res_pjsip: Deny requests when threadpool queue is backed up.

We have observed situations where the SIP threadpool may become
deadlocked. However, because incoming traffic is still arriving, the SIP
threadpool's queue can continue to grow, eventually running the system
out of memory.

This change makes it so that incoming traffic gets rejected with a 503
response if the queue is backed up too much.

Change-Id: I4e736d48a2ba79fd1f8056c0dcd330e38e6a3816

diff --git a/include/asterisk/res_pjsip.h b/include/asterisk/res_pjsip.h
index af11ea4b6fb7c2abc510f7447b1f18a18394a205..0ce50912c038c3956a02e5afb12d9655c754a796 100644 (file)
--- a/include/asterisk/res_pjsip.h
+++ b/include/asterisk/res_pjsip.h
@@ -2105,4 +2105,10 @@ int ast_sip_get_host_ip(int af, pj_sockaddr *addr);
  */
 const char *ast_sip_get_host_ip_string(int af);
 
+/*!
+ * \brief Return the size of the SIP threadpool's task queue
+ * \since 13.7.0
+ */
+long ast_sip_threadpool_queue_size(void);
+
 #endif /* _RES_PJSIP_H */

diff --git a/include/asterisk/taskprocessor.h b/include/asterisk/taskprocessor.h
index f16f144cb558ef9fcdc130a4ce91da89d8987dfc..06368867a8f309d24666b1a7941d6fa3380a4108 100644 (file)
--- a/include/asterisk/taskprocessor.h
+++ b/include/asterisk/taskprocessor.h
@@ -262,4 +262,10 @@ int ast_taskprocessor_is_task(struct ast_taskprocessor *tps);
  */
 const char *ast_taskprocessor_name(struct ast_taskprocessor *tps);
 
+/*!
+ * \brief Return the current size of the taskprocessor queue
+ * \since 13.7.0
+ */
+long ast_taskprocessor_size(struct ast_taskprocessor *tps);
+
 #endif /* __AST_TASKPROCESSOR_H__ */

diff --git a/include/asterisk/threadpool.h b/include/asterisk/threadpool.h
index 1c6705856b1b7a03f1a2c6e40eb225303838c8d1..9a09d17d4466dc69b9122cbc92d7c10346a25dbb 100644 (file)
--- a/include/asterisk/threadpool.h
+++ b/include/asterisk/threadpool.h
@@ -239,4 +239,10 @@ struct ast_taskprocessor *ast_threadpool_serializer_get_current(void);
  */
 struct ast_taskprocessor *ast_threadpool_serializer(const char *name, struct ast_threadpool *pool);
 
+/*!
+ * \brief Return the size of the threadpool's task queue
+ * \since 13.7.0
+ */
+long ast_threadpool_queue_size(struct ast_threadpool *pool);
+
 #endif /* ASTERISK_THREADPOOL_H */

diff --git a/main/taskprocessor.c b/main/taskprocessor.c
index 227fc3e81f14335496c9b501715406370a87d11e..cdac2c8a002fbb1e7a759993b4e258d9e6da9143 100644 (file)
--- a/main/taskprocessor.c
+++ b/main/taskprocessor.c
@@ -130,9 +130,6 @@ static int tps_ping_handler(void *datap);
 /*! \brief Remove the front task off the taskprocessor queue */
 static struct tps_task *tps_taskprocessor_pop(struct ast_taskprocessor *tps);
 
-/*! \brief Return the size of the taskprocessor queue */
-static int tps_taskprocessor_depth(struct ast_taskprocessor *tps);
-
 static char *cli_tps_ping(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a);
 static char *cli_tps_report(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a);
 
@@ -508,7 +505,7 @@ static struct tps_task *tps_taskprocessor_pop(struct ast_taskprocessor *tps)
        return task;
 }
 
-static int tps_taskprocessor_depth(struct ast_taskprocessor *tps)
+long ast_taskprocessor_size(struct ast_taskprocessor *tps)
 {
        return (tps) ? tps->tps_queue_size : -1;
 }
@@ -766,7 +763,7 @@ int ast_taskprocessor_execute(struct ast_taskprocessor *tps)
 {
        struct ast_taskprocessor_local local;
        struct tps_task *t;
-       int size;
+       long size;
 
        ao2_lock(tps);
        t = tps_taskprocessor_pop(tps);
@@ -798,7 +795,7 @@ int ast_taskprocessor_execute(struct ast_taskprocessor *tps)
         * after we pop an empty stack.
         */
        tps->executing = 0;
-       size = tps_taskprocessor_depth(tps);
+       size = ast_taskprocessor_size(tps);
        /* If we executed a task, bump the stats */
        if (tps->stats) {
                tps->stats->_tasks_processed_count++;

diff --git a/main/threadpool.c b/main/threadpool.c
index 229528c030bfbc6bce28db6587bdec70720b9bd7..81d76c4457ecc4ff2c84b4e367bb4da2290846cf 100644 (file)
--- a/main/threadpool.c
+++ b/main/threadpool.c
@@ -1276,3 +1276,8 @@ struct ast_taskprocessor *ast_threadpool_serializer(const char *name, struct ast
 
        return tps;
 }
+
+long ast_threadpool_queue_size(struct ast_threadpool *pool)
+{
+       return ast_taskprocessor_size(pool->tps);
+}

diff --git a/res/res_pjsip.c b/res/res_pjsip.c
index 0b541afe989fe3fd9bf7d79787ce32d6815aed0a..d5b68bccdee8de0faeb5d45f102c67a78a06e20c 100644 (file)
--- a/res/res_pjsip.c
+++ b/res/res_pjsip.c
@@ -3408,6 +3408,11 @@ static void remove_request_headers(pjsip_endpoint *endpt)
        }
 }
 
+long ast_sip_threadpool_queue_size(void)
+{
+       return ast_threadpool_queue_size(sip_threadpool);
+}
+
 AST_TEST_DEFINE(xml_sanitization_end_null)
 {
        char sanitized[8];

diff --git a/res/res_pjsip/pjsip_distributor.c b/res/res_pjsip/pjsip_distributor.c
index 1294379bcb9778dd67853d7937331294504a3780..3b6c262ecd09a94d02546fdc1a07fe905e2cb734 100644 (file)
--- a/res/res_pjsip/pjsip_distributor.c
+++ b/res/res_pjsip/pjsip_distributor.c
@@ -245,6 +245,8 @@ static pjsip_module endpoint_mod = {
        .on_rx_request = endpoint_lookup,
 };
 
+#define SIP_MAX_QUEUE 500L
+
 static pj_bool_t distributor(pjsip_rx_data *rdata)
 {
        pjsip_dialog *dlg = find_dialog(rdata);
@@ -279,7 +281,17 @@ static pj_bool_t distributor(pjsip_rx_data *rdata)
                clone->endpt_info.mod_data[endpoint_mod.id] = ao2_bump(dist->endpoint);
        }
 
-       ast_sip_push_task(serializer, distribute, clone);
+       if (ast_sip_threadpool_queue_size() > SIP_MAX_QUEUE) {
+               /* When the threadpool is backed up this much, there is a good chance that we have encountered
+                * some sort of terrible condition and don't need to be adding more work to the threadpool.
+                * It's in our best interest to send back a 503 response and be done with it.
+                */
+               pjsip_endpt_respond_stateless(ast_sip_get_pjsip_endpoint(), rdata, 503, NULL, NULL, NULL);
+               ao2_cleanup(clone->endpt_info.mod_data[endpoint_mod.id]);
+               pjsip_rx_data_free_cloned(clone);
+       } else {
+               ast_sip_push_task(serializer, distribute, clone);
+       }
 
 end:
        if (dlg) {