On Junos devices with the BGP graceful restart helper mode enabled or the BGP graceful restart mechanism enabled, a certain sequence of BGP session restart on a remote peer that has the graceful restart mechanism enabled may cause the local routing protocol daemon (RPD) process to crash and restart. Repeated crashes of the RPD process can cause prolonged Denial of Service (DoS). Graceful restart helper mode for BGP is enabled by default. No other Juniper Networks products or platforms are affected by this issue. Affected releases are Juniper Networks Junos OS: 16.1 versions prior to 16.1R7-S3; 16.2 versions prior to 16.2R2-S9; 17.1 versions prior to 17.1R3; 17.2 versions prior to 17.2R3; 17.2X75 versions prior to 17.2X75-D105; 17.3 versions prior to 17.3R3-S2; 17.4 versions prior to 17.4R1
The following software releases have been updated to resolve this specific issue: 16.1R7-S3, 16.2R2-S9, 17.1R3, 17.2R3, 17.2X75-D105, 17.3R3-S2, 17.4R1-S7, 17.4R2-S2, 17.4R3, 18.1R3-S2, 18.2R2, 18.2X75-D12, 18.2X75-D30, 18.3R1-S4, 18.3R2, 18.4R1, and all subsequent releases.
This issue can be prevented by disabling the BGP graceful restart mechanism, including graceful restart helper mode: [protocols bgp graceful-restart disable] Furthermore, the risk associated with this issue can be mitigated by limiting BGP sessions only from trusted peers.
| Product | Vulnerable range | Fixed version | Advisory |
|---|---|---|---|
| Juniper Networks Junos OS | — | not specified | advisory ↗ |