[Asterisk-code-review] res pjsip distributor: Don't send 503 response to responses. (asterisk[certified/13.1])
Mark Michelson
asteriskteam at digium.com
Thu Nov 12 11:21:29 CST 2015
Mark Michelson has uploaded a new change for review.
https://gerrit.asterisk.org/1618
Change subject: res_pjsip distributor: Don't send 503 response to responses.
......................................................................
res_pjsip distributor: Don't send 503 response to responses.
When the SIP threadpool is backed up with tasks, we send 503 responses
to ensure that we don't try to overload ourselves. The problem is that
we were not insuring that we were not trying to send a 503 to an
incoming SIP response.
This change makes it so that we only send the 503 on incoming requests.
Change-Id: Ie2b418d89c0e453cc6c2b5c7d543651c981e1404
---
M res/res_pjsip/pjsip_distributor.c
1 file changed, 3 insertions(+), 1 deletion(-)
git pull ssh://gerrit.asterisk.org:29418/asterisk refs/changes/18/1618/1
diff --git a/res/res_pjsip/pjsip_distributor.c b/res/res_pjsip/pjsip_distributor.c
index 3b6c262..32f6154 100644
--- a/res/res_pjsip/pjsip_distributor.c
+++ b/res/res_pjsip/pjsip_distributor.c
@@ -286,7 +286,9 @@
* 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);
+ if (rdata->msg_info.msg->type == PJSIP_REQUEST_MSG) {
+ 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 {
--
To view, visit https://gerrit.asterisk.org/1618
To unsubscribe, visit https://gerrit.asterisk.org/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: Ie2b418d89c0e453cc6c2b5c7d543651c981e1404
Gerrit-PatchSet: 1
Gerrit-Project: asterisk
Gerrit-Branch: certified/13.1
Gerrit-Owner: Mark Michelson <mmichelson at digium.com>
More information about the asterisk-code-review
mailing list