Changes with Apache 2.4.24
+ *) core: avoid adding multiple subrequest filters when there are nested
+ subrequests. PR58292
+
*) mod_http2: connection shutdown revisited: corrected edge cases on
shutting down ongoing streams, changed log warnings to be less noisy
when waiting on long running tasks. [Stefan Eissing]
PATCHES ACCEPTED TO BACKPORT FROM TRUNK:
[ start all new proposals below, under PATCHES PROPOSED. ]
- *) core: avoid adding multiple subrequest filters when there are nested
- subrequests. PR58292
- trunk patch: http://svn.apache.org/r1698239
- 2.4.x patch: trunk works
- +1: covener, ylavic, jim
-
*) mod_remoteip: Pick up where we left off during a subrequest rather
than running with the modified XFF but original TCP address. PR49839/PR60251
trunk patch: http://svn.apache.org/r1688399
update_r_in_filters(r->output_filters, rr, r);
if (r->main) {
- ap_add_output_filter_handle(ap_subreq_core_filter_handle,
- NULL, r, r->connection);
+ ap_filter_t *next = r->output_filters;
+ while (next && (next != r->proto_output_filters)) {
+ if (next->frec == ap_subreq_core_filter_handle) {
+ break;
+ }
+ }
+ if (!next || next == r->proto_output_filters) {
+ ap_add_output_filter_handle(ap_subreq_core_filter_handle,
+ NULL, r, r->connection);
+ }
}
else {
/*
/* start with the same set of output filters */
if (next_filter) {
+ ap_filter_t *scan = next_filter;
+
/* while there are no input filters for a subrequest, we will
* try to insert some, so if we don't have valid data, the code
* will seg fault.
rnew->proto_input_filters = r->proto_input_filters;
rnew->output_filters = next_filter;
rnew->proto_output_filters = r->proto_output_filters;
- ap_add_output_filter_handle(ap_subreq_core_filter_handle,
- NULL, rnew, rnew->connection);
+ while (scan && (scan != r->proto_output_filters)) {
+ if (scan->frec == ap_subreq_core_filter_handle) {
+ break;
+ }
+ }
+ if (!scan || scan == r->proto_output_filters) {
+ ap_add_output_filter_handle(ap_subreq_core_filter_handle,
+ NULL, rnew, rnew->connection);
+ }
}
else {
/* If NULL - we are expecting to be internal_fast_redirect'ed