5:55 a.m.
This is a fix for commit db488c79173b240459c7754f38c3c6af9b432970
dnsmasq main process exits without waiting for DAD, this is dnsmasq
daemon's task. So we periodically poll the kernel using netlink and
check whether there are any IPv6 addresses assigned to bridge
which have 'tentative' state. After DAD is finished, execution continues.
I guess that is what dnsmasq was assumed to do.
---
Difference to v1: fixed syntax (comments and alignment).
src/network/bridge_driver.c | 113 +++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 112 insertions(+), 1 deletion(-)
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 3d6721b..fb69ae5 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -2025,6 +2025,111 @@ networkAddRouteToBridge(virNetworkObjPtr network,
return 0;
}
+/* return whether there is a known address with 'tentative' flag set */
+static int
+networkParseDadStatus(struct nlmsghdr *nlh, int len,
+ virNetworkObjPtr network)
+{
+ struct ifaddrmsg *ifaddrmsg_ptr;
+ unsigned int ifaddrmsg_len;
+ struct rtattr *rtattr_ptr;
+ size_t i;
+ virNetworkIpDefPtr ipdef;
+ unsigned char prefix;
+ struct in6_addr *addr;
+ for (; NLMSG_OK(nlh, len); nlh = NLMSG_NEXT(nlh, len)) {
+ if (NLMSG_PAYLOAD(nlh, 0) < sizeof(struct ifaddrmsg)) {
+ /* Message without payload is the last one. */
+ break;
+ }
+
+ ifaddrmsg_ptr = (struct ifaddrmsg *)NLMSG_DATA(nlh);
+ if (!(ifaddrmsg_ptr->ifa_flags & IFA_F_TENTATIVE)) {
+ /* Not tentative: we are not interested in this entry. */
+ continue;
+ }
+
+ ifaddrmsg_len = IFA_PAYLOAD(nlh);
+ rtattr_ptr = (struct rtattr *) IFA_RTA(ifaddrmsg_ptr);
+ for (; RTA_OK(rtattr_ptr, ifaddrmsg_len);
+ rtattr_ptr = RTA_NEXT(rtattr_ptr, ifaddrmsg_len)) {
+ if (RTA_PAYLOAD(rtattr_ptr) != sizeof(struct in6_addr)) {
+ /* No address: ignore. */
+ continue;
+ }
+
+ /* We check only known addresses. */
+ for (i = 0;
+ (ipdef = virNetworkDefGetIpByIndex(network->def, AF_INET6, i));
+ i++) {
+
+ prefix = virNetworkIpDefPrefix(ipdef);
+ addr = &ipdef->address.data.inet6.sin6_addr;
+
+ if (prefix == ifaddrmsg_ptr->ifa_prefixlen &&
+ !memcmp(addr, RTA_DATA(rtattr_ptr),
+ sizeof(struct in6_addr))) {
+ /* We found matching tentative address. */
+ return 1;
+ }
+ }
+ }
+ }
+ return 0;
+}
+
+/* return after DAD finishes for all known IPv6 addresses or an error */
+static int
+networkWaitDadFinish(virNetworkObjPtr network)
+{
+ struct nl_msg *nlmsg = NULL;
+ struct ifaddrmsg ifa;
+ struct nlmsghdr *resp = NULL;
+ unsigned int recvbuflen;
+ int ret = -1, dad = 1;
+
+ if (!(nlmsg = nlmsg_alloc_simple(RTM_GETADDR,
+ NLM_F_REQUEST | NLM_F_DUMP))) {
+ virReportOOMError();
+ return -1;
+ }
+
+ memset(&ifa, 0, sizeof(ifa));
+ /* DAD is for IPv6 adresses only. */
+ ifa.ifa_family = AF_INET6;
+ if (nlmsg_append(nlmsg, &ifa, sizeof(ifa), NLMSG_ALIGNTO) < 0) {
+ virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
+ _("allocated netlink buffer is too small"));
+ goto cleanup;
+ }
+
+ /* Periodically query netlink until DAD finishes on all known addresses. */
+ while (dad) {
+ if (virNetlinkCommand(nlmsg, &resp, &recvbuflen, 0, 0,
+ NETLINK_ROUTE, 0) < 0)
+ goto cleanup;
+
+ if (virNetlinkGetErrorCode(resp, recvbuflen) < 0) {
+ virReportError(VIR_ERR_SYSTEM_ERROR, "%s",
+ _("error reading DAD state information"));
+ goto cleanup;
+ }
+
+ /* Parse response. */
+ dad = networkParseDadStatus(resp, recvbuflen, network);
+ if (dad)
+ usleep(1000 * 10);
+
+ VIR_FREE(resp);
+ }
+ ret = 0;
+
+ cleanup:
+ VIR_FREE(resp);
+ nlmsg_free(nlmsg);
+ return ret;
+}
+
static int
networkStartNetworkVirtual(virNetworkDriverStatePtr driver,
virNetworkObjPtr network)
@@ -2159,7 +2264,13 @@ networkStartNetworkVirtual(virNetworkDriverStatePtr driver,
if (v6present && networkStartRadvd(driver, network) < 0)
goto err4;
- /* DAD has happened (dnsmasq waits for it), dnsmasq is now bound to the
+ /* dnsmasq main process does not wait for DAD to complete,
+ * so we need to wait for it ourselves.
+ */
+ if (v6present && networkWaitDadFinish(network) < 0)
+ goto err4;
+
+ /* DAD has happened, dnsmasq is now bound to the
* bridge's IPv6 address, so we can now set the dummy tun down.
*/
if (tapfd >= 0) {
--
Sincerely,
Maxim Perevedentsev