[
https://issues.apache.org/jira/browse/DISPATCH-1732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17167825#comment-17167825
]
Gordon Sim commented on DISPATCH-1732:
--------------------------------------
The following fixes the segfault. Not sure if it is the correct fix:
{noformat}
diff --git a/src/policy.c b/src/policy.c
index ae093b13..7486dd7c 100644
--- a/src/policy.c
+++ b/src/policy.c
@@ -1294,7 +1294,7 @@ void qd_policy_amqp_open(qd_connection_t *qd_conn) {
if (cf && cf->multi_tenant) {
char vhost_name_buf[SETTINGS_NAME_SIZE];
if (qd_policy_lookup_vhost_alias(policy, vhost,
vhost_name_buf, SETTINGS_NAME_SIZE)) {
- if (!strcmp(pcrh, vhost_name_buf)) {
+ if (pcrh && !strcmp(pcrh, vhost_name_buf)) {
// Default condition: use proton connection value;
no action here
} else {
// Policy used a name different from what came in
the AMQP Open hostname.
{noformat}
> segfault when connecting to multitenant listener with policy but no hostname
> in open
> -------------------------------------------------------------------------------------
>
> Key: DISPATCH-1732
> URL: https://issues.apache.org/jira/browse/DISPATCH-1732
> Project: Qpid Dispatch
> Issue Type: Improvement
> Components: Policy Engine
> Affects Versions: 1.13.0
> Reporter: Gordon Sim
> Priority: Major
> Fix For: 1.13.0
>
>
> Start router with following config file and then run qpid-receive against it
> (which by default does not set a hostname in open).
>
> {noformat}
> [
> ["listener", { "host": "0.0.0.0", "port": 5672, "multiTenant": true ,
> "policyVhost": "myhost"}],
> ["policy", {"enableVhostPolicy": true }],
> ["vhost", { "hostname": "myhost", "allowUnknownUser": true,
> "groups":{"$default":{"remoteHosts":"*","sources":"*","targets":"*","allowDynamicSource":true,"allowAnonymousSender":true}}}]
> ] {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
