(gdb) frame 3
#3 0x0000560855ec770f in HttpTransact::is_stale_cache_response_returnable(HttpTransact::State*) (s=0x7f898856f8a0) at HttpTransact.cc:6034
6034 if (!s->cache_info.directives.does_client_permit_lookup) {
(gdb) p s
$1 = (struct State *) 0x7f898856f8a0
(gdb) p *s
$2 = <incomplete type>
(gdb) p s->cache_info
There is no member named cache_info.
On frame 4, s->cache_info.directives.does_client_permit_lookup is accessible and it's true
(gdb) frame 4
#4 0x0000560855ec50cc in HttpTransact::OSDNSLookup(HttpTransact::State*) (s=0x7f898856f8a0) at HttpTransact.cc:1951
1951 if (is_cache_hit(s->cache_lookup_result) && is_stale_cache_response_returnable(s)) {
(gdb) p s->cache_info.directives.does_client_permit_lookup
$7 = true
This issue has been automatically marked as stale because it has not had recent activity. Marking it stale to flag it for further consideration by the community.
This crash happens intermittently, so this doesn't block the 9.2.0 release, IMO.
Frame 3 reports that the SEGV happens on this line, but it's a bit odd. https://github.com/apache/trafficserver/blob/c26d71cc35acff7ad8a9a98de45433305be9318c/proxy/http/HttpTransact.cc#L6034
On frame 4,
s->cache_info.directives.does_client_permit_lookup
is accessible and it'strue