Title: kernel de Linux
Description: En el kernel de Linux, se resolvió la siguiente vulnerabilidad: gso: no omita el encabezado de IP externo en caso de ipip y net_failover. Nos encontramos con un problema de caída de TCP en nuestro entorno de nube. El paquete GROed en el host se reenvía a una NIC virtio_net de VM con net_failover habilitado. VM actúa como IPVS LB con encapsulación ipip. La ruta completa como: host gro -> vm virtio_net rx -> net_failover rx -> ipvs fullnat -> ipip encap -> net_failover tx -> virtio_net tx Cuando net_failover transmite un paquete ipip (gso_type = 0x0103, que significa SKB_GSO_TCPV4, SKB_GSO_DODGY y SKB_GSO_IPXIP 4 ), no existe gso porque admite TSO y GSO_IPXIP4. Pero network_header apunta al encabezado IP interno. Seguimiento de llamadas: tcp4_gso_segment ------> return NULL inet_gso_segment ------> iph interno, network_header apunta a ipip_gso_segment inet_gso_segment ------> iph externo skb_mac_gso_segment Luego, virtio_net transmite el paquete, solo se muestra el encabezado de IP interno modificado. Y el exterior simplemente se mantiene sin cambios. El paquete se colocará en el host remoto. Seguimiento de llamadas: inet_gso_segment ------> iph interno, se omite el iph externo skb_mac_gso_segment __skb_gso_segment validar_xmit_skb validar_xmit_skb_list sch_direct_xmit __qdisc_run __dev_queue_xmit ------> virtio_net dev_hard_start_xmit __dev_queue_xmit --- ---> net_failover ip_finish_output2 ip_output iptunnel_xmit ip_tunnel_xmit ipip_tunnel_xmit -- ----> ipip dev_hard_start_xmit __dev_queue_xmit ip_finish_output2 ip_output ip_forward ip_rcv __netif_receive_skb_one_core netif_receive_skb_internal napi_gro_receiveceived_buf virtnet_poll net_rx_action La causa raíz de este problema es específica de la rara combinación de SKB_GSO_DODGY y un dispositivo de túnel que agrega una opción de túnel SKB_GSO_. SKB_GSO_DODGY se configura desde virtio_net externo. Necesitamos restablecer el encabezado de la red cuando callbacks.gso_segment() devuelve NULL. Este parche también incluye ipv6_gso_segment(), considerando SIT, etc.

OR
     
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 5.16 from (excluding) 5.16.12
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 5.11 from (excluding) 5.15.26
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 5.5 from (excluding) 5.10.103
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 4.20 from (excluding) 5.4.182
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 4.15 from (excluding) 4.19.232
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 4.10 from (excluding) 4.14.269
          *cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* versions from (including) 3.13 from (excluding) 4.9.304

NIST AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

NIST NVD-CWE-noinfo

In the Linux kernel, the following vulnerability has been resolved:

gso: do not skip outer ip header in case of ipip and net_failover

We encounter a tcp drop issue in our cloud environment. Packet GROed in
host forwards to a VM virtio_net nic with net_failover enabled. VM acts
as a IPVS LB with ipip encapsulation. The full path like:
host gro -> vm virtio_net rx -> net_failover rx -> ipvs fullnat
 -> ipip encap -> net_failover tx -> virtio_net tx

When net_failover transmits a ipip pkt (gso_type = 0x0103, which means
SKB_GSO_TCPV4, SKB_GSO_DODGY and SKB_GSO_IPXIP4), there is no gso
did because it supports TSO and GSO_IPXIP4. But network_header points to
inner ip header.

Call Trace:
 tcp4_gso_segment        ------> return NULL
 inet_gso_segment        ------> inner iph, network_header points to
 ipip_gso_segment
 inet_gso_segment        ------> outer iph
 skb_mac_gso_segment

Afterwards virtio_net transmits the pkt, only inner ip header is modified.
And the outer one just keeps unchanged. The pkt will be dropped in remote
host.

Call Trace:
 inet_gso_segment        ------> inner iph, outer iph is skipped
 skb_mac_gso_segment
 __skb_gso_segment
 validate_xmit_skb
 validate_xmit_skb_list
 sch_direct_xmit
 __qdisc_run
 __dev_queue_xmit        ------> virtio_net
 dev_hard_start_xmit
 __dev_queue_xmit        ------> net_failover
 ip_finish_output2
 ip_output
 iptunnel_xmit
 ip_tunnel_xmit
 ipip_tunnel_xmit        ------> ipip
 dev_hard_start_xmit
 __dev_queue_xmit
 ip_finish_output2
 ip_output
 ip_forward
 ip_rcv
 __netif_receive_skb_one_core
 netif_receive_skb_internal
 napi_gro_receive
 receive_buf
 virtnet_poll
 net_rx_action

The root cause of this issue is specific with the rare combination of
SKB_GSO_DODGY and a tunnel device that adds an SKB_GSO_ tunnel option.
SKB_GSO_DODGY is set from external virtio_net. We need to reset network
header when callbacks.gso_segment() returns NULL.

This patch also includes ipv6_gso_segment(), considering SIT, etc.

Rejected reason: This CVE ID has been rejected or withdrawn by its CVE Numbering Authority.

kernel.org https://git.kernel.org/stable/c/2b3cdd70ea5f5a694f95ea1788393fb3b83071ea

kernel.org https://git.kernel.org/stable/c/45d006c2c7ed7baf1fa258fa7b5bc9923d3a983e

kernel.org https://git.kernel.org/stable/c/7840e559799a08a8588ee6de27516a991cb2e5e7

kernel.org https://git.kernel.org/stable/c/899e56a1ad435261812355550ae869d8be3df395

kernel.org https://git.kernel.org/stable/c/a739963f43269297c3f438b776194542e2a97499

kernel.org https://git.kernel.org/stable/c/cc20cced0598d9a5ff91ae4ab147b3b5e99ee819

kernel.org https://git.kernel.org/stable/c/dac2490d9ee0b89dffc72f1172b8bbeb60eaec39

kernel.org https://git.kernel.org/stable/c/e9ffbe63f6f32f526a461756309b61c395168d73