- Fix a couple of segfaults that may happen on reload

This commit is contained in:
Simo Sorce 2009-06-22 14:49:00 +00:00
parent d0eb246884
commit 7522c31552
3 changed files with 157 additions and 1 deletions

View File

@ -0,0 +1,113 @@
From 673c2ce9b3371241de872b2bd206f732485888cb Mon Sep 17 00:00:00 2001
From: Stephen Gallagher <sgallagh@redhat.com>
Date: Fri, 19 Jun 2009 11:09:33 -0400
Subject: [PATCH] Fix segfault in update_monitor_config
We were stealing the memory context of only the first value in
the linked-list of domains (and also services). This patch adds a
memory context to hold the lists so that can be stolen along with
all of the entries.
---
server/confdb/confdb.c | 4 ++++
server/monitor/monitor.c | 34 ++++++++++++++++++++++++++--------
2 files changed, 30 insertions(+), 8 deletions(-)
diff --git a/server/confdb/confdb.c b/server/confdb/confdb.c
index 8eefcfb..8b8dc74 100644
--- a/server/confdb/confdb.c
+++ b/server/confdb/confdb.c
@@ -709,6 +709,10 @@ int confdb_get_domain(struct confdb_ctx *cdb,
}
domain = talloc_zero(mem_ctx, struct sss_domain_info);
+ if (!domain) {
+ ret = ENOMEM;
+ goto done;
+ }
tmp = ldb_msg_find_attr_as_string(res->msgs[0], "cn", NULL);
if (!tmp) {
diff --git a/server/monitor/monitor.c b/server/monitor/monitor.c
index 906d157..e4fca25 100644
--- a/server/monitor/monitor.c
+++ b/server/monitor/monitor.c
@@ -84,7 +84,9 @@ struct mt_svc {
struct mt_ctx {
struct tevent_context *ev;
struct confdb_ctx *cdb;
+ TALLOC_CTX *domain_ctx; /* Memory context for domain list */
struct sss_domain_info *domains;
+ TALLOC_CTX *service_ctx; /* Memory context for services */
char **services;
struct mt_svc *svc_list;
struct sbus_srv_ctx *sbus_srv;
@@ -619,8 +621,13 @@ int get_monitor_config(struct mt_ctx *ctx)
return ret;
}
- ret = confdb_get_string_as_list(ctx->cdb, ctx, SERVICE_CONF_ENTRY,
- "activeServices", &ctx->services);
+ ctx->service_ctx = talloc_new(ctx);
+ if(!ctx->service_ctx) {
+ return ENOMEM;
+ }
+ ret = confdb_get_string_as_list(ctx->cdb, ctx->service_ctx,
+ SERVICE_CONF_ENTRY, "activeServices",
+ &ctx->services);
if (ret != EOK) {
DEBUG(0, ("No services configured!\n"));
return EINVAL;
@@ -631,7 +638,11 @@ int get_monitor_config(struct mt_ctx *ctx)
return ret;
}
- ret = confdb_get_domains(ctx->cdb, ctx, &ctx->domains);
+ ctx->domain_ctx = talloc_new(ctx);
+ if(!ctx->domain_ctx) {
+ return ENOMEM;
+ }
+ ret = confdb_get_domains(ctx->cdb, ctx->domain_ctx, &ctx->domains);
if (ret != EOK) {
DEBUG(2, ("No domains configured. LOCAL should always exist!\n"));
return ret;
@@ -861,7 +872,12 @@ static int update_monitor_config(struct mt_ctx *ctx)
struct mt_svc *cur_svc;
struct mt_svc *new_svc;
struct sss_domain_info *dom, *new_dom;
- struct mt_ctx *new_config = talloc_zero(NULL, struct mt_ctx);
+ struct mt_ctx *new_config;
+
+ new_config = talloc_zero(NULL, struct mt_ctx);
+ if(!new_config) {
+ return ENOMEM;
+ }
new_config->ev = ctx->ev;
new_config->cdb = ctx->cdb;
@@ -953,8 +969,9 @@ static int update_monitor_config(struct mt_ctx *ctx)
}
/* Replace the old service list with the new one */
- talloc_free(ctx->services);
- ctx->services = talloc_steal(ctx, new_config->services);
+ talloc_free(ctx->service_ctx);
+ ctx->service_ctx = talloc_steal(ctx, new_config->service_ctx);
+ ctx->services = new_config->services;
/* Compare data providers */
/* Have any providers been disabled? */
@@ -1040,8 +1057,9 @@ static int update_monitor_config(struct mt_ctx *ctx)
}
/* Replace the old domain list with the new one */
- talloc_free(ctx->domains);
- ctx->domains = talloc_steal(ctx, new_config->domains);
+ talloc_free(ctx->domain_ctx);
+ ctx->domain_ctx = talloc_steal(ctx, new_config->domain_ctx);
+ ctx->domains = new_config->domains;
/* Signal all services to reload their configuration */
for(cur_svc = ctx->svc_list; cur_svc; cur_svc = cur_svc->next) {
--
1.6.2.2

View File

@ -0,0 +1,36 @@
From 12cbba5545aefa59e27f683e17e05b8e80063718 Mon Sep 17 00:00:00 2001
From: Stephen Gallagher <sgallagh@redhat.com>
Date: Fri, 19 Jun 2009 11:28:49 -0400
Subject: [PATCH] Protect against segfault in service_signal_reload
There is a potential race condition where the monitor may attempt
to signal a reload of a child process before the communication
sbus channel is available. If this happens, we will just exit this
function and let the monitor kill and restart the child process.
---
server/monitor/monitor.c | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/server/monitor/monitor.c b/server/monitor/monitor.c
index e4fca25..5cc73c8 100644
--- a/server/monitor/monitor.c
+++ b/server/monitor/monitor.c
@@ -525,6 +525,15 @@ static int service_signal_reload(struct mt_svc *svc)
return EOK;
}
+ if (!svc->mt_conn) {
+ /* Avoid a race condition where we are trying to
+ * order a service to reload that hasn't started
+ * yet.
+ */
+ DEBUG(1,("Could not reload service [%s].\n", svc->name));
+ return EIO;
+ }
+
conn = sbus_get_connection(svc->mt_conn->conn_ctx);
msg = dbus_message_new_method_call(NULL,
SERVICE_PATH,
--
1.6.2.2

View File

@ -1,6 +1,6 @@
Name: sssd
Version: 0.4.1
Release: 1%{?dist}
Release: 2%{?dist}
Group: Applications/System
Summary: System Security Services Daemon
@ -14,6 +14,8 @@ BuildRoot: %(mktemp -ud %{_tmppath}/%{name}-%{version}-%{release}-XXXXXX)
### Patches ###
Patch010: sssd-0.4.1-debug_fn.patch
Patch011: sssd-0.4.1-conf_check.patch
Patch012: sssd-0.4.1-reload_conf.patch
Patch013: sssd-0.4.1-reload_conf_2.patch
### Dependencies ###
@ -58,6 +60,8 @@ services for projects like FreeIPA.
%patch010 -p1 -b .debug_fn
%patch011 -p1 -b .conf_check
%patch012 -p1 -b .reload_conf
%patch013 -p1 -b .reload_conf_2
%build
%configure \
@ -131,6 +135,9 @@ if [ $1 -ge 1 ] ; then
fi
%changelog
* Mon Jun 22 2009 Simo Sorce <ssorce@redhat.com> - 0.4.1-2
- Fix a couple of segfaults that may happen on reload
* Thu Jun 11 2009 Simo Sorce <ssorce@redhat.com> - 0.4.1-1
- add missing configure check that broke stopping the daemon
- also fix default config to add a missing required option