Remove IP address dependency for management server communication

Allows the management.server.address config to be a list of hostnames instead of static IPs.
Rebalancing, cluster propagation, alerts, and stats collection all detect the format at runtime via ManagementServerAddressUtil and use the matching lookup (listNonUpStateMsHostnames vs listNonUpStateMsIPs) when computing the "avoid" list sent to agents.

Hostnames survive MS restarts and IP reassignments where static IPs do not, which is required in deployment environments where management server instances can be rescheduled or replaced.

Author: mprokopchuk

api/src/main/java/org/apache/cloudstack/management/ManagementServerHost.java

@@ -16,6 +16,7 @@
 // under the License.
 package org.apache.cloudstack.management;
 
+import com.cloud.utils.StringUtils;
 import org.apache.cloudstack.acl.ControlledEntity;
 import org.apache.cloudstack.api.Identity;
 import org.apache.cloudstack.api.InternalIdentity;
@@ -34,4 +35,15 @@ enum State {
     String getVersion();
 
     String getServiceIP();
+
+    /**
+     * Returns the node's hostname when set, otherwise falls back to the service IP, so log and
+     * alert messages always carry a usable identifier even though the name column is nullable.
+     *
+     * @return the hostname if not blank, else the service IP
+     */
+    default String getHostIdentifier() {
+        String name = getName();
+        return StringUtils.isNotBlank(name) ? name : getServiceIP();
+    }
 }

engine/components-api/src/main/java/com/cloud/agent/AgentManager.java

@@ -16,6 +16,7 @@
 // under the License.
 package com.cloud.agent;
 
+import java.util.List;
 import java.util.Map;
 
 import org.apache.cloudstack.framework.config.ConfigKey;
@@ -178,4 +179,6 @@ enum TapAgentsAction {
     boolean transferDirectAgentsFromMS(String fromMsUuid, long fromMsId, long timeoutDurationInMs, boolean excludeHostsInMaintenance);
 
     int getHostSshPort(HostVO host);
+
+    List<String> getAvoidMsList();
 }

engine/orchestration/src/main/java/com/cloud/agent/manager/AgentManagerImpl.java

@@ -49,6 +49,7 @@
 
 import com.cloud.agent.api.AgentConnectStatusAnswer;
 import com.cloud.agent.api.AgentConnectStatusCommand;
+import com.cloud.cluster.ManagementServerAddressUtil;
 import com.cloud.event.EventTypes;
 import com.cloud.utils.DateUtil;
 import com.cloud.utils.Profiler;
@@ -1859,10 +1860,16 @@ private void updateReadyCommandWithMSList(HostVO host, ReadyCommand ready, Start
         }
     }
 
+    public List<String> getAvoidMsList() {
+        // Detect config format and build avoid list using matching format
+        boolean isUsingHostnames = ManagementServerAddressUtil.isManagementServerAddressListUsingHostnames();
+        return isUsingHostnames ? _mshostDao.listNonUpStateMsHostnames() : _mshostDao.listNonUpStateMsIPs();
+    }
+
     private void setReadyCommandMSList(HostVO host, ReadyCommand ready) {
         List<String> newMSList = indirectAgentLB.getManagementServerList(host.getId(), host.getDataCenterId(), null);
         ready.setMsHostList(newMSList);
-        List<String> avoidMsList = _mshostDao.listNonUpStateMsIPs();
+        List<String> avoidMsList = getAvoidMsList();
         ready.setAvoidMsHostList(avoidMsList);
         ready.setLbAlgorithm(indirectAgentLB.getLBAlgorithmName());
         ready.setLbCheckInterval(indirectAgentLB.getLBPreferredHostCheckInterval(host.getClusterId()));
@@ -2381,10 +2388,6 @@ private boolean sendRequestStartupCommand(long hostId, HostVO host) {
             return requestStartup;
         }
 
-        private List<String> getAvoidMsList() {
-            return _mshostDao.listNonUpStateMsIPs();
-        }
-
         protected void processResponse(final Link link, final Response response) {
             final AgentAttache attache = (AgentAttache)link.attachment();
             if (attache == null) {

engine/orchestration/src/main/java/com/cloud/agent/manager/ClusteredAgentManagerImpl.java

@@ -43,6 +43,7 @@
 import javax.net.ssl.SSLEngine;
 
 import com.cloud.resource.ResourceState;
+import com.cloud.utils.StringUtils;
 import org.apache.cloudstack.ca.CAManager;
 import org.apache.cloudstack.framework.config.ConfigDepot;
 import org.apache.cloudstack.framework.config.ConfigKey;
@@ -541,14 +542,9 @@ public SocketChannel connectToPeer(final String peerName, final SocketChannel pr
                     logger.info("Unable to find peer: {}",  peerName);
                     return null;
                 }
-                final String ip = ms.getServiceIP();
-                InetAddress addr;
-                int port = Port.value();
-                try {
-                    addr = InetAddress.getByName(ip);
-                } catch (final UnknownHostException e) {
-                    throw new CloudRuntimeException("Unable to resolve " + ip);
-                }
+                final String hostName = getHostIdentifier(ms);
+                final int port = Port.value();
+                final InetAddress addr = resolveManagementServerAddress(ms);
                 SocketChannel ch1 = null;
                 try {
                     ch1 = SocketChannel.open(new InetSocketAddress(addr, port));
@@ -557,20 +553,20 @@ public SocketChannel connectToPeer(final String peerName, final SocketChannel pr
                     ch1.socket().setSoTimeout(60 * 1000);
                     try {
                         SSLContext sslContext = Link.initManagementSSLContext(caService);
-                        sslEngine = sslContext.createSSLEngine(ip, port);
+                        sslEngine = sslContext.createSSLEngine(hostName, port);
                         sslEngine.setUseClientMode(true);
                         sslEngine.setEnabledProtocols(SSLUtils.getSupportedProtocols(sslEngine.getEnabledProtocols()));
                         sslEngine.beginHandshake();
                         if (!Link.doHandshake(ch1, sslEngine)) {
                             ch1.close();
-                            throw new IOException(String.format("SSL: Handshake failed with peer management server '%s' on %s:%d ", peerName, ip, port));
+                            throw new IOException(String.format("SSL: Handshake failed with peer management server '%s' on %s:%d ", peerName, hostName, port));
                         }
-                        logger.info("SSL: Handshake done with peer management server '{}' on {}:{} ", peerName, ip, port);
+                        logger.info("SSL: Handshake done with peer management server '{}' on {}:{} ", peerName, hostName, port);
                     } catch (final Exception e) {
                         ch1.close();
                         throw new IOException("SSL: Fail to init SSL! " + e);
                     }
-                    logger.debug("Connection to peer opened: {}, IP: {}", peerName, ip);
+                    logger.debug("Connection to peer opened: {}, host: {}", peerName, hostName);
                     _peers.put(peerName, ch1);
                     _sslEngines.put(peerName, sslEngine);
                     return ch1;
@@ -582,7 +578,7 @@ public SocketChannel connectToPeer(final String peerName, final SocketChannel pr
                             logger.error("failed to close failed peer socket: {}",  ex);
                         }
                     }
-                    logger.warn("Unable to connect to peer management server: {}, IP {} due to {}", peerName, ip, e.getMessage(), e);
+                    logger.warn("Unable to connect to peer management server: {}, host {} due to {}", peerName, hostName, e.getMessage(), e);
                     return null;
                 }
             }
@@ -592,6 +588,38 @@ public SocketChannel connectToPeer(final String peerName, final SocketChannel pr
         }
     }
 
+    /**
+     * Gets the hostname or IP address to use for connecting to a management server.
+     * Prefers hostname (for CNAME support) but falls back to IP if hostname is null.
+     *
+     * @param host the management server host
+     * @return hostname if available, otherwise service IP
+     * @throws CloudRuntimeException if both hostname and IP are null
+     */
+    private String getHostIdentifier(final ManagementServerHost host) {
+        final String hostName = host.getName();
+        if (StringUtils.isNotBlank(hostName)) {
+            return hostName;
+        }
+
+        final String serviceIP = host.getServiceIP();
+        if (StringUtils.isNotBlank(serviceIP)) {
+            return serviceIP;
+        }
+
+        throw new CloudRuntimeException("Management server host has neither hostname nor IP address: " + host);
+    }
+
+    private InetAddress resolveManagementServerAddress(final ManagementServerHost host) {
+        final String addressToResolve = getHostIdentifier(host);
+
+        try {
+            return InetAddress.getByName(addressToResolve);
+        } catch (final UnknownHostException e) {
+            throw new CloudRuntimeException("Unable to resolve " + addressToResolve, e);
+        }
+    }
+
     public SocketChannel connectToPeer(final long hostId, final SocketChannel prevCh) {
         final String peerName = getPeerName(hostId);
         if (peerName == null) {