BUG/MAJOR: leastconn: do not loop forever when facing saturated servers

Since commit 9fe72bba3 ("MAJOR: leastconn; Revamp the way servers are
ordered."), there's no way to escape the loop visiting the mt_list heads
in fwlc_get_next_server if all servers in the list are saturated,
resulting in a watchdog panic. It can be reproduced with this config
and injecting with more than 2 concurrent conns:

    balance leastconn
    server s1 127.0.0.1:8000 maxconn 1
    server s2 127.0.0.1:8000 maxconn 1

Here we count the number of saturated servers that were encountered, and
escape the loop once the number of remaining servers exceeds the number
of saturated ones. No backport is needed since this arrived in 3.2.

diff --git a/src/lb_fwlc.c b/src/lb_fwlc.c
index 1c5717ba69190d36a5f1f654269da4bc53b7efb9..9023a7aaf6ad25296c4f7adfdda0fa84ae401dd5 100644 (file)
--- a/src/lb_fwlc.c
+++ b/src/lb_fwlc.c
@@ -762,13 +762,14 @@ redo:
        while (node) {
                struct fwlc_tree_elt *tree_elt;
                struct server *s;
+               int unusable = 0;
                int orig_nb;
                int i = 0;
 
                tree_elt = eb32_entry(node, struct fwlc_tree_elt, lb_node);
                orig_nb = statistical_prng_range(FWLC_LISTS_NB);
 
-               while (_HA_ATOMIC_LOAD(&tree_elt->elements) > 0) {
+               while (_HA_ATOMIC_LOAD(&tree_elt->elements) > unusable) {
                        struct mt_list mt_list;
                        mt_list.next = _HA_ATOMIC_LOAD(&tree_elt->srv_list[(i + orig_nb) % FWLC_LISTS_NB].next);
 
@@ -802,6 +803,8 @@ redo:
                                         }
                                        avoided = s;
                                }
+                               else
+                                       unusable++;
                                i++;
                        } else if (mt_list.next == &tree_elt->srv_list[(i + orig_nb) % FWLC_LISTS_NB]) {
                                i++;