17:8b:54:3e),eth_type(0x0800),ipv4(src=8.8.8.8/0.0.0.0,dst=172.27.45.101,proto=17/0,tos=0/0,ttl=55/0,frag=no <http://8.8.8.8/0.0.0.0,dst=172.27.45.101,proto=17/0,tos=0/0,ttl=55/0,frag=no>),udp(src=53/0,dst=38447/0), actions:6* *--* I have also seen similar errors in other logs, with multiple instances of DNS requests failing to resolve, as shown below: *ubuntu@dnstest:~$ while true; do time host google.com <http://google.com>; date; sleep 2; done* *google.com <http://google.com> mail is handled by 10 smtp.google.com <http://smtp.google.com>.* *Thu Sep 19 22:15:03 UTC 2024* *google.com <http://google.com> has address 172.253.115.113* *google.com <http://google.com> has address 172.253.115.138* *google.com <http://google.com> has address 172.253.115.102* *google.com <http://google.com> has address 172.253.115.101* *google.com <http://google.com> has address 172.253.115.139* *google.com <http://google.com> has address 172.253.115.100* *;; communications error to 127.0.0.53#53: timed out* *^[[B;; communications error to 127.0.0.53#53: timed out* *;; no servers could be reached* *google.com <http://google.com> mail is handled by 10 smtp.google.com <http://smtp.google.com>.* *Thu Sep 19 22:15:15 UTC 2024* *google.com <http://google.com> has address 172.253.115.113* *google.com <http://google.com> has address 172.253.115.138* *google.com <http://google.com> has address 172.253.115.101* *google.com <http://google.com> has address 172.253.115.100* *google.com <http://google.com> has address 172.253.115.102* *google.com <http://google.com> has address 172.253.115.139* *;; communications error to 127.0.0.53#53: timed out* *google.com <http://google.com> has IPv6 address 2607:f8b0:4004:c06::65* *google.com <http://google.com> has IPv6 address 2607:f8b0:4004:c06::8b* *google.com <http://google.com> has IPv6 address 2607:f8b0:4004:c06::64* *google.com <http://google.com> has IPv6 address 2607:f8b0:4004:c06::8a* *google.com <http://google.com> mail is handled by 10 smtp.google.com <http://smtp.google.com>.* *Thu Sep 19 22:15:26 UTC 2024* *;; communications error to 127.0.0.53#53: timed out* *google.com <http://google.com> has address 172.253.115.139* *Thu Sep 19 23:04:26 UTC 2024* *;; communications error to 127.0.0.53#53: timed out* This issue seems to occur intermittently, and while many requests succeed, others are timing out. I suspect the flow creation failure could be contributing to the DNS timeouts. Is this a known issue with OVN/OVS, or has anyone else experienced something similar? If so, are there any workarounds or solutions available to mitigate these errors? I would appreciate any insights or suggestions the community might have. Best regards, Srini
_______________________________________________ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
