Initial description: Router PODs frequently getting restarted and haproxy process is receiving the segfault but it is not generating coredump even though the core file size is unlimited.

Upstream bug: https://github.com/haproxy/haproxy/issues/1972

An issue was discovered in HAProxy before 1.8.8. The incoming H2 frame length was checked against the maxframesize setting instead of being checked against the bufsize. The maxframesize only applies to outgoing traffic and not to incoming, so if a large enough frame size is advertised in the SETTINGS frame, a wrapped frame will be defragmented into a temporary allocated buffer where the second fragment may overflow the heap by up to 16 kB. It is very unlikely that this can be exploited for code execution given that buffers are very short lived and their addresses not realistically predictable in production, but the likelihood of an immediate crash is absolutely certain.

HAProxy statistics in openstack-tripleo-image-elements are non-authenticated over the network.

It was found that HAProxy statistics are non-authenticated over network.

elements/haproxy/os-apply-config/etc/haproxy/haproxy.cfg:

listen haproxy.stats :{{#stats.port}}{{stats.port}}{{/stats.port}}{{^stats.port}}1993{{/stats.port}}

Multiple integer overflows in the httprequestforwardbody function in protohttp.c in HAProxy 1.5-dev23 before 1.5.4 allow remote attackers to cause a denial of service (crash) via a large stream of data, which triggers a buffer overflow and an out-of-bounds read.

Buffer overflow in HAProxy 1.4 through 1.4.22 and 1.5-dev through 1.5-dev17, when HTTP keep-alive is enabled, using HTTP keywords in TCP inspection rules, and running with rewrite rules that appends to requests, allows remote attackers to cause a denial of service (crash) and possibly execute arbitrary code via crafted pipelined HTTP requests that prevent request realignment from occurring.