aboutsummaryrefslogtreecommitdiff
path: root/contrib/bind9/bin/named/server.c
diff options
context:
space:
mode:
Diffstat (limited to 'contrib/bind9/bin/named/server.c')
-rw-r--r--contrib/bind9/bin/named/server.c2454
1 files changed, 2184 insertions, 270 deletions
diff --git a/contrib/bind9/bin/named/server.c b/contrib/bind9/bin/named/server.c
index bc7fc17c3296..5bbf94b9b604 100644
--- a/contrib/bind9/bin/named/server.c
+++ b/contrib/bind9/bin/named/server.c
@@ -15,7 +15,7 @@
* PERFORMANCE OF THIS SOFTWARE.
*/
-/* $Id: server.c,v 1.520.12.21 2011-01-14 23:45:49 tbox Exp $ */
+/* $Id: server.c,v 1.599.8.4 2011-02-16 19:46:12 each Exp $ */
/*! \file */
@@ -23,6 +23,10 @@
#include <stdlib.h>
#include <unistd.h>
+#include <limits.h>
+#include <ctype.h>
+#include <sys/types.h>
+#include <sys/stat.h>
#include <isc/app.h>
#include <isc/base64.h>
@@ -36,7 +40,9 @@
#include <isc/portset.h>
#include <isc/print.h>
#include <isc/resource.h>
+#include <isc/sha2.h>
#include <isc/socket.h>
+#include <isc/stat.h>
#include <isc/stats.h>
#include <isc/stdio.h>
#include <isc/string.h>
@@ -57,9 +63,11 @@
#ifdef DLZ
#include <dns/dlz.h>
#endif
+#include <dns/dns64.h>
#include <dns/forward.h>
#include <dns/journal.h>
#include <dns/keytable.h>
+#include <dns/keyvalues.h>
#include <dns/lib.h>
#include <dns/master.h>
#include <dns/masterdump.h>
@@ -102,6 +110,10 @@
#include <stdlib.h>
#endif
+#ifndef PATH_MAX
+#define PATH_MAX 1024
+#endif
+
/*%
* Check an operation for failure. Assumes that the function
* using it has a 'result' variable and a 'cleanup' label.
@@ -143,6 +155,14 @@
fatal(msg, result); \
} while (0) \
+/*%
+ * Maximum ADB size for views that share a cache. Use this limit to suppress
+ * the total of memory footprint, which should be the main reason for sharing
+ * a cache. Only effective when a finite max-cache-size is specified.
+ * This is currently defined to be 8MB.
+ */
+#define MAX_ADB_SIZE_FOR_CACHESHARE 8388608
+
struct ns_dispatch {
isc_sockaddr_t addr;
unsigned int dispatchgen;
@@ -150,6 +170,14 @@ struct ns_dispatch {
ISC_LINK(struct ns_dispatch) link;
};
+struct ns_cache {
+ dns_cache_t *cache;
+ dns_view_t *primaryview;
+ isc_boolean_t needflush;
+ isc_boolean_t adbsizeadjusted;
+ ISC_LINK(ns_cache_t) link;
+};
+
struct dumpcontext {
isc_mem_t *mctx;
isc_boolean_t dumpcache;
@@ -176,6 +204,17 @@ struct zonelistentry {
ISC_LINK(struct zonelistentry) link;
};
+/*%
+ * Configuration context to retain for each view that allows
+ * new zones to be added at runtime
+ */
+struct cfg_context {
+ isc_mem_t * mctx;
+ cfg_obj_t * config;
+ cfg_parser_t * parser;
+ cfg_aclconfctx_t actx;
+};
+
/*
* These zones should not leak onto the Internet.
*/
@@ -230,8 +269,8 @@ static const struct {
{ NULL, ISC_FALSE }
};
-static void
-fatal(const char *msg, isc_result_t result);
+ISC_PLATFORM_NORETURN_PRE static void
+fatal(const char *msg, isc_result_t result) ISC_PLATFORM_NORETURN_POST;
static void
ns_server_reload(isc_task_t *task, isc_event_t *event);
@@ -256,19 +295,25 @@ configure_alternates(const cfg_obj_t *config, dns_view_t *view,
static isc_result_t
configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
const cfg_obj_t *vconfig, isc_mem_t *mctx, dns_view_t *view,
- cfg_aclconfctx_t *aclconf);
+ cfg_aclconfctx_t *aclconf, isc_boolean_t added);
+
+static isc_result_t
+add_keydata_zone(dns_view_t *view, const char *directory, isc_mem_t *mctx);
static void
end_reserved_dispatches(ns_server_t *server, isc_boolean_t all);
+static void
+cfgctx_destroy(void **cfgp);
+
/*%
* Configure a single view ACL at '*aclp'. Get its configuration from
* 'vconfig' (for per-view configuration) and maybe from 'config'
*/
static isc_result_t
configure_view_acl(const cfg_obj_t *vconfig, const cfg_obj_t *config,
- const char *aclname, cfg_aclconfctx_t *actx,
- isc_mem_t *mctx, dns_acl_t **aclp)
+ const char *aclname, const char *acltuplename,
+ cfg_aclconfctx_t *actx, isc_mem_t *mctx, dns_acl_t **aclp)
{
isc_result_t result;
const cfg_obj_t *maps[3];
@@ -294,13 +339,21 @@ configure_view_acl(const cfg_obj_t *vconfig, const cfg_obj_t *config,
*/
return (ISC_R_SUCCESS);
+ if (acltuplename != NULL) {
+ /*
+ * If the ACL is given in an optional tuple, retrieve it.
+ * The parser should have ensured that a valid object be
+ * returned.
+ */
+ aclobj = cfg_tuple_get(aclobj, acltuplename);
+ }
+
result = cfg_acl_fromconfig(aclobj, config, ns_g_lctx,
actx, mctx, 0, aclp);
return (result);
}
-
/*%
* Configure a sortlist at '*aclp'. Essentially the same as
* configure_view_acl() except it calls cfg_acl_fromconfig with a
@@ -345,8 +398,88 @@ configure_view_sortlist(const cfg_obj_t *vconfig, const cfg_obj_t *config,
}
static isc_result_t
-configure_view_dnsseckey(const cfg_obj_t *vconfig, const cfg_obj_t *key,
- dns_keytable_t *keytable, isc_mem_t *mctx)
+configure_view_nametable(const cfg_obj_t *vconfig, const cfg_obj_t *config,
+ const char *confname, const char *conftuplename,
+ isc_mem_t *mctx, dns_rbt_t **rbtp)
+{
+ isc_result_t result;
+ const cfg_obj_t *maps[3];
+ const cfg_obj_t *obj = NULL;
+ const cfg_listelt_t *element;
+ int i = 0;
+ dns_fixedname_t fixed;
+ dns_name_t *name;
+ isc_buffer_t b;
+ const char *str;
+ const cfg_obj_t *nameobj;
+
+ if (*rbtp != NULL)
+ dns_rbt_destroy(rbtp);
+ if (vconfig != NULL)
+ maps[i++] = cfg_tuple_get(vconfig, "options");
+ if (config != NULL) {
+ const cfg_obj_t *options = NULL;
+ (void)cfg_map_get(config, "options", &options);
+ if (options != NULL)
+ maps[i++] = options;
+ }
+ maps[i] = NULL;
+
+ (void)ns_config_get(maps, confname, &obj);
+ if (obj == NULL)
+ /*
+ * No value available. *rbtp == NULL.
+ */
+ return (ISC_R_SUCCESS);
+
+ if (conftuplename != NULL) {
+ obj = cfg_tuple_get(obj, conftuplename);
+ if (cfg_obj_isvoid(obj))
+ return (ISC_R_SUCCESS);
+ }
+
+ result = dns_rbt_create(mctx, NULL, NULL, rbtp);
+ if (result != ISC_R_SUCCESS)
+ return (result);
+
+ dns_fixedname_init(&fixed);
+ name = dns_fixedname_name(&fixed);
+ for (element = cfg_list_first(obj);
+ element != NULL;
+ element = cfg_list_next(element)) {
+ nameobj = cfg_listelt_value(element);
+ str = cfg_obj_asstring(nameobj);
+ isc_buffer_init(&b, str, strlen(str));
+ isc_buffer_add(&b, strlen(str));
+ CHECK(dns_name_fromtext(name, &b, dns_rootname, 0, NULL));
+ /*
+ * We don't need the node data, but need to set dummy data to
+ * avoid a partial match with an empty node. For example, if
+ * we have foo.example.com and bar.example.com, we'd get a match
+ * for baz.example.com, which is not the expected result.
+ * We simply use (void *)1 as the dummy data.
+ */
+ result = dns_rbt_addname(*rbtp, name, (void *)1);
+ if (result != ISC_R_SUCCESS) {
+ cfg_obj_log(nameobj, ns_g_lctx, ISC_LOG_ERROR,
+ "failed to add %s for %s: %s",
+ str, confname, isc_result_totext(result));
+ goto cleanup;
+ }
+
+ }
+
+ return (result);
+
+ cleanup:
+ dns_rbt_destroy(rbtp);
+ return (result);
+
+}
+
+static isc_result_t
+dstkey_fromconfig(const cfg_obj_t *vconfig, const cfg_obj_t *key,
+ isc_boolean_t managed, dst_key_t **target, isc_mem_t *mctx)
{
dns_rdataclass_t viewclass;
dns_rdata_dnskey_t keystruct;
@@ -363,12 +496,28 @@ configure_view_dnsseckey(const cfg_obj_t *vconfig, const cfg_obj_t *key,
isc_result_t result;
dst_key_t *dstkey = NULL;
+ INSIST(target != NULL && *target == NULL);
+
flags = cfg_obj_asuint32(cfg_tuple_get(key, "flags"));
proto = cfg_obj_asuint32(cfg_tuple_get(key, "protocol"));
alg = cfg_obj_asuint32(cfg_tuple_get(key, "algorithm"));
keyname = dns_fixedname_name(&fkeyname);
keynamestr = cfg_obj_asstring(cfg_tuple_get(key, "name"));
+ if (managed) {
+ const char *initmethod;
+ initmethod = cfg_obj_asstring(cfg_tuple_get(key, "init"));
+
+ if (strcasecmp(initmethod, "initial-key") != 0) {
+ cfg_obj_log(key, ns_g_lctx, ISC_LOG_ERROR,
+ "managed key '%s': "
+ "invalid initialization method '%s'",
+ keynamestr, initmethod);
+ result = ISC_R_FAILURE;
+ goto cleanup;
+ }
+ }
+
if (vconfig == NULL)
viewclass = dns_rdataclass_in;
else {
@@ -408,7 +557,8 @@ configure_view_dnsseckey(const cfg_obj_t *vconfig, const cfg_obj_t *key,
keystruct.algorithm == DST_ALG_RSAMD5) &&
r.length > 1 && r.base[0] == 1 && r.base[1] == 3)
cfg_obj_log(key, ns_g_lctx, ISC_LOG_WARNING,
- "trusted key '%s' has a weak exponent",
+ "%s key '%s' has a weak exponent",
+ managed ? "managed" : "trusted",
keynamestr);
CHECK(dns_rdata_fromstruct(NULL,
@@ -418,25 +568,28 @@ configure_view_dnsseckey(const cfg_obj_t *vconfig, const cfg_obj_t *key,
dns_fixedname_init(&fkeyname);
isc_buffer_init(&namebuf, keynamestr, strlen(keynamestr));
isc_buffer_add(&namebuf, strlen(keynamestr));
- CHECK(dns_name_fromtext(keyname, &namebuf,
- dns_rootname, ISC_FALSE,
- NULL));
+ CHECK(dns_name_fromtext(keyname, &namebuf, dns_rootname, 0, NULL));
CHECK(dst_key_fromdns(keyname, viewclass, &rrdatabuf,
mctx, &dstkey));
- CHECK(dns_keytable_add(keytable, &dstkey));
- INSIST(dstkey == NULL);
+ *target = dstkey;
return (ISC_R_SUCCESS);
cleanup:
if (result == DST_R_NOCRYPTO) {
cfg_obj_log(key, ns_g_lctx, ISC_LOG_ERROR,
- "ignoring trusted key for '%s': no crypto support",
+ "ignoring %s key for '%s': no crypto support",
+ managed ? "managed" : "trusted",
keynamestr);
- result = ISC_R_SUCCESS;
+ } else if (result == DST_R_UNSUPPORTEDALG) {
+ cfg_obj_log(key, ns_g_lctx, ISC_LOG_WARNING,
+ "skipping %s key for '%s': %s",
+ managed ? "managed" : "trusted",
+ keynamestr, isc_result_totext(result));
} else {
cfg_obj_log(key, ns_g_lctx, ISC_LOG_ERROR,
- "configuring trusted key for '%s': %s",
+ "configuring %s key for '%s': %s",
+ managed ? "managed" : "trusted",
keynamestr, isc_result_totext(result));
result = ISC_R_FAILURE;
}
@@ -447,63 +600,215 @@ configure_view_dnsseckey(const cfg_obj_t *vconfig, const cfg_obj_t *key,
return (result);
}
+static isc_result_t
+load_view_keys(const cfg_obj_t *keys, const cfg_obj_t *vconfig,
+ dns_view_t *view, isc_boolean_t managed,
+ dns_name_t *keyname, isc_mem_t *mctx)
+{
+ const cfg_listelt_t *elt, *elt2;
+ const cfg_obj_t *key, *keylist;
+ dst_key_t *dstkey = NULL;
+ isc_result_t result;
+ dns_keytable_t *secroots = NULL;
+
+ CHECK(dns_view_getsecroots(view, &secroots));
+
+ for (elt = cfg_list_first(keys);
+ elt != NULL;
+ elt = cfg_list_next(elt)) {
+ keylist = cfg_listelt_value(elt);
+
+ for (elt2 = cfg_list_first(keylist);
+ elt2 != NULL;
+ elt2 = cfg_list_next(elt2)) {
+ key = cfg_listelt_value(elt2);
+ result = dstkey_fromconfig(vconfig, key, managed,
+ &dstkey, mctx);
+ if (result == DST_R_UNSUPPORTEDALG) {
+ result = ISC_R_SUCCESS;
+ continue;
+ }
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+
+ /*
+ * If keyname was specified, we only add that key.
+ */
+ if (keyname != NULL &&
+ !dns_name_equal(keyname, dst_key_name(dstkey)))
+ {
+ dst_key_free(&dstkey);
+ continue;
+ }
+
+ CHECK(dns_keytable_add(secroots, managed, &dstkey));
+ }
+ }
+
+ cleanup:
+ if (dstkey != NULL)
+ dst_key_free(&dstkey);
+ if (secroots != NULL)
+ dns_keytable_detach(&secroots);
+ if (result == DST_R_NOCRYPTO)
+ result = ISC_R_SUCCESS;
+ return (result);
+}
+
/*%
- * Configure DNSSEC keys for a view. Currently used only for
- * the security roots.
+ * Configure DNSSEC keys for a view.
*
* The per-view configuration values and the server-global defaults are read
- * from 'vconfig' and 'config'. The variable to be configured is '*target'.
+ * from 'vconfig' and 'config'.
*/
static isc_result_t
-configure_view_dnsseckeys(const cfg_obj_t *vconfig, const cfg_obj_t *config,
- isc_mem_t *mctx, dns_keytable_t **target)
+configure_view_dnsseckeys(dns_view_t *view, const cfg_obj_t *vconfig,
+ const cfg_obj_t *config, const cfg_obj_t *bindkeys,
+ isc_boolean_t auto_dlv, isc_boolean_t auto_root,
+ isc_mem_t *mctx)
{
- isc_result_t result;
- const cfg_obj_t *keys = NULL;
+ isc_result_t result = ISC_R_SUCCESS;
+ const cfg_obj_t *view_keys = NULL;
+ const cfg_obj_t *global_keys = NULL;
+ const cfg_obj_t *view_managed_keys = NULL;
+ const cfg_obj_t *global_managed_keys = NULL;
+ const cfg_obj_t *maps[4];
const cfg_obj_t *voptions = NULL;
- const cfg_listelt_t *element, *element2;
- const cfg_obj_t *keylist;
- const cfg_obj_t *key;
- dns_keytable_t *keytable = NULL;
+ const cfg_obj_t *options = NULL;
+ const cfg_obj_t *obj = NULL;
+ const char *directory;
+ int i = 0;
- CHECK(dns_keytable_create(mctx, &keytable));
+ /* We don't need trust anchors for the _bind view */
+ if (strcmp(view->name, "_bind") == 0 &&
+ view->rdclass == dns_rdataclass_chaos) {
+ return (ISC_R_SUCCESS);
+ }
- if (vconfig != NULL)
+ if (vconfig != NULL) {
voptions = cfg_tuple_get(vconfig, "options");
+ if (voptions != NULL) {
+ (void) cfg_map_get(voptions, "trusted-keys",
+ &view_keys);
+ (void) cfg_map_get(voptions, "managed-keys",
+ &view_managed_keys);
+ maps[i++] = voptions;
+ }
+ }
- keys = NULL;
- if (voptions != NULL)
- (void)cfg_map_get(voptions, "trusted-keys", &keys);
- if (keys == NULL)
- (void)cfg_map_get(config, "trusted-keys", &keys);
+ if (config != NULL) {
+ (void)cfg_map_get(config, "trusted-keys", &global_keys);
+ (void)cfg_map_get(config, "managed-keys", &global_managed_keys);
+ (void)cfg_map_get(config, "options", &options);
+ if (options != NULL) {
+ maps[i++] = options;
+ }
+ }
- for (element = cfg_list_first(keys);
- element != NULL;
- element = cfg_list_next(element))
- {
- keylist = cfg_listelt_value(element);
- for (element2 = cfg_list_first(keylist);
- element2 != NULL;
- element2 = cfg_list_next(element2))
- {
- key = cfg_listelt_value(element2);
- CHECK(configure_view_dnsseckey(vconfig, key,
- keytable, mctx));
+ maps[i++] = ns_g_defaults;
+ maps[i] = NULL;
+
+ result = dns_view_initsecroots(view, mctx);
+ if (result != ISC_R_SUCCESS) {
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "couldn't create keytable");
+ return (ISC_R_UNEXPECTED);
+ }
+
+ if (auto_dlv && view->rdclass == dns_rdataclass_in) {
+ const cfg_obj_t *builtin_keys = NULL;
+ const cfg_obj_t *builtin_managed_keys = NULL;
+
+ isc_log_write(ns_g_lctx, DNS_LOGCATEGORY_SECURITY,
+ NS_LOGMODULE_SERVER, ISC_LOG_WARNING,
+ "using built-in DLV key for view %s",
+ view->name);
+
+ /*
+ * If bind.keys exists, it overrides the managed-keys
+ * clause hard-coded in ns_g_config.
+ */
+ if (bindkeys != NULL) {
+ (void)cfg_map_get(bindkeys, "trusted-keys",
+ &builtin_keys);
+ (void)cfg_map_get(bindkeys, "managed-keys",
+ &builtin_managed_keys);
+ } else {
+ (void)cfg_map_get(ns_g_config, "trusted-keys",
+ &builtin_keys);
+ (void)cfg_map_get(ns_g_config, "managed-keys",
+ &builtin_managed_keys);
}
+
+ if (builtin_keys != NULL)
+ CHECK(load_view_keys(builtin_keys, vconfig, view,
+ ISC_FALSE, view->dlv, mctx));
+ if (builtin_managed_keys != NULL)
+ CHECK(load_view_keys(builtin_managed_keys, vconfig,
+ view, ISC_TRUE, view->dlv, mctx));
}
- dns_keytable_detach(target);
- *target = keytable; /* Transfer ownership. */
- keytable = NULL;
- result = ISC_R_SUCCESS;
+ if (auto_root && view->rdclass == dns_rdataclass_in) {
+ const cfg_obj_t *builtin_keys = NULL;
+ const cfg_obj_t *builtin_managed_keys = NULL;
- cleanup:
+ isc_log_write(ns_g_lctx, DNS_LOGCATEGORY_SECURITY,
+ NS_LOGMODULE_SERVER, ISC_LOG_WARNING,
+ "using built-in root key for view %s",
+ view->name);
+
+ /*
+ * If bind.keys exists, it overrides the managed-keys
+ * clause hard-coded in ns_g_config.
+ */
+ if (bindkeys != NULL) {
+ (void)cfg_map_get(bindkeys, "trusted-keys",
+ &builtin_keys);
+ (void)cfg_map_get(bindkeys, "managed-keys",
+ &builtin_managed_keys);
+ } else {
+ (void)cfg_map_get(ns_g_config, "trusted-keys",
+ &builtin_keys);
+ (void)cfg_map_get(ns_g_config, "managed-keys",
+ &builtin_managed_keys);
+ }
+
+ if (builtin_keys != NULL)
+ CHECK(load_view_keys(builtin_keys, vconfig, view,
+ ISC_FALSE, dns_rootname, mctx));
+ if (builtin_managed_keys != NULL)
+ CHECK(load_view_keys(builtin_managed_keys, vconfig,
+ view, ISC_TRUE, dns_rootname,
+ mctx));
+ }
+
+ CHECK(load_view_keys(view_keys, vconfig, view, ISC_FALSE,
+ NULL, mctx));
+ CHECK(load_view_keys(view_managed_keys, vconfig, view, ISC_TRUE,
+ NULL, mctx));
+
+ if (view->rdclass == dns_rdataclass_in) {
+ CHECK(load_view_keys(global_keys, vconfig, view, ISC_FALSE,
+ NULL, mctx));
+ CHECK(load_view_keys(global_managed_keys, vconfig, view,
+ ISC_TRUE, NULL, mctx));
+ }
+
+ /*
+ * Add key zone for managed-keys.
+ */
+ obj = NULL;
+ (void)ns_config_get(maps, "managed-keys-directory", &obj);
+ directory = obj != NULL ? cfg_obj_asstring(obj) : NULL;
+ CHECK(add_keydata_zone(view, directory, ns_g_mctx));
+
+ cleanup:
return (result);
}
static isc_result_t
-mustbesecure(const cfg_obj_t *mbs, dns_resolver_t *resolver)
-{
+mustbesecure(const cfg_obj_t *mbs, dns_resolver_t *resolver) {
const cfg_listelt_t *element;
const cfg_obj_t *obj;
const char *str;
@@ -523,8 +828,7 @@ mustbesecure(const cfg_obj_t *mbs, dns_resolver_t *resolver)
str = cfg_obj_asstring(cfg_tuple_get(obj, "name"));
isc_buffer_init(&b, str, strlen(str));
isc_buffer_add(&b, strlen(str));
- CHECK(dns_name_fromtext(name, &b, dns_rootname,
- ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &b, dns_rootname, 0, NULL));
value = cfg_obj_asboolean(cfg_tuple_get(obj, "value"));
CHECK(dns_resolver_setmustbesecure(resolver, name, value));
}
@@ -684,7 +988,7 @@ configure_order(dns_order_t *order, const cfg_obj_t *ent) {
isc_buffer_add(&b, strlen(str));
dns_fixedname_init(&fixed);
result = dns_name_fromtext(dns_fixedname_name(&fixed), &b,
- dns_rootname, ISC_FALSE, NULL);
+ dns_rootname, 0, NULL);
if (result != ISC_R_SUCCESS)
return (result);
@@ -868,7 +1172,7 @@ disable_algorithms(const cfg_obj_t *disabled, dns_resolver_t *resolver) {
str = cfg_obj_asstring(cfg_tuple_get(disabled, "name"));
isc_buffer_init(&b, str, strlen(str));
isc_buffer_add(&b, strlen(str));
- CHECK(dns_name_fromtext(name, &b, dns_rootname, ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &b, dns_rootname, 0, NULL));
algorithms = cfg_tuple_get(disabled, "algorithms");
for (element = cfg_list_first(algorithms);
@@ -921,7 +1225,7 @@ on_disable_list(const cfg_obj_t *disablelist, dns_name_t *zonename) {
isc_buffer_init(&b, str, strlen(str));
isc_buffer_add(&b, strlen(str));
result = dns_name_fromtext(name, &b, dns_rootname,
- ISC_TRUE, NULL);
+ 0, NULL);
RUNTIME_CHECK(result == ISC_R_SUCCESS);
if (dns_name_equal(name, zonename))
return (ISC_TRUE);
@@ -979,6 +1283,20 @@ setquerystats(dns_zone_t *zone, isc_mem_t *mctx, isc_boolean_t on) {
return (ISC_R_SUCCESS);
}
+static ns_cache_t *
+cachelist_find(ns_cachelist_t *cachelist, const char *cachename) {
+ ns_cache_t *nsc;
+
+ for (nsc = ISC_LIST_HEAD(*cachelist);
+ nsc != NULL;
+ nsc = ISC_LIST_NEXT(nsc, link)) {
+ if (strcmp(dns_cache_getname(nsc->cache), cachename) == 0)
+ return (nsc);
+ }
+
+ return (NULL);
+}
+
static isc_boolean_t
cache_reusable(dns_view_t *originview, dns_view_t *view,
isc_boolean_t new_zero_no_soattl)
@@ -996,6 +1314,238 @@ cache_reusable(dns_view_t *originview, dns_view_t *view,
return (ISC_TRUE);
}
+static isc_boolean_t
+cache_sharable(dns_view_t *originview, dns_view_t *view,
+ isc_boolean_t new_zero_no_soattl,
+ unsigned int new_cleaning_interval,
+ isc_uint32_t new_max_cache_size)
+{
+ /*
+ * If the cache cannot even reused for the same view, it cannot be
+ * shared with other views.
+ */
+ if (!cache_reusable(originview, view, new_zero_no_soattl))
+ return (ISC_FALSE);
+
+ /*
+ * Check other cache related parameters that must be consistent among
+ * the sharing views.
+ */
+ if (dns_cache_getcleaninginterval(originview->cache) !=
+ new_cleaning_interval ||
+ dns_cache_getcachesize(originview->cache) != new_max_cache_size) {
+ return (ISC_FALSE);
+ }
+
+ return (ISC_TRUE);
+}
+
+#ifdef DLZ
+/*
+ * Callback from DLZ configure when the driver sets up a writeable zone
+ */
+static isc_result_t
+dlzconfigure_callback(dns_view_t *view, dns_zone_t *zone) {
+ dns_name_t *origin = dns_zone_getorigin(zone);
+ dns_rdataclass_t zclass = view->rdclass;
+ isc_result_t result;
+
+ result = dns_zonemgr_managezone(ns_g_server->zonemgr, zone);
+ if (result != ISC_R_SUCCESS)
+ return result;
+ dns_zone_setstats(zone, ns_g_server->zonestats);
+
+ return ns_zone_configure_writeable_dlz(view->dlzdatabase,
+ zone, zclass, origin);
+}
+#endif
+
+static isc_result_t
+dns64_reverse(dns_view_t *view, isc_mem_t *mctx, isc_netaddr_t *na,
+ unsigned int prefixlen, const char *server,
+ const char *contact)
+{
+ char *cp;
+ char reverse[48+sizeof("ip6.arpa.")];
+ const char *dns64_dbtype[4] = { "_builtin", "dns64", ".", "." };
+ const char *sep = ": view ";
+ const char *viewname = view->name;
+ const unsigned char *s6;
+ dns_fixedname_t fixed;
+ dns_name_t *name;
+ dns_zone_t *zone = NULL;
+ int dns64_dbtypec = 4;
+ isc_buffer_t b;
+ isc_result_t result;
+
+ REQUIRE(prefixlen == 32 || prefixlen == 40 || prefixlen == 48 ||
+ prefixlen == 56 || prefixlen == 64 || prefixlen == 96);
+
+ if (!strcmp(viewname, "_default")) {
+ sep = "";
+ viewname = "";
+ }
+
+ /*
+ * Construct the reverse name of the zone.
+ */
+ cp = reverse;
+ s6 = na->type.in6.s6_addr;
+ while (prefixlen > 0) {
+ prefixlen -= 8;
+ sprintf(cp, "%x.%x.", s6[prefixlen/8] & 0xf,
+ (s6[prefixlen/8] >> 4) & 0xf);
+ cp += 4;
+ }
+ strcat(cp, "ip6.arpa.");
+
+ /*
+ * Create the actual zone.
+ */
+ if (server != NULL)
+ dns64_dbtype[2] = server;
+ if (contact != NULL)
+ dns64_dbtype[3] = contact;
+ dns_fixedname_init(&fixed);
+ name = dns_fixedname_name(&fixed);
+ isc_buffer_init(&b, reverse, strlen(reverse));
+ isc_buffer_add(&b, strlen(reverse));
+ CHECK(dns_name_fromtext(name, &b, dns_rootname, 0, NULL));
+ CHECK(dns_zone_create(&zone, mctx));
+ CHECK(dns_zone_setorigin(zone, name));
+ dns_zone_setview(zone, view);
+ CHECK(dns_zonemgr_managezone(ns_g_server->zonemgr, zone));
+ dns_zone_setclass(zone, view->rdclass);
+ dns_zone_settype(zone, dns_zone_master);
+ dns_zone_setstats(zone, ns_g_server->zonestats);
+ CHECK(dns_zone_setdbtype(zone, dns64_dbtypec, dns64_dbtype));
+ if (view->queryacl != NULL)
+ dns_zone_setqueryacl(zone, view->queryacl);
+ if (view->queryonacl != NULL)
+ dns_zone_setqueryonacl(zone, view->queryonacl);
+ dns_zone_setdialup(zone, dns_dialuptype_no);
+ dns_zone_setnotifytype(zone, dns_notifytype_no);
+ dns_zone_setoption(zone, DNS_ZONEOPT_NOCHECKNS, ISC_TRUE);
+ CHECK(setquerystats(zone, mctx, ISC_FALSE)); /* XXXMPA */
+ CHECK(dns_view_addzone(view, zone));
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL, NS_LOGMODULE_SERVER,
+ ISC_LOG_INFO, "dns64 reverse zone%s%s: %s", sep,
+ viewname, reverse);
+
+cleanup:
+ if (zone != NULL)
+ dns_zone_detach(&zone);
+ return (result);
+}
+
+static isc_result_t
+configure_rpz(dns_view_t *view, const cfg_listelt_t *element) {
+ const cfg_obj_t *rpz_obj, *policy_obj;
+ const char *str;
+ dns_fixedname_t fixed;
+ dns_name_t *origin;
+ dns_rpz_zone_t *old, *new;
+ dns_zone_t *zone = NULL;
+ isc_result_t result;
+ unsigned int l1, l2;
+
+ new = isc_mem_get(view->mctx, sizeof(*new));
+ if (new == NULL) {
+ result = ISC_R_NOMEMORY;
+ goto cleanup;
+ }
+
+ memset(new, 0, sizeof(*new));
+ dns_name_init(&new->nsdname, NULL);
+ dns_name_init(&new->origin, NULL);
+ dns_name_init(&new->cname, NULL);
+ ISC_LIST_INITANDAPPEND(view->rpz_zones, new, link);
+
+ rpz_obj = cfg_listelt_value(element);
+ policy_obj = cfg_tuple_get(rpz_obj, "policy");
+ if (cfg_obj_isvoid(policy_obj)) {
+ new->policy = DNS_RPZ_POLICY_GIVEN;
+ } else {
+ str = cfg_obj_asstring(policy_obj);
+ new->policy = dns_rpz_str2policy(str);
+ INSIST(new->policy != DNS_RPZ_POLICY_ERROR);
+ }
+
+ dns_fixedname_init(&fixed);
+ origin = dns_fixedname_name(&fixed);
+ str = cfg_obj_asstring(cfg_tuple_get(rpz_obj, "name"));
+ result = dns_name_fromstring(origin, str, DNS_NAME_DOWNCASE, NULL);
+ if (result != ISC_R_SUCCESS) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "invalid zone '%s'", str);
+ goto cleanup;
+ }
+
+ result = dns_name_fromstring2(&new->nsdname, DNS_RPZ_NSDNAME_ZONE,
+ origin, DNS_NAME_DOWNCASE, view->mctx);
+ if (result != ISC_R_SUCCESS) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "invalid zone '%s'", str);
+ goto cleanup;
+ }
+
+ /*
+ * The origin is part of 'nsdname' so we don't need to keep it
+ * seperately.
+ */
+ l1 = dns_name_countlabels(&new->nsdname);
+ l2 = dns_name_countlabels(origin);
+ dns_name_getlabelsequence(&new->nsdname, l1 - l2, l2, &new->origin);
+
+ /*
+ * Are we configured to with the reponse policy zone?
+ */
+ result = dns_view_findzone(view, &new->origin, &zone);
+ if (result != ISC_R_SUCCESS) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "unknown zone '%s'", str);
+ goto cleanup;
+ }
+
+ if (dns_zone_gettype(zone) != dns_zone_master &&
+ dns_zone_gettype(zone) != dns_zone_slave) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "zone '%s' is neither master nor slave", str);
+ dns_zone_detach(&zone);
+ result = DNS_R_NOTMASTER;
+ goto cleanup;
+ }
+ dns_zone_detach(&zone);
+
+ for (old = ISC_LIST_HEAD(view->rpz_zones);
+ old != new;
+ old = ISC_LIST_NEXT(old, link)) {
+ ++new->num;
+ if (dns_name_equal(&old->origin, &new->origin)) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "duplicate '%s'", str);
+ result = DNS_R_DUPLICATE;
+ goto cleanup;
+ }
+ }
+
+ if (new->policy == DNS_RPZ_POLICY_CNAME) {
+ str = cfg_obj_asstring(cfg_tuple_get(rpz_obj, "cname"));
+ result = dns_name_fromstring(&new->cname, str, 0, view->mctx);
+ if (result != ISC_R_SUCCESS) {
+ cfg_obj_log(rpz_obj, ns_g_lctx, DNS_RPZ_ERROR_LEVEL,
+ "invalid cname '%s'", str);
+ goto cleanup;
+ }
+ }
+
+ return (ISC_R_SUCCESS);
+
+ cleanup:
+ dns_rpz_view_destroy(view);
+ return (result);
+}
+
/*
* Configure 'view' according to 'vconfig', taking defaults from 'config'
* where values are missing in 'vconfig'.
@@ -1004,12 +1554,15 @@ cache_reusable(dns_view_t *originview, dns_view_t *view,
* global defaults in 'config' used exclusively.
*/
static isc_result_t
-configure_view(dns_view_t *view, const cfg_obj_t *config,
- const cfg_obj_t *vconfig, isc_mem_t *mctx,
- cfg_aclconfctx_t *actx, isc_boolean_t need_hints)
+configure_view(dns_view_t *view, cfg_parser_t* parser,
+ cfg_obj_t *config, cfg_obj_t *vconfig,
+ ns_cachelist_t *cachelist, const cfg_obj_t *bindkeys,
+ isc_mem_t *mctx, cfg_aclconfctx_t *actx,
+ isc_boolean_t need_hints)
{
const cfg_obj_t *maps[4];
const cfg_obj_t *cfgmaps[3];
+ const cfg_obj_t *optionmaps[3];
const cfg_obj_t *options = NULL;
const cfg_obj_t *voptions = NULL;
const cfg_obj_t *forwardtype;
@@ -1028,17 +1581,20 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
dns_cache_t *cache = NULL;
isc_result_t result;
isc_uint32_t max_adb_size;
+ unsigned int cleaning_interval;
isc_uint32_t max_cache_size;
isc_uint32_t max_acache_size;
isc_uint32_t lame_ttl;
- dns_tsig_keyring_t *ring;
+ dns_tsig_keyring_t *ring = NULL;
dns_view_t *pview = NULL; /* Production view */
isc_mem_t *cmctx;
dns_dispatch_t *dispatch4 = NULL;
dns_dispatch_t *dispatch6 = NULL;
isc_boolean_t reused_cache = ISC_FALSE;
- int i;
+ isc_boolean_t shared_cache = ISC_FALSE;
+ int i = 0, j = 0, k = 0;
const char *str;
+ const char *cachename = NULL;
dns_order_t *order = NULL;
isc_uint32_t udpsize;
unsigned int resopts = 0;
@@ -1052,7 +1608,14 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
const cfg_obj_t *disablelist = NULL;
isc_stats_t *resstats = NULL;
dns_stats_t *resquerystats = NULL;
+ isc_boolean_t auto_dlv = ISC_FALSE;
+ isc_boolean_t auto_root = ISC_FALSE;
+ ns_cache_t *nsc;
isc_boolean_t zero_no_soattl;
+ cfg_parser_t *newzones_parser = NULL;
+ cfg_obj_t *nzfconf = NULL;
+ dns_acl_t *clients = NULL, *mapped = NULL, *excluded = NULL;
+ unsigned int query_timeout;
REQUIRE(DNS_VIEW_VALID(view));
@@ -1061,22 +1624,28 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
if (config != NULL)
(void)cfg_map_get(config, "options", &options);
- i = 0;
+ /*
+ * maps: view options, options, defaults
+ * cfgmaps: view options, config
+ * optionmaps: view options, options
+ */
if (vconfig != NULL) {
voptions = cfg_tuple_get(vconfig, "options");
maps[i++] = voptions;
+ optionmaps[j++] = voptions;
+ cfgmaps[k++] = voptions;
}
- if (options != NULL)
+ if (options != NULL) {
maps[i++] = options;
+ optionmaps[j++] = options;
+ }
+
maps[i++] = ns_g_defaults;
maps[i] = NULL;
-
- i = 0;
- if (voptions != NULL)
- cfgmaps[i++] = voptions;
+ optionmaps[j] = NULL;
if (config != NULL)
- cfgmaps[i++] = config;
- cfgmaps[i] = NULL;
+ cfgmaps[k++] = config;
+ cfgmaps[k] = NULL;
if (!strcmp(viewname, "_default")) {
sep = "";
@@ -1137,12 +1706,12 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
dns_acache_setcachesize(view->acache, max_acache_size);
}
- CHECK(configure_view_acl(vconfig, config, "allow-query", actx,
+ CHECK(configure_view_acl(vconfig, config, "allow-query", NULL, actx,
ns_g_mctx, &view->queryacl));
-
if (view->queryacl == NULL) {
- CHECK(configure_view_acl(NULL, ns_g_config, "allow-query", actx,
- ns_g_mctx, &view->queryacl));
+ CHECK(configure_view_acl(NULL, ns_g_config, "allow-query",
+ NULL, actx, ns_g_mctx,
+ &view->queryacl));
}
/*
@@ -1159,7 +1728,62 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
{
const cfg_obj_t *zconfig = cfg_listelt_value(element);
CHECK(configure_zone(config, zconfig, vconfig, mctx, view,
- actx));
+ actx, ISC_FALSE));
+ }
+
+ /*
+ * Are we allowing zones to be added and deleted dynamically?
+ */
+ obj = NULL;
+ result = ns_config_get(maps, "allow-new-zones", &obj);
+ if (result == ISC_R_SUCCESS) {
+ isc_boolean_t allow = cfg_obj_asboolean(obj);
+ struct cfg_context *cfg = NULL;
+ if (allow) {
+ cfg = isc_mem_get(view->mctx, sizeof(*cfg));
+ if (cfg == NULL) {
+ result = ISC_R_NOMEMORY;
+ goto cleanup;
+ }
+ memset(cfg, 0, sizeof(*cfg));
+ isc_mem_attach(view->mctx, &cfg->mctx);
+ if (config != NULL)
+ cfg_obj_attach(config, &cfg->config);
+ cfg_parser_attach(parser, &cfg->parser);
+ cfg_aclconfctx_clone(actx, &cfg->actx);
+ }
+ dns_view_setnewzones(view, allow, cfg, cfgctx_destroy);
+ }
+
+ /*
+ * If we're allowing added zones, then load zone configuration
+ * from the newzone file for zones that were added during previous
+ * runs.
+ */
+ if (view->new_zone_file != NULL) {
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "loading additional zones for view '%s'",
+ view->name);
+
+ CHECK(cfg_parser_create(view->mctx, ns_g_lctx,
+ &newzones_parser));
+ result = cfg_parse_file(newzones_parser, view->new_zone_file,
+ &cfg_type_newzones, &nzfconf);
+ if (result == ISC_R_SUCCESS) {
+ zonelist = NULL;
+ cfg_map_get(nzfconf, "zone", &zonelist);
+ for (element = cfg_list_first(zonelist);
+ element != NULL;
+ element = cfg_list_next(element))
+ {
+ const cfg_obj_t *zconfig =
+ cfg_listelt_value(element);
+ CHECK(configure_zone(config, zconfig, vconfig,
+ mctx, view, actx,
+ ISC_TRUE));
+ }
+ }
}
#ifdef DLZ
@@ -1197,6 +1821,14 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
isc_mem_put(mctx, dlzargv, dlzargc * sizeof(*dlzargv));
if (result != ISC_R_SUCCESS)
goto cleanup;
+
+ /*
+ * If the dlz backend supports configuration,
+ * then call its configure method now.
+ */
+ result = dns_dlzconfigure(view, dlzconfigure_callback);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
}
}
#endif
@@ -1205,6 +1837,32 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
* Obtain configuration parameters that affect the decision of whether
* we can reuse/share an existing cache.
*/
+ obj = NULL;
+ result = ns_config_get(maps, "cleaning-interval", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ cleaning_interval = cfg_obj_asuint32(obj) * 60;
+
+ obj = NULL;
+ result = ns_config_get(maps, "max-cache-size", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ if (cfg_obj_isstring(obj)) {
+ str = cfg_obj_asstring(obj);
+ INSIST(strcasecmp(str, "unlimited") == 0);
+ max_cache_size = ISC_UINT32_MAX;
+ } else {
+ isc_resourcevalue_t value;
+ value = cfg_obj_asuint64(obj);
+ if (value > ISC_UINT32_MAX) {
+ cfg_obj_log(obj, ns_g_lctx, ISC_LOG_ERROR,
+ "'max-cache-size "
+ "%" ISC_PRINT_QUADFORMAT "d' is too large",
+ value);
+ result = ISC_R_RANGE;
+ goto cleanup;
+ }
+ max_cache_size = (isc_uint32_t)value;
+ }
+
/* Check-names. */
obj = NULL;
result = ns_checknames_get(maps, "response", &obj);
@@ -1229,6 +1887,109 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
zero_no_soattl = cfg_obj_asboolean(obj);
obj = NULL;
+ result = ns_config_get(maps, "dns64", &obj);
+ if (result == ISC_R_SUCCESS && strcmp(view->name, "_bind") &&
+ strcmp(view->name, "_meta")) {
+ const cfg_listelt_t *element;
+ isc_netaddr_t na, suffix, *sp;
+ unsigned int prefixlen;
+ const char *server, *contact;
+ const cfg_obj_t *myobj;
+
+ myobj = NULL;
+ result = ns_config_get(maps, "dns64-server", &myobj);
+ if (result == ISC_R_SUCCESS)
+ server = cfg_obj_asstring(myobj);
+ else
+ server = NULL;
+
+ myobj = NULL;
+ result = ns_config_get(maps, "dns64-contact", &myobj);
+ if (result == ISC_R_SUCCESS)
+ contact = cfg_obj_asstring(myobj);
+ else
+ contact = NULL;
+
+ for (element = cfg_list_first(obj);
+ element != NULL;
+ element = cfg_list_next(element))
+ {
+ const cfg_obj_t *map = cfg_listelt_value(element);
+ dns_dns64_t *dns64 = NULL;
+ unsigned int dns64options = 0;
+
+ cfg_obj_asnetprefix(cfg_map_getname(map), &na,
+ &prefixlen);
+
+ obj = NULL;
+ (void)cfg_map_get(map, "suffix", &obj);
+ if (obj != NULL) {
+ sp = &suffix;
+ isc_netaddr_fromsockaddr(sp,
+ cfg_obj_assockaddr(obj));
+ } else
+ sp = NULL;
+
+ clients = mapped = excluded = NULL;
+ obj = NULL;
+ (void)cfg_map_get(map, "clients", &obj);
+ if (obj != NULL) {
+ result = cfg_acl_fromconfig(obj, config,
+ ns_g_lctx, actx,
+ mctx, 0, &clients);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ }
+ obj = NULL;
+ (void)cfg_map_get(map, "mapped", &obj);
+ if (obj != NULL) {
+ result = cfg_acl_fromconfig(obj, config,
+ ns_g_lctx, actx,
+ mctx, 0, &mapped);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ }
+ obj = NULL;
+ (void)cfg_map_get(map, "exclude", &obj);
+ if (obj != NULL) {
+ result = cfg_acl_fromconfig(obj, config,
+ ns_g_lctx, actx,
+ mctx, 0, &excluded);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ }
+
+ obj = NULL;
+ (void)cfg_map_get(map, "recursive-only", &obj);
+ if (obj != NULL && cfg_obj_asboolean(obj))
+ dns64options |= DNS_DNS64_RECURSIVE_ONLY;
+
+ obj = NULL;
+ (void)cfg_map_get(map, "break-dnssec", &obj);
+ if (obj != NULL && cfg_obj_asboolean(obj))
+ dns64options |= DNS_DNS64_BREAK_DNSSEC;
+
+ result = dns_dns64_create(mctx, &na, prefixlen, sp,
+ clients, mapped, excluded,
+ dns64options, &dns64);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ dns_dns64_append(&view->dns64, dns64);
+ view->dns64cnt++;
+ result = dns64_reverse(view, mctx, &na, prefixlen,
+ server, contact);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ if (clients != NULL)
+ dns_acl_detach(&clients);
+ if (mapped != NULL)
+ dns_acl_detach(&mapped);
+ if (excluded != NULL)
+ dns_acl_detach(&excluded);
+ }
+ }
+
+ obj = NULL;
result = ns_config_get(maps, "dnssec-accept-expired", &obj);
INSIST(result == ISC_R_SUCCESS);
view->acceptexpired = cfg_obj_asboolean(obj);
@@ -1236,7 +1997,13 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
obj = NULL;
result = ns_config_get(maps, "dnssec-validation", &obj);
INSIST(result == ISC_R_SUCCESS);
- view->enablevalidation = cfg_obj_asboolean(obj);
+ if (cfg_obj_isboolean(obj)) {
+ view->enablevalidation = cfg_obj_asboolean(obj);
+ } else {
+ /* If dnssec-validation is not boolean, it must be "auto" */
+ view->enablevalidation = ISC_TRUE;
+ auto_root = ISC_TRUE;
+ }
obj = NULL;
result = ns_config_get(maps, "max-cache-ttl", &obj);
@@ -1251,53 +2018,113 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
view->maxncachettl = 7 * 24 * 3600;
/*
- * Configure the view's cache. Try to reuse an existing
- * cache if possible, otherwise create a new cache.
- * Note that the ADB is not preserved in either case.
- * When a matching view is found, the associated statistics are
- * also retrieved and reused.
+ * Configure the view's cache.
+ *
+ * First, check to see if there are any attach-cache options. If yes,
+ * attempt to lookup an existing cache at attach it to the view. If
+ * there is not one, then try to reuse an existing cache if possible;
+ * otherwise create a new cache.
*
- * XXX Determining when it is safe to reuse a cache is tricky.
+ * Note that the ADB is not preserved or shared in either case.
+ *
+ * When a matching view is found, the associated statistics are also
+ * retrieved and reused.
+ *
+ * XXX Determining when it is safe to reuse or share a cache is tricky.
* When the view's configuration changes, the cached data may become
* invalid because it reflects our old view of the world. We check
- * some of the configuration parameters that could invalidate the cache,
- * but there are other configuration options that should be checked.
- * For example, if a view uses a forwarder, changes in the forwarder
- * configuration may invalidate the cache. At the moment, it's the
- * administrator's responsibility to ensure these configuration options
- * don't invalidate reusing.
+ * some of the configuration parameters that could invalidate the cache
+ * or otherwise make it unsharable, but there are other configuration
+ * options that should be checked. For example, if a view uses a
+ * forwarder, changes in the forwarder configuration may invalidate
+ * the cache. At the moment, it's the administrator's responsibility to
+ * ensure these configuration options don't invalidate reusing/sharing.
*/
- result = dns_viewlist_find(&ns_g_server->viewlist,
- view->name, view->rdclass,
- &pview);
- if (result != ISC_R_NOTFOUND && result != ISC_R_SUCCESS)
- goto cleanup;
- if (pview != NULL) {
- if (cache_reusable(pview, view, zero_no_soattl)) {
- INSIST(pview->cache != NULL);
- isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
- NS_LOGMODULE_SERVER, ISC_LOG_DEBUG(3),
- "reusing existing cache");
- reused_cache = ISC_TRUE;
- dns_cache_attach(pview->cache, &cache);
- } else {
+ obj = NULL;
+ result = ns_config_get(maps, "attach-cache", &obj);
+ if (result == ISC_R_SUCCESS)
+ cachename = cfg_obj_asstring(obj);
+ else
+ cachename = view->name;
+ cache = NULL;
+ nsc = cachelist_find(cachelist, cachename);
+ if (nsc != NULL) {
+ if (!cache_sharable(nsc->primaryview, view, zero_no_soattl,
+ cleaning_interval, max_cache_size)) {
isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
- NS_LOGMODULE_SERVER, ISC_LOG_DEBUG(1),
- "cache cannot be reused for view %s "
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "views %s and %s can't share the cache "
"due to configuration parameter mismatch",
- view->name);
+ nsc->primaryview->name, view->name);
+ result = ISC_R_FAILURE;
+ goto cleanup;
}
- dns_view_getresstats(pview, &resstats);
- dns_view_getresquerystats(pview, &resquerystats);
- dns_view_detach(&pview);
- }
- if (cache == NULL) {
- CHECK(isc_mem_create(0, 0, &cmctx));
- CHECK(dns_cache_create(cmctx, ns_g_taskmgr, ns_g_timermgr,
- view->rdclass, "rbt", 0, NULL, &cache));
- isc_mem_setname(cmctx, "cache", NULL);
+ dns_cache_attach(nsc->cache, &cache);
+ shared_cache = ISC_TRUE;
+ } else {
+ if (strcmp(cachename, view->name) == 0) {
+ result = dns_viewlist_find(&ns_g_server->viewlist,
+ cachename, view->rdclass,
+ &pview);
+ if (result != ISC_R_NOTFOUND && result != ISC_R_SUCCESS)
+ goto cleanup;
+ if (pview != NULL) {
+ if (!cache_reusable(pview, view,
+ zero_no_soattl)) {
+ isc_log_write(ns_g_lctx,
+ NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER,
+ ISC_LOG_DEBUG(1),
+ "cache cannot be reused "
+ "for view %s due to "
+ "configuration parameter "
+ "mismatch", view->name);
+ } else {
+ INSIST(pview->cache != NULL);
+ isc_log_write(ns_g_lctx,
+ NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER,
+ ISC_LOG_DEBUG(3),
+ "reusing existing cache");
+ reused_cache = ISC_TRUE;
+ dns_cache_attach(pview->cache, &cache);
+ }
+ dns_view_getresstats(pview, &resstats);
+ dns_view_getresquerystats(pview,
+ &resquerystats);
+ dns_view_detach(&pview);
+ }
+ }
+ if (cache == NULL) {
+ /*
+ * Create a cache with the desired name. This normally
+ * equals the view name, but may also be a forward
+ * reference to a view that share the cache with this
+ * view but is not yet configured. If it is not the
+ * view name but not a forward reference either, then it
+ * is simply a named cache that is not shared.
+ */
+ CHECK(isc_mem_create(0, 0, &cmctx));
+ isc_mem_setname(cmctx, "cache", NULL);
+ CHECK(dns_cache_create2(cmctx, ns_g_taskmgr,
+ ns_g_timermgr, view->rdclass,
+ cachename, "rbt", 0, NULL,
+ &cache));
+ }
+ nsc = isc_mem_get(mctx, sizeof(*nsc));
+ if (nsc == NULL) {
+ result = ISC_R_NOMEMORY;
+ goto cleanup;
+ }
+ nsc->cache = NULL;
+ dns_cache_attach(cache, &nsc->cache);
+ nsc->primaryview = view;
+ nsc->needflush = ISC_FALSE;
+ nsc->adbsizeadjusted = ISC_FALSE;
+ ISC_LINK_INIT(nsc, link);
+ ISC_LIST_APPEND(*cachelist, nsc, link);
}
- dns_view_setcache(view, cache);
+ dns_view_setcache2(view, cache, shared_cache);
/*
* cache-file cannot be inherited if views are present, but this
@@ -1307,35 +2134,11 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
result = ns_config_get(maps, "cache-file", &obj);
if (result == ISC_R_SUCCESS && strcmp(view->name, "_bind") != 0) {
CHECK(dns_cache_setfilename(cache, cfg_obj_asstring(obj)));
- if (!reused_cache)
+ if (!reused_cache && !shared_cache)
CHECK(dns_cache_load(cache));
}
- obj = NULL;
- result = ns_config_get(maps, "cleaning-interval", &obj);
- INSIST(result == ISC_R_SUCCESS);
- dns_cache_setcleaninginterval(cache, cfg_obj_asuint32(obj) * 60);
-
- obj = NULL;
- result = ns_config_get(maps, "max-cache-size", &obj);
- INSIST(result == ISC_R_SUCCESS);
- if (cfg_obj_isstring(obj)) {
- str = cfg_obj_asstring(obj);
- INSIST(strcasecmp(str, "unlimited") == 0);
- max_cache_size = ISC_UINT32_MAX;
- } else {
- isc_resourcevalue_t value;
- value = cfg_obj_asuint64(obj);
- if (value > ISC_UINT32_MAX) {
- cfg_obj_log(obj, ns_g_lctx, ISC_LOG_ERROR,
- "'max-cache-size "
- "%" ISC_PRINT_QUADFORMAT "d' is too large",
- value);
- result = ISC_R_RANGE;
- goto cleanup;
- }
- max_cache_size = (isc_uint32_t)value;
- }
+ dns_cache_setcleaninginterval(cache, cleaning_interval);
dns_cache_setcachesize(cache, max_cache_size);
dns_cache_detach(&cache);
@@ -1373,13 +2176,23 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
dns_view_setresquerystats(view, resquerystats);
/*
- * Set the ADB cache size to 1/8th of the max-cache-size.
+ * Set the ADB cache size to 1/8th of the max-cache-size or
+ * MAX_ADB_SIZE_FOR_CACHESHARE when the cache is shared.
*/
max_adb_size = 0;
if (max_cache_size != 0) {
max_adb_size = max_cache_size / 8;
if (max_adb_size == 0)
max_adb_size = 1; /* Force minimum. */
+ if (view != nsc->primaryview &&
+ max_adb_size > MAX_ADB_SIZE_FOR_CACHESHARE) {
+ max_adb_size = MAX_ADB_SIZE_FOR_CACHESHARE;
+ if (!nsc->adbsizeadjusted) {
+ dns_adb_setadbsize(nsc->primaryview->adb,
+ MAX_ADB_SIZE_FOR_CACHESHARE);
+ nsc->adbsizeadjusted = ISC_TRUE;
+ }
+ }
}
dns_adb_setadbsize(view->adb, max_adb_size);
@@ -1395,6 +2208,18 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
dns_resolver_setlamettl(view->resolver, lame_ttl);
/*
+ * Set the resolver's query timeout.
+ */
+ obj = NULL;
+ result = ns_config_get(maps, "resolver-query-timeout", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ query_timeout = cfg_obj_asuint32(obj);
+ dns_resolver_settimeout(view->resolver, query_timeout);
+
+ /* Specify whether to use 0-TTL for negative response for SOA query */
+ dns_resolver_setzeronosoattl(view->resolver, zero_no_soattl);
+
+ /*
* Set the resolver's EDNS UDP size.
*/
obj = NULL;
@@ -1484,9 +2309,29 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
/*
* Configure the view's TSIG keys.
*/
- ring = NULL;
CHECK(ns_tsigkeyring_fromconfig(config, vconfig, view->mctx, &ring));
+ if (ns_g_server->sessionkey != NULL) {
+ CHECK(dns_tsigkeyring_add(ring, ns_g_server->session_keyname,
+ ns_g_server->sessionkey));
+ }
dns_view_setkeyring(view, ring);
+ dns_tsigkeyring_detach(&ring);
+
+ /*
+ * See if we can re-use a dynamic key ring.
+ */
+ result = dns_viewlist_find(&ns_g_server->viewlist, view->name,
+ view->rdclass, &pview);
+ if (result != ISC_R_NOTFOUND && result != ISC_R_SUCCESS)
+ goto cleanup;
+ if (pview != NULL) {
+ dns_view_getdynamickeyring(pview, &ring);
+ if (ring != NULL)
+ dns_view_setdynamickeyring(view, ring);
+ dns_tsigkeyring_detach(&ring);
+ dns_view_detach(&pview);
+ } else
+ dns_view_restorekeyring(view);
/*
* Configure the view's peer list.
@@ -1543,10 +2388,10 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
/*
* Configure the "match-clients" and "match-destinations" ACL.
*/
- CHECK(configure_view_acl(vconfig, config, "match-clients", actx,
+ CHECK(configure_view_acl(vconfig, config, "match-clients", NULL, actx,
ns_g_mctx, &view->matchclients));
- CHECK(configure_view_acl(vconfig, config, "match-destinations", actx,
- ns_g_mctx, &view->matchdestinations));
+ CHECK(configure_view_acl(vconfig, config, "match-destinations", NULL,
+ actx, ns_g_mctx, &view->matchdestinations));
/*
* Configure the "match-recursive-only" option.
@@ -1618,20 +2463,20 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
* "allow-recursion", and "allow-recursion-on" acls if
* configured in named.conf.
*/
- CHECK(configure_view_acl(vconfig, config, "allow-query-cache",
+ CHECK(configure_view_acl(vconfig, config, "allow-query-cache", NULL,
actx, ns_g_mctx, &view->cacheacl));
- CHECK(configure_view_acl(vconfig, config, "allow-query-cache-on",
+ CHECK(configure_view_acl(vconfig, config, "allow-query-cache-on", NULL,
actx, ns_g_mctx, &view->cacheonacl));
if (view->cacheonacl == NULL)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-query-cache-on", actx,
+ "allow-query-cache-on", NULL, actx,
ns_g_mctx, &view->cacheonacl));
if (strcmp(view->name, "_bind") != 0) {
CHECK(configure_view_acl(vconfig, config, "allow-recursion",
- actx, ns_g_mctx,
+ NULL, actx, ns_g_mctx,
&view->recursionacl));
CHECK(configure_view_acl(vconfig, config, "allow-recursion-on",
- actx, ns_g_mctx,
+ NULL, actx, ns_g_mctx,
&view->recursiononacl));
}
@@ -1643,8 +2488,14 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
*/
if (view->cacheacl == NULL && view->recursionacl != NULL)
dns_acl_attach(view->recursionacl, &view->cacheacl);
+ /*
+ * XXXEACH: This call to configure_view_acl() is redundant. We
+ * are leaving it as it is because we are making a minimal change
+ * for a patch release. In the future this should be changed to
+ * dns_acl_attach(view->queryacl, &view->cacheacl).
+ */
if (view->cacheacl == NULL && view->recursion)
- CHECK(configure_view_acl(vconfig, config, "allow-query",
+ CHECK(configure_view_acl(vconfig, config, "allow-query", NULL,
actx, ns_g_mctx, &view->cacheacl));
if (view->recursion &&
view->recursionacl == NULL && view->cacheacl != NULL)
@@ -1656,24 +2507,44 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
*/
if (view->recursionacl == NULL && view->recursion)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-recursion",
+ "allow-recursion", NULL,
actx, ns_g_mctx,
&view->recursionacl));
if (view->recursiononacl == NULL && view->recursion)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-recursion-on",
+ "allow-recursion-on", NULL,
actx, ns_g_mctx,
&view->recursiononacl));
if (view->cacheacl == NULL) {
if (view->recursion)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-query-cache", actx,
- ns_g_mctx, &view->cacheacl));
+ "allow-query-cache", NULL,
+ actx, ns_g_mctx,
+ &view->cacheacl));
else
- CHECK(dns_acl_none(ns_g_mctx, &view->cacheacl));
+ CHECK(dns_acl_none(mctx, &view->cacheacl));
}
/*
+ * Filter setting on addresses in the answer section.
+ */
+ CHECK(configure_view_acl(vconfig, config, "deny-answer-addresses",
+ "acl", actx, ns_g_mctx, &view->denyansweracl));
+ CHECK(configure_view_nametable(vconfig, config, "deny-answer-addresses",
+ "except-from", ns_g_mctx,
+ &view->answeracl_exclude));
+
+ /*
+ * Filter setting on names (CNAME/DNAME targets) in the answer section.
+ */
+ CHECK(configure_view_nametable(vconfig, config, "deny-answer-aliases",
+ "name", ns_g_mctx,
+ &view->denyanswernames));
+ CHECK(configure_view_nametable(vconfig, config, "deny-answer-aliases",
+ "except-from", ns_g_mctx,
+ &view->answernames_exclude));
+
+ /*
* Configure sortlist, if set
*/
CHECK(configure_view_sortlist(vconfig, config, actx, ns_g_mctx,
@@ -1686,19 +2557,19 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
*/
if (view->notifyacl == NULL)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-notify", actx,
+ "allow-notify", NULL, actx,
ns_g_mctx, &view->notifyacl));
if (view->transferacl == NULL)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-transfer", actx,
+ "allow-transfer", NULL, actx,
ns_g_mctx, &view->transferacl));
if (view->updateacl == NULL)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-update", actx,
+ "allow-update", NULL, actx,
ns_g_mctx, &view->updateacl));
if (view->upfwdacl == NULL)
CHECK(configure_view_acl(NULL, ns_g_config,
- "allow-update-forwarding", actx,
+ "allow-update-forwarding", NULL, actx,
ns_g_mctx, &view->upfwdacl));
obj = NULL;
@@ -1728,13 +2599,47 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
cfg_obj_asuint32(obj),
max_clients_per_query);
+#ifdef ALLOW_FILTER_AAAA_ON_V4
+ obj = NULL;
+ result = ns_config_get(maps, "filter-aaaa-on-v4", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ if (cfg_obj_isboolean(obj)) {
+ if (cfg_obj_asboolean(obj))
+ view->v4_aaaa = dns_v4_aaaa_filter;
+ else
+ view->v4_aaaa = dns_v4_aaaa_ok;
+ } else {
+ const char *v4_aaaastr = cfg_obj_asstring(obj);
+ if (strcasecmp(v4_aaaastr, "break-dnssec") == 0)
+ view->v4_aaaa = dns_v4_aaaa_break_dnssec;
+ else
+ INSIST(0);
+ }
+ CHECK(configure_view_acl(vconfig, config, "filter-aaaa", NULL,
+ actx, ns_g_mctx, &view->v4_aaaa_acl));
+#endif
+
obj = NULL;
result = ns_config_get(maps, "dnssec-enable", &obj);
INSIST(result == ISC_R_SUCCESS);
view->enablednssec = cfg_obj_asboolean(obj);
obj = NULL;
- result = ns_config_get(maps, "dnssec-lookaside", &obj);
+ result = ns_config_get(optionmaps, "dnssec-lookaside", &obj);
+ if (result == ISC_R_SUCCESS) {
+ /* If set to "auto", use the version from the defaults */
+ const cfg_obj_t *dlvobj;
+ dlvobj = cfg_listelt_value(cfg_list_first(obj));
+ if (!strcmp(cfg_obj_asstring(cfg_tuple_get(dlvobj, "domain")),
+ "auto") &&
+ cfg_obj_isvoid(cfg_tuple_get(dlvobj, "trust-anchor"))) {
+ auto_dlv = ISC_TRUE;
+ obj = NULL;
+ result = cfg_map_get(ns_g_defaults,
+ "dnssec-lookaside", &obj);
+ }
+ }
+
if (result == ISC_R_SUCCESS) {
for (element = cfg_list_first(obj);
element != NULL;
@@ -1745,31 +2650,13 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
dns_name_t *dlv;
obj = cfg_listelt_value(element);
-#if 0
- dns_fixedname_t fixed;
- dns_name_t *name;
-
- /*
- * When we support multiple dnssec-lookaside
- * entries this is how to find the domain to be
- * checked. XXXMPA
- */
- dns_fixedname_init(&fixed);
- name = dns_fixedname_name(&fixed);
- str = cfg_obj_asstring(cfg_tuple_get(obj,
- "domain"));
- isc_buffer_init(&b, str, strlen(str));
- isc_buffer_add(&b, strlen(str));
- CHECK(dns_name_fromtext(name, &b, dns_rootname,
- ISC_TRUE, NULL));
-#endif
str = cfg_obj_asstring(cfg_tuple_get(obj,
"trust-anchor"));
isc_buffer_init(&b, str, strlen(str));
isc_buffer_add(&b, strlen(str));
dlv = dns_fixedname_name(&view->dlv_fixed);
CHECK(dns_name_fromtext(dlv, &b, dns_rootname,
- ISC_TRUE, NULL));
+ DNS_NAME_DOWNCASE, NULL));
view->dlv = dns_fixedname_name(&view->dlv_fixed);
}
} else
@@ -1779,8 +2666,8 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
* For now, there is only one kind of trusted keys, the
* "security roots".
*/
- CHECK(configure_view_dnsseckeys(vconfig, config, mctx,
- &view->secroots));
+ CHECK(configure_view_dnsseckeys(view, vconfig, config, bindkeys,
+ auto_dlv, auto_root, mctx));
dns_resolver_resetmustbesecure(view->resolver);
obj = NULL;
result = ns_config_get(maps, "dnssec-must-be-secure", &obj);
@@ -1821,7 +2708,7 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
isc_buffer_init(&b, str, strlen(str));
isc_buffer_add(&b, strlen(str));
CHECK(dns_name_fromtext(name, &b, dns_rootname,
- ISC_FALSE, NULL));
+ 0, NULL));
CHECK(dns_view_excludedelegationonly(view,
name));
}
@@ -1874,8 +2761,8 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
str = cfg_obj_asstring(obj);
isc_buffer_init(&buffer, str, strlen(str));
isc_buffer_add(&buffer, strlen(str));
- CHECK(dns_name_fromtext(name, &buffer, dns_rootname,
- ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &buffer, dns_rootname, 0,
+ NULL));
isc_buffer_init(&buffer, server, sizeof(server) - 1);
CHECK(dns_name_totext(name, ISC_FALSE, &buffer));
server[isc_buffer_usedlength(&buffer)] = 0;
@@ -1889,8 +2776,8 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
str = cfg_obj_asstring(obj);
isc_buffer_init(&buffer, str, strlen(str));
isc_buffer_add(&buffer, strlen(str));
- CHECK(dns_name_fromtext(name, &buffer, dns_rootname,
- ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &buffer, dns_rootname, 0,
+ NULL));
isc_buffer_init(&buffer, contact, sizeof(contact) - 1);
CHECK(dns_name_totext(name, ISC_FALSE, &buffer));
contact[isc_buffer_usedlength(&buffer)] = 0;
@@ -1916,8 +2803,8 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
/*
* Look for zone on drop list.
*/
- CHECK(dns_name_fromtext(name, &buffer, dns_rootname,
- ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &buffer, dns_rootname, 0,
+ NULL));
if (disablelist != NULL &&
on_disable_list(disablelist, name))
continue;
@@ -2012,9 +2899,40 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
}
}
+ /*
+ * Make the list of response policy zone names for views that
+ * are used for real lookups and so care about hints.
+ */
+ zonelist = NULL;
+ if (view->rdclass == dns_rdataclass_in && need_hints) {
+ obj = NULL;
+ result = ns_config_get(maps, "response-policy", &obj);
+ if (result == ISC_R_SUCCESS)
+ cfg_map_get(obj, "zone", &zonelist);
+ }
+ if (zonelist != NULL) {
+
+ for (element = cfg_list_first(zonelist);
+ element != NULL;
+ element = cfg_list_next(element)) {
+ result = configure_rpz(view, element);
+ if (result != ISC_R_SUCCESS)
+ goto cleanup;
+ dns_rpz_set_need(ISC_TRUE);
+ }
+ }
+
result = ISC_R_SUCCESS;
cleanup:
+ if (clients != NULL)
+ dns_acl_detach(&clients);
+ if (mapped != NULL)
+ dns_acl_detach(&mapped);
+ if (excluded != NULL)
+ dns_acl_detach(&excluded);
+ if (ring != NULL)
+ dns_tsigkeyring_detach(&ring);
if (zone != NULL)
dns_zone_detach(&zone);
if (dispatch4 != NULL)
@@ -2033,6 +2951,12 @@ configure_view(dns_view_t *view, const cfg_obj_t *config,
if (cache != NULL)
dns_cache_detach(&cache);
+ if (newzones_parser != NULL) {
+ if (nzfconf != NULL)
+ cfg_obj_destroy(newzones_parser, &nzfconf);
+ cfg_parser_destroy(&newzones_parser);
+ }
+
return (result);
}
@@ -2105,8 +3029,8 @@ configure_alternates(const cfg_obj_t *config, dns_view_t *view,
isc_buffer_add(&buffer, strlen(str));
dns_fixedname_init(&fixed);
name = dns_fixedname_name(&fixed);
- CHECK(dns_name_fromtext(name, &buffer, dns_rootname,
- ISC_FALSE, NULL));
+ CHECK(dns_name_fromtext(name, &buffer, dns_rootname, 0,
+ NULL));
portobj = cfg_tuple_get(alternate, "port");
if (cfg_obj_isuint32(portobj)) {
@@ -2286,7 +3210,7 @@ create_view(const cfg_obj_t *vconfig, dns_viewlist_t *viewlist,
static isc_result_t
configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
const cfg_obj_t *vconfig, isc_mem_t *mctx, dns_view_t *view,
- cfg_aclconfctx_t *aclconf)
+ cfg_aclconfctx_t *aclconf, isc_boolean_t added)
{
dns_view_t *pview = NULL; /* Production view */
dns_zone_t *zone = NULL; /* New or reused zone */
@@ -2319,7 +3243,7 @@ configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
isc_buffer_add(&buffer, strlen(zname));
dns_fixedname_init(&fixorigin);
CHECK(dns_name_fromtext(dns_fixedname_name(&fixorigin),
- &buffer, dns_rootname, ISC_FALSE, NULL));
+ &buffer, dns_rootname, 0, NULL));
origin = dns_fixedname_name(&fixorigin);
CHECK(ns_config_getclass(cfg_tuple_get(zconfig, "class"),
@@ -2349,7 +3273,7 @@ configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
ztypestr = cfg_obj_asstring(typeobj);
/*
- * "hints zones" aren't zones. If we've got one,
+ * "hints zones" aren't zones. If we've got one,
* configure it and return.
*/
if (strcasecmp(ztypestr, "hint") == 0) {
@@ -2500,6 +3424,11 @@ configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
}
/*
+ * Mark whether the zone was originally added at runtime or not
+ */
+ dns_zone_setadded(zone, added);
+
+ /*
* Configure the zone.
*/
CHECK(ns_zone_configure(config, vconfig, zconfig, aclconf, zone));
@@ -2519,6 +3448,95 @@ configure_zone(const cfg_obj_t *config, const cfg_obj_t *zconfig,
}
/*
+ * Configure built-in zone for storing managed-key data.
+ */
+
+#define KEYZONE "managed-keys.bind"
+#define MKEYS ".mkeys"
+
+static isc_result_t
+add_keydata_zone(dns_view_t *view, const char *directory, isc_mem_t *mctx) {
+ isc_result_t result;
+ dns_view_t *pview = NULL;
+ dns_zone_t *zone = NULL;
+ dns_acl_t *none = NULL;
+ char filename[PATH_MAX];
+ char buffer[ISC_SHA256_DIGESTSTRINGLENGTH + sizeof(MKEYS)];
+ int n;
+
+ REQUIRE(view != NULL);
+
+ /* See if we can re-use an existing keydata zone. */
+ result = dns_viewlist_find(&ns_g_server->viewlist,
+ view->name, view->rdclass,
+ &pview);
+ if (result != ISC_R_NOTFOUND &&
+ result != ISC_R_SUCCESS)
+ return (result);
+
+ if (pview != NULL && pview->managed_keys != NULL) {
+ dns_zone_attach(pview->managed_keys, &view->managed_keys);
+ dns_zone_setview(pview->managed_keys, view);
+ dns_view_detach(&pview);
+ return (ISC_R_SUCCESS);
+ }
+
+ /* No existing keydata zone was found; create one */
+ CHECK(dns_zone_create(&zone, mctx));
+ CHECK(dns_zone_setorigin(zone, dns_rootname));
+
+ isc_sha256_data((void *)view->name, strlen(view->name), buffer);
+ strcat(buffer, MKEYS);
+ n = snprintf(filename, sizeof(filename), "%s%s%s",
+ directory ? directory : "", directory ? "/" : "",
+ strcmp(view->name, "_default") == 0 ? KEYZONE : buffer);
+ if (n < 0 || (size_t)n >= sizeof(filename)) {
+ result = (n < 0) ? ISC_R_FAILURE : ISC_R_NOSPACE;
+ goto cleanup;
+ }
+ CHECK(dns_zone_setfile(zone, filename));
+
+ dns_zone_setview(zone, view);
+ dns_zone_settype(zone, dns_zone_key);
+ dns_zone_setclass(zone, view->rdclass);
+
+ CHECK(dns_zonemgr_managezone(ns_g_server->zonemgr, zone));
+
+ if (view->acache != NULL)
+ dns_zone_setacache(zone, view->acache);
+
+ CHECK(dns_acl_none(mctx, &none));
+ dns_zone_setqueryacl(zone, none);
+ dns_zone_setqueryonacl(zone, none);
+ dns_acl_detach(&none);
+
+ dns_zone_setdialup(zone, dns_dialuptype_no);
+ dns_zone_setnotifytype(zone, dns_notifytype_no);
+ dns_zone_setoption(zone, DNS_ZONEOPT_NOCHECKNS, ISC_TRUE);
+ dns_zone_setjournalsize(zone, 0);
+
+ dns_zone_setstats(zone, ns_g_server->zonestats);
+ CHECK(setquerystats(zone, mctx, ISC_FALSE));
+
+ if (view->managed_keys != NULL)
+ dns_zone_detach(&view->managed_keys);
+ dns_zone_attach(zone, &view->managed_keys);
+
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "set up managed keys zone for view %s, file '%s'",
+ view->name, filename);
+
+cleanup:
+ if (zone != NULL)
+ dns_zone_detach(&zone);
+ if (none != NULL)
+ dns_acl_detach(&none);
+
+ return (result);
+}
+
+/*
* Configure a single server quota.
*/
static void
@@ -2914,13 +3932,226 @@ removed(dns_zone_t *zone, void *uap) {
return (ISC_R_SUCCESS);
}
+static void
+cleanup_session_key(ns_server_t *server, isc_mem_t *mctx) {
+ if (server->session_keyfile != NULL) {
+ isc_file_remove(server->session_keyfile);
+ isc_mem_free(mctx, server->session_keyfile);
+ server->session_keyfile = NULL;
+ }
+
+ if (server->session_keyname != NULL) {
+ if (dns_name_dynamic(server->session_keyname))
+ dns_name_free(server->session_keyname, mctx);
+ isc_mem_put(mctx, server->session_keyname, sizeof(dns_name_t));
+ server->session_keyname = NULL;
+ }
+
+ if (server->sessionkey != NULL)
+ dns_tsigkey_detach(&server->sessionkey);
+
+ server->session_keyalg = DST_ALG_UNKNOWN;
+ server->session_keybits = 0;
+}
+
+static isc_result_t
+generate_session_key(const char *filename, const char *keynamestr,
+ dns_name_t *keyname, const char *algstr,
+ dns_name_t *algname, unsigned int algtype,
+ isc_uint16_t bits, isc_mem_t *mctx,
+ dns_tsigkey_t **tsigkeyp)
+{
+ isc_result_t result = ISC_R_SUCCESS;
+ dst_key_t *key = NULL;
+ isc_buffer_t key_txtbuffer;
+ isc_buffer_t key_rawbuffer;
+ char key_txtsecret[256];
+ char key_rawsecret[64];
+ isc_region_t key_rawregion;
+ isc_stdtime_t now;
+ dns_tsigkey_t *tsigkey = NULL;
+ FILE *fp = NULL;
+
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "generating session key for dynamic DNS");
+
+ /* generate key */
+ result = dst_key_generate(keyname, algtype, bits, 1, 0,
+ DNS_KEYPROTO_ANY, dns_rdataclass_in,
+ mctx, &key);
+ if (result != ISC_R_SUCCESS)
+ return (result);
+
+ /*
+ * Dump the key to the buffer for later use. Should be done before
+ * we transfer the ownership of key to tsigkey.
+ */
+ isc_buffer_init(&key_rawbuffer, &key_rawsecret, sizeof(key_rawsecret));
+ CHECK(dst_key_tobuffer(key, &key_rawbuffer));
+
+ isc_buffer_usedregion(&key_rawbuffer, &key_rawregion);
+ isc_buffer_init(&key_txtbuffer, &key_txtsecret, sizeof(key_txtsecret));
+ CHECK(isc_base64_totext(&key_rawregion, -1, "", &key_txtbuffer));
+
+ /* Store the key in tsigkey. */
+ isc_stdtime_get(&now);
+ CHECK(dns_tsigkey_createfromkey(dst_key_name(key), algname, key,
+ ISC_FALSE, NULL, now, now, mctx, NULL,
+ &tsigkey));
+
+ /* Dump the key to the key file. */
+ fp = ns_os_openfile(filename, S_IRUSR|S_IWUSR, ISC_TRUE);
+ if (fp == NULL) {
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "could not create %s", filename);
+ result = ISC_R_NOPERM;
+ goto cleanup;
+ }
+
+ fprintf(fp, "key \"%s\" {\n"
+ "\talgorithm %s;\n"
+ "\tsecret \"%.*s\";\n};\n", keynamestr, algstr,
+ (int) isc_buffer_usedlength(&key_txtbuffer),
+ (char*) isc_buffer_base(&key_txtbuffer));
+
+ RUNTIME_CHECK(isc_stdio_flush(fp) == ISC_R_SUCCESS);
+ RUNTIME_CHECK(isc_stdio_close(fp) == ISC_R_SUCCESS);
+
+ dst_key_free(&key);
+
+ *tsigkeyp = tsigkey;
+
+ return (ISC_R_SUCCESS);
+
+ cleanup:
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "failed to generate session key "
+ "for dynamic DNS: %s", isc_result_totext(result));
+ if (tsigkey != NULL)
+ dns_tsigkey_detach(&tsigkey);
+ if (key != NULL)
+ dst_key_free(&key);
+
+ return (result);
+}
+
+static isc_result_t
+configure_session_key(const cfg_obj_t **maps, ns_server_t *server,
+ isc_mem_t *mctx)
+{
+ const char *keyfile, *keynamestr, *algstr;
+ unsigned int algtype;
+ dns_fixedname_t fname;
+ dns_name_t *keyname, *algname;
+ isc_buffer_t buffer;
+ isc_uint16_t bits;
+ const cfg_obj_t *obj;
+ isc_boolean_t need_deleteold = ISC_FALSE;
+ isc_boolean_t need_createnew = ISC_FALSE;
+ isc_result_t result;
+
+ obj = NULL;
+ result = ns_config_get(maps, "session-keyfile", &obj);
+ if (result == ISC_R_SUCCESS) {
+ if (cfg_obj_isvoid(obj))
+ keyfile = NULL; /* disable it */
+ else
+ keyfile = cfg_obj_asstring(obj);
+ } else
+ keyfile = ns_g_defaultsessionkeyfile;
+
+ obj = NULL;
+ result = ns_config_get(maps, "session-keyname", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ keynamestr = cfg_obj_asstring(obj);
+ dns_fixedname_init(&fname);
+ isc_buffer_init(&buffer, keynamestr, strlen(keynamestr));
+ isc_buffer_add(&buffer, strlen(keynamestr));
+ keyname = dns_fixedname_name(&fname);
+ result = dns_name_fromtext(keyname, &buffer, dns_rootname, 0, NULL);
+ if (result != ISC_R_SUCCESS)
+ return (result);
+
+ obj = NULL;
+ result = ns_config_get(maps, "session-keyalg", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ algstr = cfg_obj_asstring(obj);
+ algname = NULL;
+ result = ns_config_getkeyalgorithm2(algstr, &algname, &algtype, &bits);
+ if (result != ISC_R_SUCCESS) {
+ const char *s = " (keeping current key)";
+
+ cfg_obj_log(obj, ns_g_lctx, ISC_LOG_ERROR, "session-keyalg: "
+ "unsupported or unknown algorithm '%s'%s",
+ algstr,
+ server->session_keyfile != NULL ? s : "");
+ return (result);
+ }
+
+ /* See if we need to (re)generate a new key. */
+ if (keyfile == NULL) {
+ if (server->session_keyfile != NULL)
+ need_deleteold = ISC_TRUE;
+ } else if (server->session_keyfile == NULL)
+ need_createnew = ISC_TRUE;
+ else if (strcmp(keyfile, server->session_keyfile) != 0 ||
+ !dns_name_equal(server->session_keyname, keyname) ||
+ server->session_keyalg != algtype ||
+ server->session_keybits != bits) {
+ need_deleteold = ISC_TRUE;
+ need_createnew = ISC_TRUE;
+ }
+
+ if (need_deleteold) {
+ INSIST(server->session_keyfile != NULL);
+ INSIST(server->session_keyname != NULL);
+ INSIST(server->sessionkey != NULL);
+
+ cleanup_session_key(server, mctx);
+ }
+
+ if (need_createnew) {
+ INSIST(server->sessionkey == NULL);
+ INSIST(server->session_keyfile == NULL);
+ INSIST(server->session_keyname == NULL);
+ INSIST(server->session_keyalg == DST_ALG_UNKNOWN);
+ INSIST(server->session_keybits == 0);
+
+ server->session_keyname = isc_mem_get(mctx, sizeof(dns_name_t));
+ if (server->session_keyname == NULL)
+ goto cleanup;
+ dns_name_init(server->session_keyname, NULL);
+ CHECK(dns_name_dup(keyname, mctx, server->session_keyname));
+
+ server->session_keyfile = isc_mem_strdup(mctx, keyfile);
+ if (server->session_keyfile == NULL)
+ goto cleanup;
+
+ server->session_keyalg = algtype;
+ server->session_keybits = bits;
+
+ CHECK(generate_session_key(keyfile, keynamestr, keyname, algstr,
+ algname, algtype, bits, mctx,
+ &server->sessionkey));
+ }
+
+ return (result);
+
+ cleanup:
+ cleanup_session_key(server, mctx);
+ return (result);
+}
+
static isc_result_t
load_configuration(const char *filename, ns_server_t *server,
isc_boolean_t first_time)
{
cfg_aclconfctx_t aclconfctx;
- cfg_obj_t *config;
- cfg_parser_t *parser = NULL;
+ cfg_obj_t *config = NULL, *bindkeys = NULL;
+ cfg_parser_t *conf_parser = NULL, *bindkeys_parser = NULL;
const cfg_listelt_t *element;
const cfg_obj_t *builtin_views;
const cfg_obj_t *maps[3];
@@ -2931,7 +4162,7 @@ load_configuration(const char *filename, ns_server_t *server,
dns_view_t *view = NULL;
dns_view_t *view_next;
dns_viewlist_t tmpviewlist;
- dns_viewlist_t viewlist;
+ dns_viewlist_t viewlist, builtin_viewlist;
in_port_t listen_port, udpport_low, udpport_high;
int i;
isc_interval_t interval;
@@ -2943,10 +4174,14 @@ load_configuration(const char *filename, ns_server_t *server,
isc_uint32_t interface_interval;
isc_uint32_t reserved;
isc_uint32_t udpsize;
+ ns_cachelist_t cachelist, tmpcachelist;
unsigned int maxsocks;
+ ns_cache_t *nsc;
cfg_aclconfctx_init(&aclconfctx);
ISC_LIST_INIT(viewlist);
+ ISC_LIST_INIT(builtin_viewlist);
+ ISC_LIST_INIT(cachelist);
/* Ensure exclusive access to configuration data. */
result = isc_task_beginexclusive(server->task);
@@ -2958,8 +4193,7 @@ load_configuration(const char *filename, ns_server_t *server,
if (first_time) {
CHECK(ns_config_parsedefaults(ns_g_parser, &ns_g_config));
RUNTIME_CHECK(cfg_map_get(ns_g_config, "options",
- &ns_g_defaults) ==
- ISC_R_SUCCESS);
+ &ns_g_defaults) == ISC_R_SUCCESS);
}
/*
@@ -2976,10 +4210,10 @@ load_configuration(const char *filename, ns_server_t *server,
NS_LOGCATEGORY_GENERAL, NS_LOGMODULE_SERVER,
ISC_LOG_INFO, "loading configuration from '%s'",
filename);
- CHECK(cfg_parser_create(ns_g_mctx, ns_g_lctx, &parser));
- cfg_parser_setcallback(parser, directory_callback, NULL);
- result = cfg_parse_file(parser, filename, &cfg_type_namedconf,
- &config);
+ CHECK(cfg_parser_create(ns_g_mctx, ns_g_lctx, &conf_parser));
+ cfg_parser_setcallback(conf_parser, directory_callback, NULL);
+ result = cfg_parse_file(conf_parser, filename,
+ &cfg_type_namedconf, &config);
}
/*
@@ -2994,10 +4228,10 @@ load_configuration(const char *filename, ns_server_t *server,
NS_LOGCATEGORY_GENERAL, NS_LOGMODULE_SERVER,
ISC_LOG_INFO, "loading configuration from '%s'",
lwresd_g_resolvconffile);
- if (parser != NULL)
- cfg_parser_destroy(&parser);
- CHECK(cfg_parser_create(ns_g_mctx, ns_g_lctx, &parser));
- result = ns_lwresd_parseeresolvconf(ns_g_mctx, parser,
+ if (conf_parser != NULL)
+ cfg_parser_destroy(&conf_parser);
+ CHECK(cfg_parser_create(ns_g_mctx, ns_g_lctx, &conf_parser));
+ result = ns_lwresd_parseeresolvconf(ns_g_mctx, conf_parser,
&config);
}
CHECK(result);
@@ -3019,13 +4253,38 @@ load_configuration(const char *filename, ns_server_t *server,
maps[i++] = NULL;
/*
+ * If bind.keys exists, load it. If "dnssec-lookaside auto"
+ * is turned on, the keys found there will be used as default
+ * trust anchors.
+ */
+ obj = NULL;
+ result = ns_config_get(maps, "bindkeys-file", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ CHECKM(setstring(server, &server->bindkeysfile,
+ cfg_obj_asstring(obj)), "strdup");
+
+ if (access(server->bindkeysfile, R_OK) == 0) {
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "reading built-in trusted "
+ "keys from file '%s'", server->bindkeysfile);
+
+ CHECK(cfg_parser_create(ns_g_mctx, ns_g_lctx,
+ &bindkeys_parser));
+
+ result = cfg_parse_file(bindkeys_parser, server->bindkeysfile,
+ &cfg_type_bindkeys, &bindkeys);
+ CHECK(result);
+ }
+
+ /*
* Set process limits, which (usually) needs to be done as root.
*/
set_limits(maps);
/*
* Check if max number of open sockets that the system allows is
- * sufficiently large. Failing this condition is not necessarily fatal,
+ * sufficiently large. Failing this condition is not necessarily fatal,
* but may cause subsequent runtime failures for a busy recursive
* server.
*/
@@ -3078,7 +4337,7 @@ load_configuration(const char *filename, ns_server_t *server,
else
isc_quota_soft(&server->recursionquota, 0);
- CHECK(configure_view_acl(NULL, config, "blackhole", &aclconfctx,
+ CHECK(configure_view_acl(NULL, config, "blackhole", NULL, &aclconfctx,
ns_g_mctx, &server->blackholeacl));
if (server->blackholeacl != NULL)
dns_dispatchmgr_setblackhole(ns_g_dispatchmgr,
@@ -3318,6 +4577,31 @@ load_configuration(const char *filename, ns_server_t *server,
&interval, ISC_FALSE));
/*
+ * Write the PID file.
+ */
+ obj = NULL;
+ if (ns_config_get(maps, "pid-file", &obj) == ISC_R_SUCCESS)
+ if (cfg_obj_isvoid(obj))
+ ns_os_writepidfile(NULL, first_time);
+ else
+ ns_os_writepidfile(cfg_obj_asstring(obj), first_time);
+ else if (ns_g_lwresdonly)
+ ns_os_writepidfile(lwresd_g_defaultpidfile, first_time);
+ else
+ ns_os_writepidfile(ns_g_defaultpidfile, first_time);
+
+ /*
+ * Configure the server-wide session key. This must be done before
+ * configure views because zone configuration may need to know
+ * session-keyname.
+ *
+ * Failure of session key generation isn't fatal at this time; if it
+ * turns out that a session key is really needed but doesn't exist,
+ * we'll treat it as a fatal error then.
+ */
+ (void)configure_session_key(maps, server, ns_g_mctx);
+
+ /*
* Configure and freeze all explicit views. Explicit
* views that have zones were already created at parsing
* time, but views with no zones must be created here.
@@ -3328,12 +4612,13 @@ load_configuration(const char *filename, ns_server_t *server,
element != NULL;
element = cfg_list_next(element))
{
- const cfg_obj_t *vconfig = cfg_listelt_value(element);
+ cfg_obj_t *vconfig = cfg_listelt_value(element);
view = NULL;
CHECK(create_view(vconfig, &viewlist, &view));
INSIST(view != NULL);
- CHECK(configure_view(view, config, vconfig,
+ CHECK(configure_view(view, conf_parser, config, vconfig,
+ &cachelist, bindkeys,
ns_g_mctx, &aclconfctx, ISC_TRUE));
dns_view_freeze(view);
dns_view_detach(&view);
@@ -3351,15 +4636,15 @@ load_configuration(const char *filename, ns_server_t *server,
* In either case, we need to configure and freeze it.
*/
CHECK(create_view(NULL, &viewlist, &view));
- CHECK(configure_view(view, config, NULL, ns_g_mctx,
- &aclconfctx, ISC_TRUE));
+ CHECK(configure_view(view, conf_parser, config, NULL,
+ &cachelist, bindkeys,
+ ns_g_mctx, &aclconfctx, ISC_TRUE));
dns_view_freeze(view);
dns_view_detach(&view);
}
/*
- * Create (or recreate) the built-in views. Currently
- * there is only one, the _bind view.
+ * Create (or recreate) the built-in views.
*/
builtin_views = NULL;
RUNTIME_CHECK(cfg_map_get(ns_g_config, "view",
@@ -3368,25 +4653,38 @@ load_configuration(const char *filename, ns_server_t *server,
element != NULL;
element = cfg_list_next(element))
{
- const cfg_obj_t *vconfig = cfg_listelt_value(element);
- CHECK(create_view(vconfig, &viewlist, &view));
- CHECK(configure_view(view, config, vconfig, ns_g_mctx,
- &aclconfctx, ISC_FALSE));
+ cfg_obj_t *vconfig = cfg_listelt_value(element);
+
+ CHECK(create_view(vconfig, &builtin_viewlist, &view));
+ CHECK(configure_view(view, conf_parser, config, vconfig,
+ &cachelist, bindkeys,
+ ns_g_mctx, &aclconfctx, ISC_FALSE));
dns_view_freeze(view);
dns_view_detach(&view);
view = NULL;
}
- /*
- * Swap our new view list with the production one.
- */
+ /* Now combine the two viewlists into one */
+ ISC_LIST_APPENDLIST(viewlist, builtin_viewlist, link);
+
+ /* Swap our new view list with the production one. */
tmpviewlist = server->viewlist;
server->viewlist = viewlist;
viewlist = tmpviewlist;
- /*
- * Load the TKEY information from the configuration.
- */
+ /* Make the view list available to each of the views */
+ view = ISC_LIST_HEAD(server->viewlist);
+ while (view != NULL) {
+ view->viewlist = &server->viewlist;
+ view = ISC_LIST_NEXT(view, link);
+ }
+
+ /* Swap our new cache list with the production one. */
+ tmpcachelist = server->cachelist;
+ server->cachelist = cachelist;
+ cachelist = tmpcachelist;
+
+ /* Load the TKEY information from the configuration. */
if (options != NULL) {
dns_tkeyctx_t *t = NULL;
CHECKM(ns_tkeyctx_fromconfig(options, ns_g_mctx, ns_g_entropy,
@@ -3551,16 +4849,6 @@ load_configuration(const char *filename, ns_server_t *server,
}
}
- obj = NULL;
- if (ns_config_get(maps, "pid-file", &obj) == ISC_R_SUCCESS)
- if (cfg_obj_isvoid(obj))
- ns_os_writepidfile(NULL, first_time);
- else
- ns_os_writepidfile(cfg_obj_asstring(obj), first_time);
- else if (ns_g_lwresdonly)
- ns_os_writepidfile(lwresd_g_defaultpidfile, first_time);
- else
- ns_os_writepidfile(ns_g_defaultpidfile, first_time);
obj = NULL;
if (options != NULL &&
@@ -3591,6 +4879,12 @@ load_configuration(const char *filename, ns_server_t *server,
"strdup");
obj = NULL;
+ result = ns_config_get(maps, "secroots-file", &obj);
+ INSIST(result == ISC_R_SUCCESS);
+ CHECKM(setstring(server, &server->secrootsfile, cfg_obj_asstring(obj)),
+ "strdup");
+
+ obj = NULL;
result = ns_config_get(maps, "recursing-file", &obj);
INSIST(result == ISC_R_SUCCESS);
CHECKM(setstring(server, &server->recfile, cfg_obj_asstring(obj)),
@@ -3647,12 +4941,18 @@ load_configuration(const char *filename, ns_server_t *server,
if (v6portset != NULL)
isc_portset_destroy(ns_g_mctx, &v6portset);
- cfg_aclconfctx_destroy(&aclconfctx);
+ cfg_aclconfctx_clear(&aclconfctx);
- if (parser != NULL) {
+ if (conf_parser != NULL) {
if (config != NULL)
- cfg_obj_destroy(parser, &config);
- cfg_parser_destroy(&parser);
+ cfg_obj_destroy(conf_parser, &config);
+ cfg_parser_destroy(&conf_parser);
+ }
+
+ if (bindkeys_parser != NULL) {
+ if (bindkeys != NULL)
+ cfg_obj_destroy(bindkeys_parser, &bindkeys);
+ cfg_parser_destroy(&bindkeys_parser);
}
if (view != NULL)
@@ -3675,6 +4975,13 @@ load_configuration(const char *filename, ns_server_t *server,
dns_view_detach(&view);
}
+ /* Same cleanup for cache list. */
+ while ((nsc = ISC_LIST_HEAD(cachelist)) != NULL) {
+ ISC_LIST_UNLINK(cachelist, nsc, link);
+ dns_cache_detach(&nsc->cache);
+ isc_mem_put(server->mctx, nsc, sizeof(*nsc));
+ }
+
/*
* Adjust the listening interfaces in accordance with the source
* addresses specified in views and zones.
@@ -3708,6 +5015,8 @@ load_zones(ns_server_t *server, isc_boolean_t stop) {
view = ISC_LIST_NEXT(view, link))
{
CHECK(dns_view_load(view, stop));
+ if (view->managed_keys != NULL)
+ CHECK(dns_zone_load(view->managed_keys));
}
/*
@@ -3737,11 +5046,14 @@ load_new_zones(ns_server_t *server, isc_boolean_t stop) {
view = ISC_LIST_NEXT(view, link))
{
CHECK(dns_view_loadnew(view, stop));
+
+ /* Load managed-keys data */
+ if (view->managed_keys != NULL)
+ CHECK(dns_zone_loadnew(view->managed_keys));
}
+
/*
- * Force zone maintenance. Do this after loading
- * so that we know when we need to force AXFR of
- * slave zones whose master files are missing.
+ * Resume zone XFRs.
*/
dns_zonemgr_resumexfrs(server->zonemgr);
cleanup:
@@ -3820,6 +5132,7 @@ shutdown_server(isc_task_t *task, isc_event_t *event) {
dns_view_t *view, *view_next;
ns_server_t *server = (ns_server_t *)event->ev_arg;
isc_boolean_t flush = server->flushonshutdown;
+ ns_cache_t *nsc;
UNUSED(task);
INSIST(task == server->task);
@@ -3834,6 +5147,7 @@ shutdown_server(isc_task_t *task, isc_event_t *event) {
ns_statschannels_shutdown(server);
ns_controls_shutdown(server->controls);
end_reserved_dispatches(server, ISC_TRUE);
+ cleanup_session_key(server, server->mctx);
cfg_obj_destroy(ns_g_parser, &ns_g_config);
cfg_parser_destroy(&ns_g_parser);
@@ -3849,6 +5163,12 @@ shutdown_server(isc_task_t *task, isc_event_t *event) {
dns_view_detach(&view);
}
+ while ((nsc = ISC_LIST_HEAD(server->cachelist)) != NULL) {
+ ISC_LIST_UNLINK(server->cachelist, nsc, link);
+ dns_cache_detach(&nsc->cache);
+ isc_mem_put(server->mctx, nsc, sizeof(*nsc));
+ }
+
isc_timer_detach(&server->interface_timer);
isc_timer_detach(&server->heartbeat_timer);
isc_timer_detach(&server->pps_timer);
@@ -3860,6 +5180,11 @@ shutdown_server(isc_task_t *task, isc_event_t *event) {
dns_zonemgr_shutdown(server->zonemgr);
+ if (ns_g_sessionkey != NULL) {
+ dns_tsigkey_detach(&ns_g_sessionkey);
+ dns_name_free(&ns_g_sessionkeyname, server->mctx);
+ }
+
if (server->blackholeacl != NULL)
dns_acl_detach(&server->blackholeacl);
@@ -3918,7 +5243,8 @@ ns_server_create(isc_mem_t *mctx, ns_server_t **serverp) {
ISC_R_NOMEMORY : ISC_R_SUCCESS,
"allocating reload event");
- CHECKFATAL(dst_lib_init(ns_g_mctx, ns_g_entropy, ISC_ENTROPY_GOODONLY),
+ CHECKFATAL(dst_lib_init2(ns_g_mctx, ns_g_entropy,
+ ns_g_engine, ISC_ENTROPY_GOODONLY),
"initializing DST");
server->tkeyctx = NULL;
@@ -3963,10 +5289,20 @@ ns_server_create(isc_mem_t *mctx, ns_server_t **serverp) {
"isc_stats_create");
isc_socketmgr_setstats(ns_g_socketmgr, server->sockstats);
+ server->bindkeysfile = isc_mem_strdup(server->mctx, "bind.keys");
+ CHECKFATAL(server->bindkeysfile == NULL ? ISC_R_NOMEMORY :
+ ISC_R_SUCCESS,
+ "isc_mem_strdup");
+
server->dumpfile = isc_mem_strdup(server->mctx, "named_dump.db");
CHECKFATAL(server->dumpfile == NULL ? ISC_R_NOMEMORY : ISC_R_SUCCESS,
"isc_mem_strdup");
+ server->secrootsfile = isc_mem_strdup(server->mctx, "named.secroots");
+ CHECKFATAL(server->secrootsfile == NULL ? ISC_R_NOMEMORY :
+ ISC_R_SUCCESS,
+ "isc_mem_strdup");
+
server->recfile = isc_mem_strdup(server->mctx, "named.recursing");
CHECKFATAL(server->recfile == NULL ? ISC_R_NOMEMORY : ISC_R_SUCCESS,
"isc_mem_strdup");
@@ -4008,6 +5344,14 @@ ns_server_create(isc_mem_t *mctx, ns_server_t **serverp) {
ISC_LIST_INIT(server->statschannels);
+ ISC_LIST_INIT(server->cachelist);
+
+ server->sessionkey = NULL;
+ server->session_keyfile = NULL;
+ server->session_keyname = NULL;
+ server->session_keyalg = DST_ALG_UNKNOWN;
+ server->session_keybits = 0;
+
server->magic = NS_SERVER_MAGIC;
*serverp = server;
}
@@ -4027,7 +5371,9 @@ ns_server_destroy(ns_server_t **serverp) {
isc_stats_detach(&server->sockstats);
isc_mem_free(server->mctx, server->statsfile);
+ isc_mem_free(server->mctx, server->bindkeysfile);
isc_mem_free(server->mctx, server->dumpfile);
+ isc_mem_free(server->mctx, server->secrootsfile);
isc_mem_free(server->mctx, server->recfile);
if (server->version != NULL)
@@ -4047,6 +5393,7 @@ ns_server_destroy(ns_server_t **serverp) {
isc_event_free(&server->reload_event);
INSIST(ISC_LIST_EMPTY(server->viewlist));
+ INSIST(ISC_LIST_EMPTY(server->cachelist));
dns_aclenv_destroy(&server->aclenv);
@@ -4278,7 +5625,9 @@ next_token(char **stringp, const char *delim) {
* set '*zonep' to NULL.
*/
static isc_result_t
-zone_from_args(ns_server_t *server, char *args, dns_zone_t **zonep) {
+zone_from_args(ns_server_t *server, char *args, dns_zone_t **zonep,
+ const char **zonename)
+{
char *input, *ptr;
const char *zonetxt;
char *classtxt;
@@ -4302,6 +5651,8 @@ zone_from_args(ns_server_t *server, char *args, dns_zone_t **zonep) {
zonetxt = next_token(&input, " \t");
if (zonetxt == NULL)
return (ISC_R_SUCCESS);
+ if (zonename)
+ *zonename = zonetxt;
/* Look for the optional class name. */
classtxt = next_token(&input, " \t");
@@ -4314,7 +5665,7 @@ zone_from_args(ns_server_t *server, char *args, dns_zone_t **zonep) {
isc_buffer_add(&buf, strlen(zonetxt));
dns_fixedname_init(&name);
result = dns_name_fromtext(dns_fixedname_name(&name),
- &buf, dns_rootname, ISC_FALSE, NULL);
+ &buf, dns_rootname, 0, NULL);
if (result != ISC_R_SUCCESS)
goto fail1;
@@ -4362,7 +5713,7 @@ ns_server_retransfercommand(ns_server_t *server, char *args) {
dns_zone_t *zone = NULL;
dns_zonetype_t type;
- result = zone_from_args(server, args, &zone);
+ result = zone_from_args(server, args, &zone, NULL);
if (result != ISC_R_SUCCESS)
return (result);
if (zone == NULL)
@@ -4386,7 +5737,7 @@ ns_server_reloadcommand(ns_server_t *server, char *args, isc_buffer_t *text) {
dns_zonetype_t type;
const char *msg = NULL;
- result = zone_from_args(server, args, &zone);
+ result = zone_from_args(server, args, &zone, NULL);
if (result != ISC_R_SUCCESS)
return (result);
if (zone == NULL) {
@@ -4446,7 +5797,7 @@ ns_server_notifycommand(ns_server_t *server, char *args, isc_buffer_t *text) {
dns_zone_t *zone = NULL;
const unsigned char msg[] = "zone notify queued";
- result = zone_from_args(server, args, &zone);
+ result = zone_from_args(server, args, &zone, NULL);
if (result != ISC_R_SUCCESS)
return (result);
if (zone == NULL)
@@ -4471,7 +5822,7 @@ ns_server_refreshcommand(ns_server_t *server, char *args, isc_buffer_t *text) {
const unsigned char msg2[] = "not a slave or stub zone";
dns_zonetype_t type;
- result = zone_from_args(server, args, &zone);
+ result = zone_from_args(server, args, &zone, NULL);
if (result != ISC_R_SUCCESS)
return (result);
if (zone == NULL)
@@ -4710,15 +6061,23 @@ dumpdone(void *arg, isc_result_t result) {
nextview:
fprintf(dctx->fp, ";\n; Start view %s\n;\n", dctx->view->view->name);
resume:
- if (dctx->zone == NULL && dctx->cache == NULL && dctx->dumpcache) {
+ if (dctx->dumpcache && dns_view_iscacheshared(dctx->view->view)) {
+ fprintf(dctx->fp,
+ ";\n; Cache of view '%s' is shared as '%s'\n",
+ dctx->view->view->name,
+ dns_cache_getname(dctx->view->view->cache));
+ } else if (dctx->zone == NULL && dctx->cache == NULL &&
+ dctx->dumpcache)
+ {
style = &dns_master_style_cache;
/* start cache dump */
if (dctx->view->view->cachedb != NULL)
dns_db_attach(dctx->view->view->cachedb, &dctx->cache);
if (dctx->cache != NULL) {
-
- fprintf(dctx->fp, ";\n; Cache dump of view '%s'\n;\n",
- dctx->view->view->name);
+ fprintf(dctx->fp,
+ ";\n; Cache dump of view '%s' (cache %s)\n;\n",
+ dctx->view->view->name,
+ dns_cache_getname(dctx->view->view->cache));
result = dns_master_dumptostreaminc(dctx->mctx,
dctx->cache, NULL,
style, dctx->fp,
@@ -4880,6 +6239,68 @@ ns_server_dumpdb(ns_server_t *server, char *args) {
}
isc_result_t
+ns_server_dumpsecroots(ns_server_t *server, char *args) {
+ dns_view_t *view;
+ dns_keytable_t *secroots = NULL;
+ isc_result_t result;
+ char *ptr;
+ FILE *fp = NULL;
+ isc_time_t now;
+ char tbuf[64];
+
+ /* Skip the command name. */
+ ptr = next_token(&args, " \t");
+ if (ptr == NULL)
+ return (ISC_R_UNEXPECTEDEND);
+ ptr = next_token(&args, " \t");
+
+ CHECKMF(isc_stdio_open(server->secrootsfile, "w", &fp),
+ "could not open secroots dump file", server->secrootsfile);
+ TIME_NOW(&now);
+ isc_time_formattimestamp(&now, tbuf, sizeof(tbuf));
+ fprintf(fp, "%s\n", tbuf);
+
+ nextview:
+ for (view = ISC_LIST_HEAD(server->viewlist);
+ view != NULL;
+ view = ISC_LIST_NEXT(view, link))
+ {
+ if (ptr != NULL && strcmp(view->name, ptr) != 0)
+ continue;
+ if (secroots != NULL)
+ dns_keytable_detach(&secroots);
+ result = dns_view_getsecroots(view, &secroots);
+ if (result == ISC_R_NOTFOUND) {
+ result = ISC_R_SUCCESS;
+ continue;
+ }
+ fprintf(fp, "\n Start view %s\n\n", view->name);
+ CHECK(dns_keytable_dump(secroots, fp));
+ }
+ if (ptr != NULL) {
+ ptr = next_token(&args, " \t");
+ if (ptr != NULL)
+ goto nextview;
+ }
+
+ cleanup:
+ if (secroots != NULL)
+ dns_keytable_detach(&secroots);
+ if (fp != NULL)
+ (void)isc_stdio_close(fp);
+ if (result == ISC_R_SUCCESS)
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "dumpsecroots complete");
+ else
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "dumpsecroots failed: %s",
+ dns_result_totext(result));
+ return (result);
+}
+
+isc_result_t
ns_server_dumprecursing(ns_server_t *server) {
FILE *fp = NULL;
isc_result_t result;
@@ -4997,6 +6418,7 @@ ns_server_flushcache(ns_server_t *server, char *args) {
isc_boolean_t flushed;
isc_boolean_t found;
isc_result_t result;
+ ns_cache_t *nsc;
/* Skip the command name. */
ptr = next_token(&args, " \t");
@@ -5010,22 +6432,96 @@ ns_server_flushcache(ns_server_t *server, char *args) {
RUNTIME_CHECK(result == ISC_R_SUCCESS);
flushed = ISC_TRUE;
found = ISC_FALSE;
- for (view = ISC_LIST_HEAD(server->viewlist);
- view != NULL;
- view = ISC_LIST_NEXT(view, link))
- {
- if (viewname != NULL && strcasecmp(viewname, view->name) != 0)
- continue;
+
+ /*
+ * Flushing a cache is tricky when caches are shared by multiple views.
+ * We first identify which caches should be flushed in the local cache
+ * list, flush these caches, and then update other views that refer to
+ * the flushed cache DB.
+ */
+ if (viewname != NULL) {
+ /*
+ * Mark caches that need to be flushed. This is an O(#view^2)
+ * operation in the very worst case, but should be normally
+ * much more lightweight because only a few (most typically just
+ * one) views will match.
+ */
+ for (view = ISC_LIST_HEAD(server->viewlist);
+ view != NULL;
+ view = ISC_LIST_NEXT(view, link))
+ {
+ if (strcasecmp(viewname, view->name) != 0)
+ continue;
+ found = ISC_TRUE;
+ for (nsc = ISC_LIST_HEAD(server->cachelist);
+ nsc != NULL;
+ nsc = ISC_LIST_NEXT(nsc, link)) {
+ if (nsc->cache == view->cache)
+ break;
+ }
+ INSIST(nsc != NULL);
+ nsc->needflush = ISC_TRUE;
+ }
+ } else
found = ISC_TRUE;
- result = dns_view_flushcache(view);
+
+ /* Perform flush */
+ for (nsc = ISC_LIST_HEAD(server->cachelist);
+ nsc != NULL;
+ nsc = ISC_LIST_NEXT(nsc, link)) {
+ if (viewname != NULL && !nsc->needflush)
+ continue;
+ nsc->needflush = ISC_TRUE;
+ result = dns_view_flushcache2(nsc->primaryview, ISC_FALSE);
if (result != ISC_R_SUCCESS) {
flushed = ISC_FALSE;
isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
"flushing cache in view '%s' failed: %s",
- view->name, isc_result_totext(result));
+ nsc->primaryview->name,
+ isc_result_totext(result));
+ }
+ }
+
+ /*
+ * Fix up views that share a flushed cache: let the views update the
+ * cache DB they're referring to. This could also be an expensive
+ * operation, but should typically be marginal: the inner loop is only
+ * necessary for views that share a cache, and if there are many such
+ * views the number of shared cache should normally be small.
+ * A worst case is that we have n views and n/2 caches, each shared by
+ * two views. Then this will be a O(n^2/4) operation.
+ */
+ for (view = ISC_LIST_HEAD(server->viewlist);
+ view != NULL;
+ view = ISC_LIST_NEXT(view, link))
+ {
+ if (!dns_view_iscacheshared(view))
+ continue;
+ for (nsc = ISC_LIST_HEAD(server->cachelist);
+ nsc != NULL;
+ nsc = ISC_LIST_NEXT(nsc, link)) {
+ if (!nsc->needflush || nsc->cache != view->cache)
+ continue;
+ result = dns_view_flushcache2(view, ISC_TRUE);
+ if (result != ISC_R_SUCCESS) {
+ flushed = ISC_FALSE;
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_ERROR,
+ "fixing cache in view '%s' "
+ "failed: %s", view->name,
+ isc_result_totext(result));
+ }
}
}
+
+ /* Cleanup the cache list. */
+ for (nsc = ISC_LIST_HEAD(server->cachelist);
+ nsc != NULL;
+ nsc = ISC_LIST_NEXT(nsc, link)) {
+ nsc->needflush = ISC_FALSE;
+ }
+
if (flushed && found) {
if (viewname != NULL)
isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
@@ -5076,7 +6572,7 @@ ns_server_flushname(ns_server_t *server, char *args) {
isc_buffer_add(&b, strlen(target));
dns_fixedname_init(&fixed);
name = dns_fixedname_name(&fixed);
- result = dns_name_fromtext(name, &b, dns_rootname, ISC_FALSE, NULL);
+ result = dns_name_fromtext(name, &b, dns_rootname, 0, NULL);
if (result != ISC_R_SUCCESS)
return (result);
@@ -5094,6 +6590,11 @@ ns_server_flushname(ns_server_t *server, char *args) {
if (viewname != NULL && strcasecmp(viewname, view->name) != 0)
continue;
found = ISC_TRUE;
+ /*
+ * It's a little inefficient to try flushing name for all views
+ * if some of the views share a single cache. But since the
+ * operation is lightweight we prefer simplicity here.
+ */
result = dns_view_flushname(view, name);
if (result != ISC_R_SUCCESS) {
flushed = ISC_FALSE;
@@ -5408,6 +6909,46 @@ ns_server_tsiglist(ns_server_t *server, isc_buffer_t *text) {
}
/*
+ * Act on a "sign" or "loadkeys" command from the command channel.
+ */
+isc_result_t
+ns_server_rekey(ns_server_t *server, char *args) {
+ isc_result_t result;
+ dns_zone_t *zone = NULL;
+ dns_zonetype_t type;
+ isc_uint16_t keyopts;
+ isc_boolean_t fullsign = ISC_FALSE;
+
+ if (strncasecmp(args, NS_COMMAND_SIGN, strlen(NS_COMMAND_SIGN)) == 0)
+ fullsign = ISC_TRUE;
+
+ result = zone_from_args(server, args, &zone, NULL);
+ if (result != ISC_R_SUCCESS)
+ return (result);
+ if (zone == NULL)
+ return (ISC_R_UNEXPECTEDEND); /* XXX: or do all zones? */
+
+ type = dns_zone_gettype(zone);
+ if (type != dns_zone_master) {
+ dns_zone_detach(&zone);
+ return (DNS_R_NOTMASTER);
+ }
+
+ keyopts = dns_zone_getkeyopts(zone);
+
+ /* "rndc loadkeys" requires "auto-dnssec maintain". */
+ if ((keyopts & DNS_ZONEKEY_ALLOW) == 0)
+ result = ISC_R_NOPERM;
+ else if ((keyopts & DNS_ZONEKEY_MAINTAIN) == 0 && !fullsign)
+ result = ISC_R_NOPERM;
+ else
+ dns_zone_rekey(zone, fullsign);
+
+ dns_zone_detach(&zone);
+ return (result);
+}
+
+/*
* Act on a "freeze" or "thaw" command from the command channel.
*/
isc_result_t
@@ -5425,7 +6966,7 @@ ns_server_freeze(ns_server_t *server, isc_boolean_t freeze, char *args,
isc_boolean_t frozen;
const char *msg = NULL;
- result = zone_from_args(server, args, &zone);
+ result = zone_from_args(server, args, &zone, NULL);
if (result != ISC_R_SUCCESS)
return (result);
if (zone == NULL) {
@@ -5451,7 +6992,7 @@ ns_server_freeze(ns_server_t *server, isc_boolean_t freeze, char *args,
type = dns_zone_gettype(zone);
if (type != dns_zone_master) {
dns_zone_detach(&zone);
- return (ISC_R_NOTFOUND);
+ return (DNS_R_NOTMASTER);
}
result = isc_task_beginexclusive(server->task);
@@ -5502,8 +7043,8 @@ ns_server_freeze(ns_server_t *server, isc_boolean_t freeze, char *args,
strlen(msg) + 1);
view = dns_zone_getview(zone);
- if (strcmp(view->name, "_bind") == 0 ||
- strcmp(view->name, "_default") == 0)
+ if (strcmp(view->name, "_default") == 0 ||
+ strcmp(view->name, "_bind") == 0)
{
vname = "";
sep = "";
@@ -5543,3 +7084,376 @@ ns_smf_add_message(isc_buffer_t *text) {
return (ISC_R_SUCCESS);
}
#endif /* HAVE_LIBSCF */
+
+/*
+ * Act on an "addzone" command from the command channel.
+ */
+isc_result_t
+ns_server_add_zone(ns_server_t *server, char *args) {
+ isc_result_t result;
+ isc_buffer_t argbuf;
+ size_t arglen;
+ cfg_parser_t *parser = NULL;
+ cfg_obj_t *config = NULL;
+ const cfg_obj_t *vconfig = NULL;
+ const cfg_obj_t *views = NULL;
+ const cfg_obj_t *parms = NULL;
+ const cfg_obj_t *obj = NULL;
+ const cfg_listelt_t *element;
+ const char *zonename;
+ const char *classname = NULL;
+ const char *argp;
+ const char *viewname = NULL;
+ dns_rdataclass_t rdclass;
+ dns_view_t *view = 0;
+ isc_buffer_t buf, *nbuf = NULL;
+ dns_name_t dnsname;
+ dns_zone_t *zone = NULL;
+ FILE *fp = NULL;
+ struct cfg_context *cfg = NULL;
+
+ /* Try to parse the argument string */
+ arglen = strlen(args);
+ isc_buffer_init(&argbuf, args, arglen);
+ isc_buffer_add(&argbuf, strlen(args));
+ CHECK(cfg_parser_create(server->mctx, ns_g_lctx, &parser));
+ CHECK(cfg_parse_buffer(parser, &argbuf, &cfg_type_addzoneconf,
+ &config));
+ CHECK(cfg_map_get(config, "addzone", &parms));
+
+ zonename = cfg_obj_asstring(cfg_tuple_get(parms, "name"));
+ isc_buffer_init(&buf, zonename, strlen(zonename));
+ isc_buffer_add(&buf, strlen(zonename));
+ dns_name_init(&dnsname, NULL);
+ isc_buffer_allocate(server->mctx, &nbuf, 256);
+ dns_name_setbuffer(&dnsname, nbuf);
+ CHECK(dns_name_fromtext(&dnsname, &buf, dns_rootname, ISC_FALSE, NULL));
+
+ /* Make sense of optional class argument */
+ obj = cfg_tuple_get(parms, "class");
+ CHECK(ns_config_getclass(obj, dns_rdataclass_in, &rdclass));
+ if (rdclass != dns_rdataclass_in && obj)
+ classname = cfg_obj_asstring(obj);
+
+ /* Make sense of optional view argument */
+ obj = cfg_tuple_get(parms, "view");
+ if (obj && cfg_obj_isstring(obj))
+ viewname = cfg_obj_asstring(obj);
+ if (viewname == NULL || *viewname == '\0')
+ viewname = "_default";
+ CHECK(dns_viewlist_find(&server->viewlist, viewname, rdclass, &view));
+
+ /* Are we accepting new zones? */
+ if (view->new_zone_file == NULL) {
+ result = ISC_R_NOPERM;
+ goto cleanup;
+ }
+
+ cfg = (struct cfg_context *) view->new_zone_config;
+ if (cfg == NULL) {
+ result = ISC_R_FAILURE;
+ goto cleanup;
+ }
+
+ /* Zone shouldn't already exist */
+ result = dns_zt_find(view->zonetable, &dnsname, 0, NULL, &zone);
+ if (result == ISC_R_SUCCESS) {
+ result = ISC_R_EXISTS;
+ goto cleanup;
+ } else if (result == DNS_R_PARTIALMATCH) {
+ /* Create our sub-zone anyway */
+ dns_zone_detach(&zone);
+ zone = NULL;
+ }
+ else if (result != ISC_R_NOTFOUND)
+ goto cleanup;
+
+ /* Find the view statement */
+ cfg_map_get(cfg->config, "view", &views);
+ for (element = cfg_list_first(views);
+ element != NULL;
+ element = cfg_list_next(element))
+ {
+ const char *vname;
+ vconfig = cfg_listelt_value(element);
+ vname = cfg_obj_asstring(cfg_tuple_get(vconfig, "name"));
+ if (vname && !strcasecmp(vname, viewname))
+ break;
+ vconfig = NULL;
+ }
+
+ /* Open save file for write configuration */
+ CHECK(isc_stdio_open(view->new_zone_file, "a", &fp));
+
+ /* Mark view unfrozen so that zone can be added */
+ dns_view_thaw(view);
+ result = configure_zone(cfg->config, parms, vconfig,
+ server->mctx, view, &cfg->actx, ISC_FALSE);
+ dns_view_freeze(view);
+ if (result != ISC_R_SUCCESS) {
+ goto cleanup;
+ }
+
+ /* Is it there yet? */
+ CHECK(dns_zt_find(view->zonetable, &dnsname, 0, NULL, &zone));
+
+ /*
+ * Load the zone from the master file. If this fails, we'll
+ * need to undo the configuration we've done already.
+ */
+ result = dns_zone_loadnew(zone);
+ if (result != ISC_R_SUCCESS) {
+ dns_db_t *dbp = NULL;
+
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "addzone failed; reverting.");
+
+ /* If the zone loaded partially, unload it */
+ if (dns_zone_getdb(zone, &dbp) == ISC_R_SUCCESS) {
+ dns_db_detach(&dbp);
+ dns_zone_unload(zone);
+ }
+
+ /* Remove the zone from the zone table */
+ dns_zt_unmount(view->zonetable, zone);
+ goto cleanup;
+ }
+
+ /* Flag the zone as having been added at runtime */
+ dns_zone_setadded(zone, ISC_TRUE);
+
+ /* Emit just the zone name from args */
+ CHECK(isc_stdio_write("zone ", 5, 1, fp, NULL));
+ CHECK(isc_stdio_write(zonename, strlen(zonename), 1, fp, NULL));
+ CHECK(isc_stdio_write(" ", 1, 1, fp, NULL));
+
+ /* Classname, if not default */
+ if (classname != NULL && *classname != '\0') {
+ CHECK(isc_stdio_write(classname, strlen(classname), 1, fp,
+ NULL));
+ CHECK(isc_stdio_write(" ", 1, 1, fp, NULL));
+ }
+
+ /* Find beginning of option block from args */
+ for (argp = args; *argp; argp++, arglen--) {
+ if (*argp == '{') { /* Assume matching '}' */
+ /* Add that to our file */
+ CHECK(isc_stdio_write(argp, arglen, 1, fp, NULL));
+
+ /* Make sure we end with a LF */
+ if (argp[arglen-1] != '\n') {
+ CHECK(isc_stdio_write("\n", 1, 1, fp, NULL));
+ }
+ break;
+ }
+ }
+
+ CHECK(isc_stdio_close(fp));
+ fp = NULL;
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "zone %s added to view %s via addzone",
+ zonename, viewname);
+
+ result = ISC_R_SUCCESS;
+
+ cleanup:
+ if (fp != NULL)
+ isc_stdio_close(fp);
+ if (parser != NULL) {
+ if (config != NULL)
+ cfg_obj_destroy(parser, &config);
+ cfg_parser_destroy(&parser);
+ }
+ if (zone != NULL)
+ dns_zone_detach(&zone);
+ if (view != NULL)
+ dns_view_detach(&view);
+ if (nbuf != NULL)
+ isc_buffer_free(&nbuf);
+
+ return (result);
+}
+
+/*
+ * Act on a "delzone" command from the command channel.
+ */
+isc_result_t
+ns_server_del_zone(ns_server_t *server, char *args) {
+ isc_result_t result;
+ dns_zone_t *zone = NULL;
+ dns_view_t *view = NULL;
+ dns_db_t *dbp = NULL;
+ const char *filename = NULL;
+ char *tmpname = NULL;
+ char buf[1024];
+ const char *zonename = NULL;
+ size_t znamelen = 0;
+ FILE *ifp = NULL, *ofp = NULL;
+
+ /* Parse parameters */
+ CHECK(zone_from_args(server, args, &zone, &zonename));
+ if (result != ISC_R_SUCCESS)
+ return (result);
+ if (zone == NULL) {
+ result = ISC_R_UNEXPECTEDEND;
+ goto cleanup;
+ }
+
+ /*
+ * Was this zone originally added at runtime?
+ * If not, we can't delete it now.
+ */
+ if (!dns_zone_getadded(zone)) {
+ result = ISC_R_NOPERM;
+ goto cleanup;
+ }
+
+ if (zonename != NULL)
+ znamelen = strlen(zonename);
+
+ /* Dig out configuration for this zone */
+ view = dns_zone_getview(zone);
+ filename = view->new_zone_file;
+ if (filename == NULL) {
+ /* No adding zones in this view */
+ result = ISC_R_FAILURE;
+ goto cleanup;
+ }
+
+ /* Rewrite zone list */
+ result = isc_stdio_open(filename, "r", &ifp);
+ if (ifp != NULL && result == ISC_R_SUCCESS) {
+ char *found = NULL, *p = NULL;
+ size_t n;
+
+ /* Create a temporary file */
+ CHECK(isc_string_printf(buf, 1023, "%s.%ld", filename,
+ (long)getpid()));
+ if (!(tmpname = isc_mem_strdup(server->mctx, buf))) {
+ result = ISC_R_NOMEMORY;
+ goto cleanup;
+ }
+ CHECK(isc_stdio_open(tmpname, "w", &ofp));
+
+ /* Look for the entry for that zone */
+ while (fgets(buf, 1024, ifp)) {
+ /* A 'zone' line */
+ if (strncasecmp(buf, "zone", 4)) {
+ fputs(buf, ofp);
+ continue;
+ }
+ p = buf+4;
+
+ /* Locate a name */
+ while (*p &&
+ ((*p == '"') || isspace((unsigned char)*p)))
+ p++;
+
+ /* Is that the zone we're looking for */
+ if (strncasecmp(p, zonename, znamelen)) {
+ fputs(buf, ofp);
+ continue;
+ }
+
+ /* And nothing else? */
+ p += znamelen;
+ if (isspace((unsigned char)*p) ||
+ *p == '"' || *p == '{') {
+ /* This must be the entry */
+ found = p;
+ break;
+ }
+
+ /* Spit it out, keep looking */
+ fputs(buf, ofp);
+ }
+
+ /* Skip over an option block (matching # of braces) */
+ if (found) {
+ int obrace = 0, cbrace = 0;
+ for (;;) {
+ while (*p) {
+ if (*p == '{') obrace++;
+ if (*p == '}') cbrace++;
+ p++;
+ }
+ if (obrace && (obrace == cbrace))
+ break;
+ if (!fgets(buf, 1024, ifp))
+ break;
+ p = buf;
+ }
+
+ /* Just spool the remainder of the file out */
+ result = isc_stdio_read(buf, 1, 1024, ifp, &n);
+ while (n > 0U) {
+ if (result == ISC_R_EOF)
+ result = ISC_R_SUCCESS;
+ CHECK(result);
+ isc_stdio_write(buf, 1, n, ofp, NULL);
+ result = isc_stdio_read(buf, 1, 1024, ifp, &n);
+ }
+
+ /* Move temporary into place */
+ CHECK(isc_file_rename(tmpname, view->new_zone_file));
+ } else {
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_WARNING,
+ "deleted zone %s was missing from "
+ "new zone file", zonename);
+ goto cleanup;
+ }
+ }
+
+ /* Stop answering for this zone */
+ if (dns_zone_getdb(zone, &dbp) == ISC_R_SUCCESS) {
+ dns_db_detach(&dbp);
+ dns_zone_unload(zone);
+ }
+
+ CHECK(dns_zt_unmount(view->zonetable, zone));
+
+ isc_log_write(ns_g_lctx, NS_LOGCATEGORY_GENERAL,
+ NS_LOGMODULE_SERVER, ISC_LOG_INFO,
+ "zone %s removed via delzone", zonename);
+
+ result = ISC_R_SUCCESS;
+
+ cleanup:
+ if (ifp != NULL)
+ isc_stdio_close(ifp);
+ if (ofp != NULL) {
+ isc_stdio_close(ofp);
+ isc_file_remove(tmpname);
+ }
+ if (tmpname != NULL)
+ isc_mem_free(server->mctx, tmpname);
+ if (zone != NULL)
+ dns_zone_detach(&zone);
+
+ return (result);
+}
+
+static void
+cfgctx_destroy(void **cfgp) {
+ struct cfg_context *cfg;
+ isc_mem_t *mctx;
+
+ REQUIRE(cfgp != NULL && *cfgp != NULL);
+ cfg = *cfgp;
+ mctx = cfg->mctx;
+ cfg->mctx = NULL;
+
+ if (cfg->parser != NULL) {
+ if (cfg->config != NULL)
+ cfg_obj_destroy(cfg->parser, &cfg->config);
+ cfg_parser_destroy(&cfg->parser);
+ }
+ cfg_aclconfctx_clear(&cfg->actx);
+
+ isc_mem_put(mctx, cfg, sizeof(*cfg));
+ isc_mem_detach(&mctx);
+ *cfgp = NULL;
+}