The flowd daemon on the primary node of an SRX Series chassis cluster may crash and restart when attempting to synchronize a multicast session created via crafted multicast packets. Upon the flowd crash, data plane redundancy groups will fail over to the secondary node in the chassis cluster while flowd on the primary node restarts. This issue only occurs in chassis cluster configurations that process transit multicast traffic. Transit multicast traffic is processed on an SRX services gateway by enabling PIM in normal Flow Mode, or via security policies permitting transit multicast traffic in L2/Transparent Mode. Juniper SIRT is not aware of any malicious exploitation of this vulnerability. No other Juniper Networks products or platforms are affected by this issue. This issue has been assigned CVE-2017-2300 .
The following software releases have been updated to resolve this specific issue: Junos OS 12.1X46-D65, 12.3X48-D40, 15.1X49-D60, and all subsequent releases. This issue is being tracked as PR 1188853 and is visible on the Customer Support website. KB16765 [juniper.net] - "In which releases are vulnerabilities fixed?" describes which release vulnerabilities are fixed as per our End of Engineering and End of Life support policies.
2017-01-11: Initial publication 2017-01-18: Devices with security policies permitting transit multicast traffic in Transparent Mode are also vulnerable
Disallow transit multicast traffic, or isolate/disable the secondary node of the chassis cluster until an upgrade can be performed.