From 972b8b71d7fb60a64e8a05e7555f9f92ede7ff40 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafael=20Weing=C3=A4rtner?= Date: Wed, 14 Mar 2018 15:27:43 -0300 Subject: [PATCH] CLOUDSTACK-8855 Improve Error Message for Host Alert State and reconnect host API. (#2387) * CLOUDSTACK-8855 Improve Error Message for Host Alert State * [CLOUDSTACK-9846] create column to save the content of alert messages Remove declaration of throws CloudRuntimeException I also removed some unused variables and comments left behind This closes #837 * Isolate a problematic test "smoke/test_certauthority_root" --- .travis.yml | 3 +- api/src/main/java/com/cloud/alert/Alert.java | 2 + .../com/cloud/resource/ResourceService.java | 21 +-- .../command/admin/host/ReconnectHostCmd.java | 28 +-- .../java/com/cloud/agent/AgentManager.java | 4 +- .../cloud/agent/manager/AgentManagerImpl.java | 173 ++++++++---------- .../manager/ClusteredAgentManagerImpl.java | 110 +++++------ .../main/java/com/cloud/alert/AlertVO.java | 12 ++ .../java/com/cloud/host/dao/HostDaoImpl.java | 1 - .../db/schema-41000to41100-cleanup.sql | 2 +- .../META-INF/db/schema-41000to41100.sql | 2 +- .../META-INF/db/schema-41100to41200.sql | 5 +- .../test/DirectAgentManagerSimpleImpl.java | 6 +- .../network/element/NetscalerElement.java | 116 ++++++------ requirements.txt | 3 +- .../com/cloud/alert/AlertManagerImpl.java | 47 +++-- .../cloud/resource/ResourceManagerImpl.java | 37 +--- .../apache/cloudstack/ca/CAManagerImpl.java | 51 +++--- .../cloudstack/ca/CAManagerImplTest.java | 2 +- 19 files changed, 286 insertions(+), 339 deletions(-) diff --git a/.travis.yml b/.travis.yml index be0bfb12210..f4b52bae945 100644 --- a/.travis.yml +++ b/.travis.yml @@ -33,10 +33,11 @@ env: - PATH=$HOME/.local/bin:$PATH matrix: # Keep the TESTS sorted by name and grouped by type + - TESTS="smoke/test_certauthority_root" + - TESTS="smoke/test_accounts smoke/test_affinity_groups smoke/test_affinity_groups_projects - smoke/test_certauthority_root smoke/test_deploy_vgpu_enabled_vm smoke/test_deploy_vm_iso smoke/test_deploy_vm_root_resize diff --git a/api/src/main/java/com/cloud/alert/Alert.java b/api/src/main/java/com/cloud/alert/Alert.java index f77442068d6..d85dca5b7b0 100644 --- a/api/src/main/java/com/cloud/alert/Alert.java +++ b/api/src/main/java/com/cloud/alert/Alert.java @@ -41,4 +41,6 @@ public interface Alert extends Identity, InternalIdentity { boolean getArchived(); String getName(); + + String getContent(); } diff --git a/api/src/main/java/com/cloud/resource/ResourceService.java b/api/src/main/java/com/cloud/resource/ResourceService.java index 854b53591d6..70823084d84 100644 --- a/api/src/main/java/com/cloud/resource/ResourceService.java +++ b/api/src/main/java/com/cloud/resource/ResourceService.java @@ -18,7 +18,6 @@ package com.cloud.resource; import java.util.List; -import com.cloud.dc.DataCenter; import org.apache.cloudstack.api.command.admin.cluster.AddClusterCmd; import org.apache.cloudstack.api.command.admin.cluster.DeleteClusterCmd; import org.apache.cloudstack.api.command.admin.host.AddHostCmd; @@ -29,6 +28,8 @@ import org.apache.cloudstack.api.command.admin.host.ReconnectHostCmd; import org.apache.cloudstack.api.command.admin.host.UpdateHostCmd; import org.apache.cloudstack.api.command.admin.host.UpdateHostPasswordCmd; +import com.cloud.dc.DataCenter; +import com.cloud.exception.AgentUnavailableException; import com.cloud.exception.DiscoveryException; import com.cloud.exception.InvalidParameterValueException; import com.cloud.exception.ResourceInUseException; @@ -41,25 +42,17 @@ public interface ResourceService { /** * Updates a host * - * @param cmd - * - the command specifying hostId - * @return hostObject - * @throws NoTransitionException + * @param cmd - the command specifying hostId */ Host updateHost(UpdateHostCmd cmd) throws NoTransitionException; Host cancelMaintenance(CancelMaintenanceCmd cmd); - Host reconnectHost(ReconnectHostCmd cmd); + Host reconnectHost(ReconnectHostCmd cmd) throws AgentUnavailableException; /** * We will automatically create an Apache CloudStack cluster to attach to the external cluster and return a hyper host to perform * host related operation within the cluster - * - * @param cmd - * @return - * @throws IllegalArgumentException - * @throws DiscoveryException */ List discoverCluster(AddClusterCmd cmd) throws IllegalArgumentException, DiscoveryException, ResourceInUseException; @@ -75,12 +68,6 @@ public interface ResourceService { /** * Deletes a host - * - * @param hostId - * TODO - * @param isForced - * TODO - * * @param true if deleted, false otherwise */ boolean deleteHost(long hostId, boolean isForced, boolean isForceDeleteStorage); diff --git a/api/src/main/java/org/apache/cloudstack/api/command/admin/host/ReconnectHostCmd.java b/api/src/main/java/org/apache/cloudstack/api/command/admin/host/ReconnectHostCmd.java index 5e156372681..34e439f3a37 100644 --- a/api/src/main/java/org/apache/cloudstack/api/command/admin/host/ReconnectHostCmd.java +++ b/api/src/main/java/org/apache/cloudstack/api/command/admin/host/ReconnectHostCmd.java @@ -16,8 +16,6 @@ // under the License. package org.apache.cloudstack.api.command.admin.host; -import org.apache.log4j.Logger; - import org.apache.cloudstack.api.APICommand; import org.apache.cloudstack.api.ApiCommandJobType; import org.apache.cloudstack.api.ApiConstants; @@ -27,13 +25,16 @@ import org.apache.cloudstack.api.Parameter; import org.apache.cloudstack.api.ServerApiException; import org.apache.cloudstack.api.response.HostResponse; import org.apache.cloudstack.context.CallContext; +import org.apache.log4j.Logger; import com.cloud.event.EventTypes; +import com.cloud.exception.AgentUnavailableException; +import com.cloud.exception.InvalidParameterValueException; import com.cloud.host.Host; import com.cloud.user.Account; +import com.cloud.utils.exception.CloudRuntimeException; -@APICommand(name = "reconnectHost", description = "Reconnects a host.", responseObject = HostResponse.class, - requestHasSensitiveInfo = false, responseHasSensitiveInfo = false) +@APICommand(name = "reconnectHost", description = "Reconnects a host.", responseObject = HostResponse.class, requestHasSensitiveInfo = false, responseHasSensitiveInfo = false) public class ReconnectHostCmd extends BaseAsyncCmd { public static final Logger s_logger = Logger.getLogger(ReconnectHostCmd.class.getName()); @@ -101,16 +102,15 @@ public class ReconnectHostCmd extends BaseAsyncCmd { public void execute() { try { Host result = _resourceService.reconnectHost(this); - if (result != null) { - HostResponse response = _responseGenerator.createHostResponse(result); - response.setResponseName(getCommandName()); - this.setResponseObject(response); - } else { - throw new ServerApiException(ApiErrorCode.INTERNAL_ERROR, "Failed to reconnect host"); - } - } catch (Exception ex) { - s_logger.warn("Exception: ", ex); - throw new ServerApiException(ApiErrorCode.RESOURCE_UNAVAILABLE_ERROR, ex.getMessage()); + HostResponse response = _responseGenerator.createHostResponse(result); + response.setResponseName(getCommandName()); + this.setResponseObject(response); + } catch (InvalidParameterValueException e) { + throw new ServerApiException(ApiErrorCode.PARAM_ERROR, e.getMessage()); + } catch (CloudRuntimeException e) { + throw new ServerApiException(ApiErrorCode.INTERNAL_ERROR, e.getMessage()); + } catch (AgentUnavailableException e) { + throw new ServerApiException(ApiErrorCode.RESOURCE_UNAVAILABLE_ERROR, e.getMessage()); } } } diff --git a/engine/components-api/src/main/java/com/cloud/agent/AgentManager.java b/engine/components-api/src/main/java/com/cloud/agent/AgentManager.java index 933c3eaef99..c51970c85f7 100644 --- a/engine/components-api/src/main/java/com/cloud/agent/AgentManager.java +++ b/engine/components-api/src/main/java/com/cloud/agent/AgentManager.java @@ -36,7 +36,7 @@ import com.cloud.resource.ServerResource; */ public interface AgentManager { static final ConfigKey Wait = new ConfigKey("Advanced", Integer.class, "wait", "1800", "Time in seconds to wait for control commands to return", - true); + true); public enum TapAgentsAction { Add, Del, Contains, @@ -143,7 +143,7 @@ public interface AgentManager { public void pullAgentOutMaintenance(long hostId); - boolean reconnect(long hostId); + void reconnect(long hostId) throws AgentUnavailableException; void rescan(); diff --git a/engine/orchestration/src/main/java/com/cloud/agent/manager/AgentManagerImpl.java b/engine/orchestration/src/main/java/com/cloud/agent/manager/AgentManagerImpl.java index b7357756c4c..4b80fb1bc11 100644 --- a/engine/orchestration/src/main/java/com/cloud/agent/manager/AgentManagerImpl.java +++ b/engine/orchestration/src/main/java/com/cloud/agent/manager/AgentManagerImpl.java @@ -38,7 +38,6 @@ import javax.inject.Inject; import javax.naming.ConfigurationException; import org.apache.cloudstack.ca.CAManager; -import com.cloud.configuration.ManagementServiceConfiguration; import org.apache.cloudstack.framework.config.ConfigKey; import org.apache.cloudstack.framework.config.Configurable; import org.apache.cloudstack.framework.config.dao.ConfigurationDao; @@ -75,6 +74,7 @@ import com.cloud.agent.api.UnsupportedAnswer; import com.cloud.agent.transport.Request; import com.cloud.agent.transport.Response; import com.cloud.alert.AlertManager; +import com.cloud.configuration.ManagementServiceConfiguration; import com.cloud.dc.ClusterVO; import com.cloud.dc.DataCenterVO; import com.cloud.dc.HostPodVO; @@ -122,7 +122,6 @@ import com.cloud.utils.time.InaccurateClock; **/ public class AgentManagerImpl extends ManagerBase implements AgentManager, HandlerFactory, Configurable { protected static final Logger s_logger = Logger.getLogger(AgentManagerImpl.class); - protected static final Logger status_logger = Logger.getLogger(Status.class); /** * _agents is a ConcurrentHashMap, but it is used from within a synchronized block. This will be reported by findbugs as JLM_JSR166_UTILCONCURRENT_MONITORENTER. Maybe a @@ -133,7 +132,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl protected List> _cmdMonitors = new ArrayList>(17); protected List> _creationMonitors = new ArrayList>(17); protected List _loadingAgents = new ArrayList(); - protected int _monitorId = 0; + private int _monitorId = 0; private final Lock _agentStatusLock = new ReentrantLock(); @Inject @@ -180,28 +179,22 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl @Inject ResourceManager _resourceMgr; - @Inject ManagementServiceConfiguration mgmtServiceConf; protected final ConfigKey Workers = new ConfigKey("Advanced", Integer.class, "workers", "5", - "Number of worker threads handling remote agent connections.", false); + "Number of worker threads handling remote agent connections.", false); protected final ConfigKey Port = new ConfigKey("Advanced", Integer.class, "port", "8250", "Port to listen on for remote agent connections.", false); protected final ConfigKey AlertWait = new ConfigKey("Advanced", Integer.class, "alert.wait", "1800", - "Seconds to wait before alerting on a disconnected agent", true); + "Seconds to wait before alerting on a disconnected agent", true); protected final ConfigKey DirectAgentLoadSize = new ConfigKey("Advanced", Integer.class, "direct.agent.load.size", "16", - "The number of direct agents to load each time", false); + "The number of direct agents to load each time", false); protected final ConfigKey DirectAgentPoolSize = new ConfigKey("Advanced", Integer.class, "direct.agent.pool.size", "500", - "Default size for DirectAgentPool", false); + "Default size for DirectAgentPool", false); protected final ConfigKey DirectAgentThreadCap = new ConfigKey("Advanced", Float.class, "direct.agent.thread.cap", "1", - "Percentage (as a value between 0 and 1) of direct.agent.pool.size to be used as upper thread cap for a single direct agent to process requests", false); - protected final ConfigKey CheckTxnBeforeSending = new ConfigKey( - "Developer", - Boolean.class, - "check.txn.before.sending.agent.commands", - "false", - "This parameter allows developers to enable a check to see if a transaction wraps commands that are sent to the resource. This is not to be enabled on production systems.", - true); + "Percentage (as a value between 0 and 1) of direct.agent.pool.size to be used as upper thread cap for a single direct agent to process requests", false); + protected final ConfigKey CheckTxnBeforeSending = new ConfigKey("Developer", Boolean.class, "check.txn.before.sending.agent.commands", "false", + "This parameter allows developers to enable a check to see if a transaction wraps commands that are sent to the resource. This is not to be enabled on production systems.", true); @Override public boolean configure(final String name, final Map params) throws ConfigurationException { @@ -241,7 +234,6 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl return true; } - @Override public Task create(final Task.Type type, final Link link, final byte[] data) { return new AgentHandler(type, link, data); @@ -436,8 +428,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl if (CheckTxnBeforeSending.value()) { if (!noDbTxn()) { throw new CloudRuntimeException("We do not allow transactions to be wrapped around commands sent to be executed on remote agents. " - + "We cannot predict how long it takes a command to complete. " - + "The transaction may be rolled back because the connection took too long."); + + "We cannot predict how long it takes a command to complete. " + "The transaction may be rolled back because the connection took too long."); } } else { assert noDbTxn() : "I know, I know. Why are we so strict as to not allow txn across an agent call? ... Why are we so cruel ... Why are we such a dictator .... Too bad... Sorry...but NO AGENT COMMANDS WRAPPED WITHIN DB TRANSACTIONS!"; @@ -564,15 +555,13 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl monitor.second().processConnect(host, cmd[i], forRebalance); } catch (final Exception e) { if (e instanceof ConnectionException) { - final ConnectionException ce = (ConnectionException) e; + final ConnectionException ce = (ConnectionException)e; if (ce.isSetupError()) { - s_logger.warn("Monitor " + monitor.second().getClass().getSimpleName() + " says there is an error in the connect process for " + hostId + - " due to " + e.getMessage()); + s_logger.warn("Monitor " + monitor.second().getClass().getSimpleName() + " says there is an error in the connect process for " + hostId + " due to " + e.getMessage()); handleDisconnectWithoutInvestigation(attache, Event.AgentDisconnected, true, true); throw ce; } else { - s_logger.info("Monitor " + monitor.second().getClass().getSimpleName() + " says not to continue the connect process for " + hostId + - " due to " + e.getMessage()); + s_logger.info("Monitor " + monitor.second().getClass().getSimpleName() + " says not to continue the connect process for " + hostId + " due to " + e.getMessage()); handleDisconnectWithoutInvestigation(attache, Event.ShutdownRequested, true, true); return attache; } @@ -580,8 +569,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl handleDisconnectWithoutInvestigation(attache, Event.ShutdownRequested, true, true); throw new CloudRuntimeException("Unable to connect " + attache.getId(), e); } else { - s_logger.error("Monitor " + monitor.second().getClass().getSimpleName() + " says there is an error in the connect process for " + hostId + - " due to " + e.getMessage(), e); + s_logger.error("Monitor " + monitor.second().getClass().getSimpleName() + " says there is an error in the connect process for " + hostId + " due to " + e.getMessage(), e); handleDisconnectWithoutInvestigation(attache, Event.AgentDisconnected, true, true); throw new CloudRuntimeException("Unable to connect " + attache.getId(), e); } @@ -634,7 +622,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl try { final Class clazz = Class.forName(resourceName); final Constructor constructor = clazz.getConstructor(); - resource = (ServerResource) constructor.newInstance(); + resource = (ServerResource)constructor.newInstance(); } catch (final ClassNotFoundException e) { s_logger.warn("Unable to find class " + host.getResource(), e); } catch (final InstantiationException e) { @@ -888,7 +876,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl final String hostDesc = "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podVO.getName(); if (host.getType() != Host.Type.SecondaryStorage && host.getType() != Host.Type.ConsoleProxy) { _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Host disconnected, " + hostDesc, - "If the agent for host [" + hostDesc + "] is not restarted within " + AlertWait + " seconds, host will go to Alert state"); + "If the agent for host [" + hostDesc + "] is not restarted within " + AlertWait + " seconds, host will go to Alert state"); } event = Status.Event.AgentDisconnected; } @@ -899,7 +887,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl final String podName = podVO != null ? podVO.getName() : "NO POD"; final String hostDesc = "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podName; _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Host in ALERT state, " + hostDesc, - "In availability zone " + host.getDataCenterId() + ", host is in alert state: " + host.getId() + "-" + host.getName()); + "In availability zone " + host.getDataCenterId() + ", host is in alert state: " + host.getId() + "-" + host.getName()); } } else { s_logger.debug("The next status of agent " + host.getId() + " is not Alert, no need to investigate what happened"); @@ -987,33 +975,30 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } @Override - public boolean reconnect(final long hostId) { - HostVO host; + public void reconnect(final long hostId) throws AgentUnavailableException { + HostVO host = _hostDao.findById(hostId); + if (host == null) { + throw new CloudRuntimeException("Unable to find host: " + hostId); + } - host = _hostDao.findById(hostId); - if (host == null || host.getRemoved() != null) { - s_logger.warn("Unable to find host " + hostId); - return false; + if (host.getRemoved() != null) { + throw new CloudRuntimeException("Host has already been removed: " + hostId); } if (host.getStatus() == Status.Disconnected) { - s_logger.info("Host is already disconnected, no work to be done"); - return true; + s_logger.debug("Host is already disconnected, no work to be done: " + hostId); + return; } if (host.getStatus() != Status.Up && host.getStatus() != Status.Alert && host.getStatus() != Status.Rebalancing) { - s_logger.info("Unable to disconnect host because it is not in the correct state: host=" + hostId + "; Status=" + host.getStatus()); - return false; + throw new CloudRuntimeException("Unable to disconnect host because it is not in the correct state: host=" + hostId + "; Status=" + host.getStatus()); } - final AgentAttache attache = findAttache(hostId); + AgentAttache attache = findAttache(hostId); if (attache == null) { - s_logger.info("Unable to disconnect host because it is not connected to this server: " + hostId); - return false; + throw new CloudRuntimeException("Unable to disconnect host because it is not connected to this server: " + hostId); } - disconnectWithoutInvestigation(attache, Event.ShutdownRequested); - return true; } @Override @@ -1049,8 +1034,15 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl handleDisconnectWithoutInvestigation(attache, Event.AgentDisconnected, true, true); } return true; - } else if (event == Event.ShutdownRequested) { - return reconnect(hostId); + } + if (event == Event.ShutdownRequested) { + try { + reconnect(hostId); + } catch (CloudRuntimeException e) { + s_logger.debug("Error on shutdown request for hostID: " + hostId, e); + return false; + } + return true; } return false; } @@ -1166,7 +1158,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl _request.logD("Processing the first command "); final StartupCommand[] startups = new StartupCommand[_cmds.length]; for (int i = 0; i < _cmds.length; i++) { - startups[i] = (StartupCommand) _cmds[i]; + startups[i] = (StartupCommand)_cmds[i]; } final AgentAttache attache = handleConnectedAgent(_link, startups, _request); @@ -1184,7 +1176,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl for (int i = 0; i < cmds.length; i++) { cmd = cmds[i]; if (cmd instanceof StartupRoutingCommand || cmd instanceof StartupProxyCommand || cmd instanceof StartupSecondaryStorageCommand || - cmd instanceof StartupStorageCommand) { + cmd instanceof StartupStorageCommand) { answers[i] = new StartupAnswer((StartupCommand) cmds[i], 0, mgmtServiceConf.getPingInterval()); break; } @@ -1205,7 +1197,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } protected void processRequest(final Link link, final Request request) { - final AgentAttache attache = (AgentAttache) link.attachment(); + final AgentAttache attache = (AgentAttache)link.attachment(); final Command[] cmds = request.getCommands(); Command cmd = cmds[0]; boolean logD = true; @@ -1256,10 +1248,9 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl final StartupStorageCommand startup = (StartupStorageCommand) cmd; answer = new StartupAnswer(startup, attache.getId(), mgmtServiceConf.getPingInterval()); } else if (cmd instanceof ShutdownCommand) { - final ShutdownCommand shutdown = (ShutdownCommand) cmd; + final ShutdownCommand shutdown = (ShutdownCommand)cmd; final String reason = shutdown.getReason(); - s_logger.info("Host " + attache.getId() + " has informed us that it is shutting down with reason " + reason + " and detail " + - shutdown.getDetail()); + s_logger.info("Host " + attache.getId() + " has informed us that it is shutting down with reason " + reason + " and detail " + shutdown.getDetail()); if (reason.equals(ShutdownCommand.Update)) { // disconnectWithoutInvestigation(attache, Event.UpdateNeeded); throw new CloudRuntimeException("Agent update not implemented"); @@ -1268,16 +1259,16 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } return; } else if (cmd instanceof AgentControlCommand) { - answer = handleControlCommand(attache, (AgentControlCommand) cmd); + answer = handleControlCommand(attache, (AgentControlCommand)cmd); } else { - handleCommands(attache, request.getSequence(), new Command[] { cmd }); + handleCommands(attache, request.getSequence(), new Command[] {cmd}); if (cmd instanceof PingCommand) { - final long cmdHostId = ((PingCommand) cmd).getHostId(); + final long cmdHostId = ((PingCommand)cmd).getHostId(); // if the router is sending a ping, verify the // gateway was pingable if (cmd instanceof PingRoutingCommand) { - final boolean gatewayAccessible = ((PingRoutingCommand) cmd).isGatewayAccessible(); + final boolean gatewayAccessible = ((PingRoutingCommand)cmd).isGatewayAccessible(); final HostVO host = _hostDao.findById(Long.valueOf(cmdHostId)); if (host != null) { @@ -1286,22 +1277,18 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl // gateway (cannot ping the default route) final DataCenterVO dcVO = _dcDao.findById(host.getDataCenterId()); final HostPodVO podVO = _podDao.findById(host.getPodId()); - final String hostDesc = - "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " - + podVO.getName(); + final String hostDesc = "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podVO.getName(); - _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_ROUTING, host.getDataCenterId(), host.getPodId(), - "Host lost connection to gateway, " + hostDesc, "Host [" + hostDesc + - "] lost connection to gateway (default route) and is possibly having network connection issues."); + _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_ROUTING, host.getDataCenterId(), host.getPodId(), "Host lost connection to gateway, " + hostDesc, + "Host [" + hostDesc + "] lost connection to gateway (default route) and is possibly having network connection issues."); } else { _alertMgr.clearAlert(AlertManager.AlertType.ALERT_TYPE_ROUTING, host.getDataCenterId(), host.getPodId()); } } else { - s_logger.debug("Not processing " + PingRoutingCommand.class.getSimpleName() + " for agent id=" + cmdHostId + - "; can't find the host in the DB"); + s_logger.debug("Not processing " + PingRoutingCommand.class.getSimpleName() + " for agent id=" + cmdHostId + "; can't find the host in the DB"); } } - answer = new PingAnswer((PingCommand) cmd); + answer = new PingAnswer((PingCommand)cmd); } else if (cmd instanceof ReadyAnswer) { final HostVO host = _hostDao.findById(attache.getId()); if (host == null) { @@ -1337,7 +1324,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } protected void processResponse(final Link link, final Response response) { - final AgentAttache attache = (AgentAttache) link.attachment(); + final AgentAttache attache = (AgentAttache)link.attachment(); if (attache == null) { s_logger.warn("Unable to process: " + response); } else if (!attache.processAnswers(response.getSequence(), response)) { @@ -1355,7 +1342,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl try { final Request event = Request.parse(data); if (event instanceof Response) { - processResponse(task.getLink(), (Response) event); + processResponse(task.getLink(), (Response)event); } else { processRequest(task.getLink(), event); } @@ -1370,7 +1357,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } else if (type == Task.Type.CONNECT) { } else if (type == Task.Type.DISCONNECT) { final Link link = task.getLink(); - final AgentAttache attache = (AgentAttache) link.attachment(); + final AgentAttache attache = (AgentAttache)link.attachment(); if (attache != null) { disconnectWithInvestigation(attache, Event.AgentDisconnected); } else { @@ -1411,23 +1398,21 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl public boolean agentStatusTransitTo(final HostVO host, final Status.Event e, final long msId) { try { _agentStatusLock.lock(); - if (status_logger.isDebugEnabled()) { + if (s_logger.isDebugEnabled()) { final ResourceState state = host.getResourceState(); final StringBuilder msg = new StringBuilder("Transition:"); msg.append("[Resource state = ").append(state); msg.append(", Agent event = ").append(e.toString()); msg.append(", Host id = ").append(host.getId()).append(", name = " + host.getName()).append("]"); - status_logger.debug(msg); + s_logger.debug(msg); } host.setManagementServerId(msId); try { return _statusStateMachine.transitTo(host, e, host.getId(), _hostDao); } catch (final NoTransitionException e1) { - status_logger.debug("Cannot transit agent status with event " + e + " for host " + host.getId() + ", name=" + host.getName() + - ", mangement server id is " + msId); - throw new CloudRuntimeException("Cannot transit agent status with event " + e + " for host " + host.getId() + ", mangement server id is " + msId + "," + - e1.getMessage()); + s_logger.debug("Cannot transit agent status with event " + e + " for host " + host.getId() + ", name=" + host.getName() + ", mangement server id is " + msId); + throw new CloudRuntimeException("Cannot transit agent status with event " + e + " for host " + host.getId() + ", mangement server id is " + msId + "," + e1.getMessage()); } } finally { _agentStatusLock.unlock(); @@ -1491,6 +1476,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } } + @Override public void disconnectWithInvestigation(final long hostId, final Status.Event event) { disconnectInternal(hostId, event, true); } @@ -1501,8 +1487,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl } @Override - public boolean handleDirectConnectAgent(final Host host, final StartupCommand[] cmds, final ServerResource resource, - final boolean forRebalance, boolean newHost) throws ConnectionException { + public boolean handleDirectConnectAgent(final Host host, final StartupCommand[] cmds, final ServerResource resource, final boolean forRebalance, boolean newHost) throws ConnectionException { AgentAttache attache; attache = createAttacheForDirectConnect(host, resource); @@ -1579,17 +1564,17 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl /* * Host is in non-operation state, so no investigation and direct put agent to Disconnected */ - status_logger.debug("Ping timeout but agent " + agentId + " is in resource state of " + resourceState + ", so no investigation"); + s_logger.debug("Ping timeout but agent " + agentId + " is in resource state of " + resourceState + ", so no investigation"); disconnectWithoutInvestigation(agentId, Event.ShutdownRequested); } else { final HostVO host = _hostDao.findById(agentId); - if (host != null && (host.getType() == Host.Type.ConsoleProxy || host.getType() == Host.Type.SecondaryStorageVM - || host.getType() == Host.Type.SecondaryStorageCmdExecutor)) { + if (host != null + && (host.getType() == Host.Type.ConsoleProxy || host.getType() == Host.Type.SecondaryStorageVM || host.getType() == Host.Type.SecondaryStorageCmdExecutor)) { s_logger.warn("Disconnect agent for CPVM/SSVM due to physical connection close. host: " + host.getId()); disconnectWithoutInvestigation(agentId, Event.ShutdownRequested); } else { - status_logger.debug("Ping timeout for agent " + agentId + ", do invstigation"); + s_logger.debug("Ping timeout for agent " + agentId + ", do invstigation"); disconnectWithInvestigation(agentId, Event.PingTimeout); } } @@ -1605,8 +1590,8 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl final DataCenterVO dcVO = _dcDao.findById(host.getDataCenterId()); final HostPodVO podVO = _podDao.findById(host.getPodId()); final String hostDesc = "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podVO.getName(); - _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Migration Complete for host " + hostDesc, "Host [" - + hostDesc + "] is ready for maintenance"); + _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Migration Complete for host " + hostDesc, + "Host [" + hostDesc + "] is ready for maintenance"); } } } catch (final Throwable th) { @@ -1708,7 +1693,7 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl @Override public ConfigKey[] getConfigKeys() { return new ConfigKey[] { CheckTxnBeforeSending, Workers, Port, Wait, AlertWait, DirectAgentLoadSize, DirectAgentPoolSize, - DirectAgentThreadCap }; + DirectAgentThreadCap }; } protected class SetHostParamsListener implements Listener { @@ -1738,20 +1723,20 @@ public class AgentManagerImpl extends ManagerBase implements AgentManager, Handl @Override public void processConnect(final Host host, final StartupCommand cmd, final boolean forRebalance) { - if (cmd instanceof StartupRoutingCommand) { - if (((StartupRoutingCommand)cmd).getHypervisorType() == HypervisorType.KVM || ((StartupRoutingCommand)cmd).getHypervisorType() == HypervisorType.LXC) { - Map params = new HashMap(); - params.put("router.aggregation.command.each.timeout", _configDao.getValue("router.aggregation.command.each.timeout")); + if (cmd instanceof StartupRoutingCommand) { + if (((StartupRoutingCommand)cmd).getHypervisorType() == HypervisorType.KVM || ((StartupRoutingCommand)cmd).getHypervisorType() == HypervisorType.LXC) { + Map params = new HashMap(); + params.put("router.aggregation.command.each.timeout", _configDao.getValue("router.aggregation.command.each.timeout")); - try { - SetHostParamsCommand cmds = new SetHostParamsCommand(params); - Commands c = new Commands(cmds); - send(host.getId(), c, this); - } catch (AgentUnavailableException e) { - s_logger.debug("Failed to send host params on host: " + host.getId()); + try { + SetHostParamsCommand cmds = new SetHostParamsCommand(params); + Commands c = new Commands(cmds); + send(host.getId(), c, this); + } catch (AgentUnavailableException e) { + s_logger.debug("Failed to send host params on host: " + host.getId()); + } } } - } } diff --git a/engine/orchestration/src/main/java/com/cloud/agent/manager/ClusteredAgentManagerImpl.java b/engine/orchestration/src/main/java/com/cloud/agent/manager/ClusteredAgentManagerImpl.java index 7a9678e6a4a..38822cba16b 100644 --- a/engine/orchestration/src/main/java/com/cloud/agent/manager/ClusteredAgentManagerImpl.java +++ b/engine/orchestration/src/main/java/com/cloud/agent/manager/ClusteredAgentManagerImpl.java @@ -46,12 +46,12 @@ import org.apache.cloudstack.ca.CAManager; import org.apache.cloudstack.framework.config.ConfigDepot; import org.apache.cloudstack.framework.config.ConfigKey; import org.apache.cloudstack.framework.config.dao.ConfigurationDao; +import org.apache.cloudstack.ha.dao.HAConfigDao; import org.apache.cloudstack.managed.context.ManagedContextRunnable; import org.apache.cloudstack.managed.context.ManagedContextTimerTask; +import org.apache.cloudstack.outofbandmanagement.dao.OutOfBandManagementDao; import org.apache.cloudstack.utils.identity.ManagementServerNode; import org.apache.cloudstack.utils.security.SSLUtils; -import org.apache.cloudstack.ha.dao.HAConfigDao; -import org.apache.cloudstack.outofbandmanagement.dao.OutOfBandManagementDao; import org.apache.log4j.Logger; import com.cloud.agent.api.Answer; @@ -134,14 +134,12 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust super(); } - protected final ConfigKey EnableLB = new ConfigKey(Boolean.class, "agent.lb.enabled", "Advanced", "false", - "Enable agent load balancing between management server nodes", true); + protected final ConfigKey EnableLB = new ConfigKey(Boolean.class, "agent.lb.enabled", "Advanced", "false", "Enable agent load balancing between management server nodes", true); protected final ConfigKey ConnectedAgentThreshold = new ConfigKey(Double.class, "agent.load.threshold", "Advanced", "0.7", "What percentage of the agents can be held by one management server before load balancing happens", true); - protected final ConfigKey LoadSize = new ConfigKey(Integer.class, "direct.agent.load.size", "Advanced", "16", - "How many agents to connect to in each round", true); - protected final ConfigKey ScanInterval = new ConfigKey(Integer.class, "direct.agent.scan.interval", "Advanced", "90", - "Interval between scans to load agents", false, ConfigKey.Scope.Global, 1000); + protected final ConfigKey LoadSize = new ConfigKey(Integer.class, "direct.agent.load.size", "Advanced", "16", "How many agents to connect to in each round", true); + protected final ConfigKey ScanInterval = new ConfigKey(Integer.class, "direct.agent.scan.interval", "Advanced", "90", "Interval between scans to load agents", false, + ConfigKey.Scope.Global, 1000); @Override public boolean configure(final String name, final Map xmlParams) throws ConfigurationException { @@ -342,8 +340,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust final HostTransferMapVO transferVO = _hostTransferDao.findById(hostId); if (transferVO != null) { if (transferVO.getFutureOwner() == _nodeId && transferVO.getState() == HostTransferState.TransferStarted) { - s_logger.debug("Not processing " + Event.AgentDisconnected + " event for the host id=" + hostId + " as the host is being connected to " + - _nodeId); + s_logger.debug("Not processing " + Event.AgentDisconnected + " event for the host id=" + hostId + " as the host is being connected to " + _nodeId); return true; } } @@ -352,8 +349,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust // don't process disconnect if the disconnect came for the host via delayed cluster notification, // but the host has already reconnected to the current management server if (!attache.forForward()) { - s_logger.debug("Not processing " + Event.AgentDisconnected + " event for the host id=" + hostId + - " as the host is directly connected to the current management server " + _nodeId); + s_logger.debug("Not processing " + Event.AgentDisconnected + " event for the host id=" + hostId + " as the host is directly connected to the current management server " + _nodeId); return true; } @@ -367,19 +363,15 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } @Override - public boolean reconnect(final long hostId) { - Boolean result; - try { - result = propagateAgentEvent(hostId, Event.ShutdownRequested); - if (result != null) { - return result; - } - } catch (final AgentUnavailableException e) { - s_logger.debug("cannot propagate agent reconnect because agent is not available", e); - return false; - } + public void reconnect(final long hostId) throws CloudRuntimeException, AgentUnavailableException { + Boolean result = propagateAgentEvent(hostId, Event.ShutdownRequested); + if (result == null) { + super.reconnect(hostId); - return super.reconnect(hostId); + } + if (!result) { + throw new CloudRuntimeException("Failed to propagating agent change request event:" + Event.ShutdownRequested + " to host:" + hostId); + } } public void notifyNodesInCluster(final AgentAttache attache) { @@ -398,18 +390,18 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } protected static void logT(final byte[] bytes, final String msg) { - s_logger.trace("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + - (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); + s_logger.trace("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + + (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); } protected static void logD(final byte[] bytes, final String msg) { - s_logger.debug("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + - (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); + s_logger.debug("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + + (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); } protected static void logI(final byte[] bytes, final String msg) { - s_logger.info("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + - (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); + s_logger.info("Seq " + Request.getAgentId(bytes) + "-" + Request.getSequence(bytes) + ": MgmtId " + Request.getManagementServerId(bytes) + ": " + + (Request.isRequest(bytes) ? "Req: " : "Resp: ") + msg); } public boolean routeToPeer(final String peer, final byte[] bytes) { @@ -443,8 +435,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust logI(bytes, "Unable to route to peer: " + Request.parse(bytes).toString() + " due to " + e.getMessage()); } catch (ClassNotFoundException | UnsupportedVersionException ex) { // Request.parse thrown exception when we try to log it, log as much as we can - logI(bytes, "Unable to route to peer due to" + e.getMessage() - + ". Also caught exception when parsing request: " + ex.getMessage()); + logI(bytes, "Unable to route to peer due to" + e.getMessage() + ". Also caught exception when parsing request: " + ex.getMessage()); } } } @@ -489,8 +480,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust try { prevCh.close(); } catch (final Exception e) { - s_logger.info("[ignored]" - + "failed to get close resource for previous channel Socket: " + e.getLocalizedMessage()); + s_logger.info("[ignored]" + "failed to get close resource for previous channel Socket: " + e.getLocalizedMessage()); } } if (ch == null || ch == prevCh) { @@ -597,7 +587,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust s_logger.info("Closing: " + ch.toString()); ch.close(); } catch (final IOException e) { - s_logger.info("[ignored] error on closing channel: " +ch.toString(), e); + s_logger.info("[ignored] error on closing channel: " + ch.toString(), e); } } } @@ -770,8 +760,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } @Override - public boolean executeRebalanceRequest(final long agentId, final long currentOwnerId, final long futureOwnerId, final Event event) throws AgentUnavailableException, - OperationTimedoutException { + public boolean executeRebalanceRequest(final long agentId, final long currentOwnerId, final long futureOwnerId, final Event event) throws AgentUnavailableException, OperationTimedoutException { boolean result = false; if (event == Event.RequestAgentRebalance) { return setToWaitForRebalance(agentId, currentOwnerId, futureOwnerId); @@ -837,8 +826,8 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust avLoad = allManagedAgents.size() / allMS.size(); } else { if (s_logger.isDebugEnabled()) { - s_logger.debug("There are no hosts to rebalance in the system. Current number of active management server nodes in the system is " + allMS.size() + - "; number of managed agents is " + allManagedAgents.size()); + s_logger.debug("There are no hosts to rebalance in the system. Current number of active management server nodes in the system is " + allMS.size() + "; number of managed agents is " + + allManagedAgents.size()); } return; } @@ -991,8 +980,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust // remove the host from re-balance list and delete from op_host_transfer DB // no need to do anything with the real attache as we haven't modified it yet final Date cutTime = DateUtil.currentGMTTime(); - final HostTransferMapVO transferMap = - _hostTransferDao.findActiveHostTransferMapByHostId(hostId, new Date(cutTime.getTime() - rebalanceTimeOut)); + final HostTransferMapVO transferMap = _hostTransferDao.findActiveHostTransferMapByHostId(hostId, new Date(cutTime.getTime() - rebalanceTimeOut)); if (transferMap == null) { s_logger.debug("Timed out waiting for the host id=" + hostId + " to be ready to transfer, skipping rebalance for the host"); @@ -1010,8 +998,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust final ManagementServerHostVO ms = _mshostDao.findByMsid(transferMap.getFutureOwner()); if (ms != null && ms.getState() != ManagementServerHost.State.Up) { - s_logger.debug("Can't transfer host " + hostId + " as it's future owner is not in UP state: " + ms + - ", skipping rebalance for the host"); + s_logger.debug("Can't transfer host " + hostId + " as it's future owner is not in UP state: " + ms + ", skipping rebalance for the host"); iterator.remove(); _hostTransferDao.completeAgentTransfer(hostId); continue; @@ -1027,8 +1014,8 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } } else { - s_logger.debug("Agent " + hostId + " can't be transfered yet as its request queue size is " + attache.getQueueSize() + - " and listener queue size is " + attache.getNonRecurringListenersSize()); + s_logger.debug("Agent " + hostId + " can't be transfered yet as its request queue size is " + attache.getQueueSize() + " and listener queue size is " + + attache.getNonRecurringListenersSize()); } } } else { @@ -1094,8 +1081,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust if (result) { if (s_logger.isDebugEnabled()) { - s_logger.debug("Loading directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + - " as a part of rebalance process"); + s_logger.debug("Loading directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + " as a part of rebalance process"); } result = loadDirectlyConnectedHost(host, true); } else { @@ -1103,17 +1089,15 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } } catch (final Exception ex) { - s_logger.warn("Failed to load directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + - " as a part of rebalance process due to:", ex); + s_logger.warn("Failed to load directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + " as a part of rebalance process due to:", + ex); result = false; } if (result) { - s_logger.debug("Successfully loaded directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + - " as a part of rebalance process"); + s_logger.debug("Successfully loaded directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + " as a part of rebalance process"); } else { - s_logger.warn("Failed to load directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + - " as a part of rebalance process"); + s_logger.warn("Failed to load directly connected host " + host.getId() + "(" + host.getName() + ") to the management server " + _nodeId + " as a part of rebalance process"); } } @@ -1144,8 +1128,7 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust // 2) Get all transfer requests and route them to peer Request requestToTransfer = forwardAttache.getRequestToTransfer(); while (requestToTransfer != null) { - s_logger.debug("Forwarding request " + requestToTransfer.getSequence() + " held in transfer attache " + hostId + " from the management server " + - _nodeId + " to " + futureOwnerId); + s_logger.debug("Forwarding request " + requestToTransfer.getSequence() + " held in transfer attache " + hostId + " from the management server " + _nodeId + " to " + futureOwnerId); final boolean routeResult = routeToPeer(Long.toString(futureOwnerId), requestToTransfer.getBytes()); if (!routeResult) { logD(requestToTransfer.getBytes(), "Failed to route request to peer"); @@ -1198,8 +1181,8 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust if (attache == null) { s_logger.warn("Attache for the agent " + hostId + " no longer exists on management server " + _nodeId + ", can't start host rebalancing"); } else { - s_logger.warn("Attache for the agent " + hostId + " has request queue size= " + attache.getQueueSize() + " and listener queue size " + - attache.getNonRecurringListenersSize() + ", can't start host rebalancing"); + s_logger.warn("Attache for the agent " + hostId + " has request queue size= " + attache.getQueueSize() + " and listener queue size " + attache.getNonRecurringListenersSize() + + ", can't start host rebalancing"); } return false; } @@ -1255,8 +1238,9 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust } catch (final Exception e) { // Scheduling host scan task in peer MS is a best effort operation during host add, regular host scan // happens at fixed intervals anyways. So handling any exceptions that may be thrown - s_logger.warn("Exception happened while trying to schedule host scan task on mgmt server " + _clusterMgr.getSelfPeerName() + - ", ignoring as regular host scan happens at fixed interval anyways", e); + s_logger.warn( + "Exception happened while trying to schedule host scan task on mgmt server " + _clusterMgr.getSelfPeerName() + ", ignoring as regular host scan happens at fixed interval anyways", + e); return null; } @@ -1372,15 +1356,15 @@ public class ClusteredAgentManagerImpl extends AgentManagerImpl implements Clust final String jsonReturn = _gson.toJson(answers); if (s_logger.isDebugEnabled()) { - s_logger.debug("Completed dispatching -> " + pdu.getAgentId() + ", json: " + pdu.getJsonPackage() + " in " + - (System.currentTimeMillis() - startTick) + " ms, return result: " + jsonReturn); + s_logger.debug("Completed dispatching -> " + pdu.getAgentId() + ", json: " + pdu.getJsonPackage() + " in " + (System.currentTimeMillis() - startTick) + " ms, return result: " + + jsonReturn); } return jsonReturn; } else { if (s_logger.isDebugEnabled()) { - s_logger.debug("Completed dispatching -> " + pdu.getAgentId() + ", json: " + pdu.getJsonPackage() + " in " + - (System.currentTimeMillis() - startTick) + " ms, return null result"); + s_logger.debug( + "Completed dispatching -> " + pdu.getAgentId() + ", json: " + pdu.getJsonPackage() + " in " + (System.currentTimeMillis() - startTick) + " ms, return null result"); } } } catch (final AgentUnavailableException e) { diff --git a/engine/schema/src/main/java/com/cloud/alert/AlertVO.java b/engine/schema/src/main/java/com/cloud/alert/AlertVO.java index 70ef469f52b..1f2cd9d8c60 100644 --- a/engine/schema/src/main/java/com/cloud/alert/AlertVO.java +++ b/engine/schema/src/main/java/com/cloud/alert/AlertVO.java @@ -53,6 +53,9 @@ public class AlertVO implements Alert { @Column(name = "subject", length = 999) private String subject; + @Column(name = "content", length = 5000) + private String content; + @Column(name = "sent_count") private int sentCount = 0; @@ -191,4 +194,13 @@ public class AlertVO implements Alert { public void setName(String name) { this.name = name; } + + @Override + public String getContent() { + return content; + } + + public void setContent(String content) { + this.content = content; + } } diff --git a/engine/schema/src/main/java/com/cloud/host/dao/HostDaoImpl.java b/engine/schema/src/main/java/com/cloud/host/dao/HostDaoImpl.java index e48bfd58a6a..5af099d4e32 100644 --- a/engine/schema/src/main/java/com/cloud/host/dao/HostDaoImpl.java +++ b/engine/schema/src/main/java/com/cloud/host/dao/HostDaoImpl.java @@ -80,7 +80,6 @@ public class HostDaoImpl extends GenericDaoBase implements HostDao private static final String LIST_CLUSTERID_FOR_HOST_TAG = "select distinct cluster_id from host join host_tags on host.id = host_tags.host_id and host_tags.tag = ?"; - protected SearchBuilder TypePodDcStatusSearch; protected SearchBuilder IdStatusSearch; diff --git a/engine/schema/src/main/resources/META-INF/db/schema-41000to41100-cleanup.sql b/engine/schema/src/main/resources/META-INF/db/schema-41000to41100-cleanup.sql index d7a080efd16..f8d9ce9b73b 100644 --- a/engine/schema/src/main/resources/META-INF/db/schema-41000to41100-cleanup.sql +++ b/engine/schema/src/main/resources/META-INF/db/schema-41000to41100-cleanup.sql @@ -66,4 +66,4 @@ CREATE VIEW `cloud`.`user_view` AS left join `cloud`.`async_job` ON async_job.instance_id = user.id and async_job.instance_type = 'User' - and async_job.job_status = 0; + and async_job.job_status = 0; \ No newline at end of file diff --git a/engine/schema/src/main/resources/META-INF/db/schema-41000to41100.sql b/engine/schema/src/main/resources/META-INF/db/schema-41000to41100.sql index 2e7f9e9fe9c..70a76d98d86 100644 --- a/engine/schema/src/main/resources/META-INF/db/schema-41000to41100.sql +++ b/engine/schema/src/main/resources/META-INF/db/schema-41000to41100.sql @@ -566,4 +566,4 @@ CREATE TABLE IF NOT EXISTS `cloud`.`external_netscaler_controlcenter` ( ) ENGINE=InnoDB DEFAULT CHARSET=utf8; ALTER TABLE `cloud`.`sslcerts` ADD COLUMN `name` varchar(255) NULL default NULL COMMENT 'Name of the Certificate'; -ALTER TABLE `cloud`.`network_offerings` ADD COLUMN `service_package_id` varchar(255) NULL default NULL COMMENT 'Netscaler ControlCenter Service Package'; +ALTER TABLE `cloud`.`network_offerings` ADD COLUMN `service_package_id` varchar(255) NULL default NULL COMMENT 'Netscaler ControlCenter Service Package'; \ No newline at end of file diff --git a/engine/schema/src/main/resources/META-INF/db/schema-41100to41200.sql b/engine/schema/src/main/resources/META-INF/db/schema-41100to41200.sql index 5667c5b3dab..97bd5ae7aae 100644 --- a/engine/schema/src/main/resources/META-INF/db/schema-41100to41200.sql +++ b/engine/schema/src/main/resources/META-INF/db/schema-41100to41200.sql @@ -20,4 +20,7 @@ --; -- [CLOUDSTACK-10314] Add reason column to ACL rule table -ALTER TABLE `cloud`.`network_acl_item` ADD COLUMN `reason` VARCHAR(2500) AFTER `display`; \ No newline at end of file +ALTER TABLE `cloud`.`network_acl_item` ADD COLUMN `reason` VARCHAR(2500) AFTER `display`; + +--[CLOUDSTACK-9846] Make provision to store content and subject for Alerts in separate columns. +ALTER TABLE `cloud`.`alert` ADD COLUMN `content` VARCHAR(5000); diff --git a/engine/storage/integration-test/src/test/java/org/apache/cloudstack/storage/test/DirectAgentManagerSimpleImpl.java b/engine/storage/integration-test/src/test/java/org/apache/cloudstack/storage/test/DirectAgentManagerSimpleImpl.java index 30cba4d6c71..c6003afee57 100644 --- a/engine/storage/integration-test/src/test/java/org/apache/cloudstack/storage/test/DirectAgentManagerSimpleImpl.java +++ b/engine/storage/integration-test/src/test/java/org/apache/cloudstack/storage/test/DirectAgentManagerSimpleImpl.java @@ -266,20 +266,16 @@ public class DirectAgentManagerSimpleImpl extends ManagerBase implements AgentMa } @Override - public boolean reconnect(long hostId) { - // TODO Auto-generated method stub - return false; + public void reconnect(long hostId) { } @Override public boolean isAgentAttached(long hostId) { - // TODO Auto-generated method stub return false; } @Override public boolean handleDirectConnectAgent(Host host, StartupCommand[] cmds, ServerResource resource, boolean forRebalance, boolean newHost) throws ConnectionException { - // TODO Auto-generated method stub return false; } diff --git a/plugins/network-elements/netscaler/src/main/java/com/cloud/network/element/NetscalerElement.java b/plugins/network-elements/netscaler/src/main/java/com/cloud/network/element/NetscalerElement.java index 38a836d2d6d..1df16404d39 100644 --- a/plugins/network-elements/netscaler/src/main/java/com/cloud/network/element/NetscalerElement.java +++ b/plugins/network-elements/netscaler/src/main/java/com/cloud/network/element/NetscalerElement.java @@ -28,12 +28,6 @@ import java.util.UUID; import javax.inject.Inject; import javax.naming.ConfigurationException; -import org.apache.log4j.Logger; -import org.json.JSONException; -import org.json.JSONObject; - -import com.google.gson.Gson; - import org.apache.cloudstack.api.ApiConstants; import org.apache.cloudstack.api.ApiErrorCode; import org.apache.cloudstack.api.ServerApiException; @@ -41,6 +35,9 @@ import org.apache.cloudstack.engine.orchestration.service.NetworkOrchestrationSe import org.apache.cloudstack.framework.config.dao.ConfigurationDao; import org.apache.cloudstack.network.ExternalNetworkDeviceManager.NetworkDevice; import org.apache.cloudstack.region.gslb.GslbServiceProvider; +import org.apache.log4j.Logger; +import org.json.JSONException; +import org.json.JSONObject; import com.cloud.agent.AgentManager; import com.cloud.agent.api.Answer; @@ -84,6 +81,7 @@ import com.cloud.deploy.DeployDestination; import com.cloud.deploy.DeploymentPlan; import com.cloud.event.ActionEvent; import com.cloud.event.EventTypes; +import com.cloud.exception.AgentUnavailableException; import com.cloud.exception.ConcurrentOperationException; import com.cloud.exception.InsufficientCapacityException; import com.cloud.exception.InsufficientNetworkCapacityException; @@ -154,10 +152,11 @@ import com.cloud.utils.net.UrlUtil; import com.cloud.vm.NicProfile; import com.cloud.vm.ReservationContext; import com.cloud.vm.VirtualMachineProfile; +import com.google.gson.Gson; public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl - implements LoadBalancingServiceProvider, NetscalerLoadBalancerElementService, ExternalLoadBalancerDeviceManager, - IpDeployer, StaticNatServiceProvider, GslbServiceProvider { +implements LoadBalancingServiceProvider, NetscalerLoadBalancerElementService, ExternalLoadBalancerDeviceManager, +IpDeployer, StaticNatServiceProvider, GslbServiceProvider { private static final Logger s_logger = Logger.getLogger(NetscalerElement.class); public static final AutoScaleCounterType AutoScaleCounterSnmp = new AutoScaleCounterType("snmp"); @@ -218,7 +217,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl boolean handleInAdvanceZone = (zone.getNetworkType() == NetworkType.Advanced && (config.getGuestType() == Network.GuestType.Isolated - || config.getGuestType() == Network.GuestType.Shared) + || config.getGuestType() == Network.GuestType.Shared) && config.getTrafficType() == TrafficType.Guest); boolean handleInBasicZone = (zone.getNetworkType() == NetworkType.Basic && config.getGuestType() == Network.GuestType.Shared && config.getTrafficType() == TrafficType.Guest); @@ -242,7 +241,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl @Override public boolean implement(Network guestConfig, NetworkOffering offering, DeployDestination dest, ReservationContext context) throws ResourceUnavailableException, ConcurrentOperationException, - InsufficientNetworkCapacityException { + InsufficientNetworkCapacityException { if (!canHandle(guestConfig, Service.Lb)) { return false; @@ -271,7 +270,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl throw new ResourceUnavailableException( "There are no NetScaler load balancer devices with the free capacity for implementing this network : " + e.getMessage(), - DataCenter.class, guestConfig.getDataCenterId()); + DataCenter.class, guestConfig.getDataCenterId()); } } @@ -286,7 +285,6 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl Map _configs; List ncc = _netscalerControlCenterDao.listAll(); HostVO hostVO = null; - Map params; if (ncc.size() > 0) { NetScalerControlCenterVO nccVO = ncc.get(0); String ipAddress = nccVO.getNccip(); @@ -414,7 +412,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl @Override public boolean prepare(Network config, NicProfile nic, VirtualMachineProfile vm, DeployDestination dest, ReservationContext context) throws ConcurrentOperationException, InsufficientNetworkCapacityException, - ResourceUnavailableException { + ResourceUnavailableException { return true; } @@ -593,45 +591,45 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl boolean dedicatedUse = (configParams.get(ApiConstants.LOAD_BALANCER_DEVICE_DEDICATED) != null) ? Boolean.parseBoolean(configParams.get(ApiConstants.LOAD_BALANCER_DEVICE_DEDICATED)) : false; - if (dedicatedUse && !deviceName.equals(NetworkDevice.NetscalerVPXLoadBalancer.getName())) { - String msg = "Only Netscaler VPX load balancers can be specified for dedicated use"; - s_logger.debug(msg); - throw new InvalidParameterValueException(msg); - } + if (dedicatedUse && !deviceName.equals(NetworkDevice.NetscalerVPXLoadBalancer.getName())) { + String msg = "Only Netscaler VPX load balancers can be specified for dedicated use"; + s_logger.debug(msg); + throw new InvalidParameterValueException(msg); + } - if (cmd.isGslbProvider()) { + if (cmd.isGslbProvider()) { - if (!deviceName.equals(NetworkDevice.NetscalerVPXLoadBalancer.getName()) - && !deviceName.equals(NetworkDevice.NetscalerMPXLoadBalancer.getName())) { - String msg = "Only Netscaler VPX or MPX load balancers can be specified as GSLB service provider"; - s_logger.debug(msg); - throw new InvalidParameterValueException(msg); - } + if (!deviceName.equals(NetworkDevice.NetscalerVPXLoadBalancer.getName()) + && !deviceName.equals(NetworkDevice.NetscalerMPXLoadBalancer.getName())) { + String msg = "Only Netscaler VPX or MPX load balancers can be specified as GSLB service provider"; + s_logger.debug(msg); + throw new InvalidParameterValueException(msg); + } - if (cmd.getSitePublicIp() == null || cmd.getSitePrivateIp() == null) { - String msg = "Public and Privae IP needs to provided for NetScaler that will be GSLB provider"; - s_logger.debug(msg); - throw new InvalidParameterValueException(msg); - } + if (cmd.getSitePublicIp() == null || cmd.getSitePrivateIp() == null) { + String msg = "Public and Privae IP needs to provided for NetScaler that will be GSLB provider"; + s_logger.debug(msg); + throw new InvalidParameterValueException(msg); + } - if (dedicatedUse) { - throw new InvalidParameterValueException( - "NetScaler provisioned to be GSLB service provider can only be configured for shared usage."); - } + if (dedicatedUse) { + throw new InvalidParameterValueException( + "NetScaler provisioned to be GSLB service provider can only be configured for shared usage."); + } - } + } - if (cmd.isExclusiveGslbProvider() && !cmd.isGslbProvider()) { - throw new InvalidParameterValueException( - "NetScaler can be provisioned to be exclusive GSLB service provider" - + " only if its being configured as GSLB service provider also."); - } + if (cmd.isExclusiveGslbProvider() && !cmd.isGslbProvider()) { + throw new InvalidParameterValueException( + "NetScaler can be provisioned to be exclusive GSLB service provider" + + " only if its being configured as GSLB service provider also."); + } - ExternalLoadBalancerDeviceVO lbDeviceVO = addExternalLoadBalancer(cmd.getPhysicalNetworkId(), cmd.getUrl(), - cmd.getUsername(), cmd.getPassword(), deviceName, new NetscalerResource(), cmd.isGslbProvider(), - cmd.isExclusiveGslbProvider(), cmd.getSitePublicIp(), cmd.getSitePrivateIp()); + ExternalLoadBalancerDeviceVO lbDeviceVO = addExternalLoadBalancer(cmd.getPhysicalNetworkId(), cmd.getUrl(), + cmd.getUsername(), cmd.getPassword(), deviceName, new NetscalerResource(), cmd.isGslbProvider(), + cmd.isExclusiveGslbProvider(), cmd.getSitePublicIp(), cmd.getSitePrivateIp()); - return lbDeviceVO; + return lbDeviceVO; } @Override @@ -759,7 +757,11 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl }); HostVO host = _hostDao.findById(lbDeviceVo.getHostId()); - _agentMgr.reconnect(host.getId()); + try { + _agentMgr.reconnect(host.getId()); + } catch (AgentUnavailableException e) { + s_logger.warn("failed to reconnect host " + host, e); + } return lbDeviceVo; } @@ -858,8 +860,9 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl boolean flag=false; try { result = _netscalerServicePackageDao.findByUuid(cmd.getId()); - if (result == null) + if (result == null) { throw new CloudRuntimeException("Record does not Exists in the Table"); + } if(_networkOfferingDao.isUsingServicePackage(result.getUuid())) { @@ -869,10 +872,11 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl flag = _netscalerServicePackageDao.remove(result.getId()); } catch (Exception e) { - if (e instanceof InvalidParameterValueException) + if (e instanceof InvalidParameterValueException) { throw new ServerApiException(ApiErrorCode.PARAM_ERROR, e.getMessage()); - else - throw e; + } else { + throw e; + } } return flag; @@ -884,18 +888,19 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl public boolean deleteNetscalerControlCenter(DeleteNetscalerControlCenterCmd cmd) throws CloudRuntimeException { NetScalerControlCenterVO result = _netscalerControlCenterDao.findByUuid(cmd.getId()); - if (result == null) + if (result == null) { throw new CloudRuntimeException("External Netscaler Control Center Table does not contain record with this ID"); - else { + } else { //ID list of Network Offering which are not removed and have service Package Uuid field not null. List servicePackageId_list = _networkOfferingDao.listNetworkOfferingID(); if (servicePackageId_list.size() != 0) { //VO list of Networks which are using Network Offering. List networkVO_list = _networkDao.listNetworkVO(servicePackageId_list); - if (networkVO_list != null && networkVO_list.size() != 0) + if (networkVO_list != null && networkVO_list.size() != 0) { throw new CloudRuntimeException( "ServicePackages published by NetScalerControlCenter are being used by NetworkOfferings. Try deleting NetworkOffering with ServicePackages and then delete NetScalerControlCenter."); + } } } try { @@ -1466,8 +1471,9 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl @DB public NetScalerControlCenterVO registerNetscalerControlCenter(RegisterNetscalerControlCenterCmd cmd) { - if (_netscalerControlCenterDao.listAll() != null && _netscalerControlCenterDao.listAll().size() != 0) + if (_netscalerControlCenterDao.listAll() != null && _netscalerControlCenterDao.listAll().size() != 0) { throw new CloudRuntimeException("One Netscaler Control Center already exist in the DataBase. At a time only one Netscaler Control Center is allowed"); + } final RegisterNetscalerControlCenterCmd cmdinfo = cmd; String ipAddress = cmd.getIpaddress(); @@ -1511,7 +1517,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl Long serviceOfferingId = cmd.getServiceOfferingId(); DeploymentPlan plan = new DataCenterDeployment(dest.getDataCenter().getId()); try { - resp = _netScalerVMManager.deployNsVpx(cmd.getAccount(), dest, plan, serviceOfferingId, templateId); + resp = _netScalerVMManager.deployNsVpx(cmd.getAccount(), dest, plan, serviceOfferingId, templateId); } catch (InsufficientCapacityException e) { e.printStackTrace(); } @@ -1520,7 +1526,7 @@ public class NetscalerElement extends ExternalLoadBalancerDeviceManagerImpl @Override public VirtualRouter stopNetscalerServiceVm(Long id, boolean forced, Account callingAccount, long callingUserId) throws ConcurrentOperationException, - ResourceUnavailableException { + ResourceUnavailableException { return _netScalerVMManager.stopNetScalerVm(id, forced, callingAccount, callingUserId); } } \ No newline at end of file diff --git a/requirements.txt b/requirements.txt index 5f2551e9f4f..4485f5d73f9 100644 --- a/requirements.txt +++ b/requirements.txt @@ -21,5 +21,4 @@ cloudmonkey # Marvin dependencies are installed via its bundle # Install the SolidFire SDK for Python -solidfire-sdk-python - +solidfire-sdk-python \ No newline at end of file diff --git a/server/src/main/java/com/cloud/alert/AlertManagerImpl.java b/server/src/main/java/com/cloud/alert/AlertManagerImpl.java index a58a4f8b0f9..5a7a8b436e0 100644 --- a/server/src/main/java/com/cloud/alert/AlertManagerImpl.java +++ b/server/src/main/java/com/cloud/alert/AlertManagerImpl.java @@ -39,12 +39,6 @@ import javax.mail.URLName; import javax.mail.internet.InternetAddress; import javax.naming.ConfigurationException; -import org.apache.log4j.Logger; - -import com.sun.mail.smtp.SMTPMessage; -import com.sun.mail.smtp.SMTPSSLTransport; -import com.sun.mail.smtp.SMTPTransport; - import org.apache.cloudstack.framework.config.ConfigDepot; import org.apache.cloudstack.framework.config.ConfigKey; import org.apache.cloudstack.framework.config.Configurable; @@ -52,6 +46,7 @@ import org.apache.cloudstack.framework.config.dao.ConfigurationDao; import org.apache.cloudstack.managed.context.ManagedContextTimerTask; import org.apache.cloudstack.storage.datastore.db.PrimaryDataStoreDao; import org.apache.cloudstack.storage.datastore.db.StoragePoolVO; +import org.apache.log4j.Logger; import com.cloud.alert.dao.AlertDao; import com.cloud.api.ApiDBUtils; @@ -85,10 +80,12 @@ import com.cloud.utils.NumbersUtil; import com.cloud.utils.component.ManagerBase; import com.cloud.utils.concurrency.NamedThreadFactory; import com.cloud.utils.db.SearchCriteria; +import com.sun.mail.smtp.SMTPMessage; +import com.sun.mail.smtp.SMTPSSLTransport; +import com.sun.mail.smtp.SMTPTransport; public class AlertManagerImpl extends ManagerBase implements AlertManager, Configurable { private static final Logger s_logger = Logger.getLogger(AlertManagerImpl.class.getName()); - private static final Logger s_alertsLogger = Logger.getLogger("org.apache.cloudstack.alerts"); private static final long INITIAL_CAPACITY_CHECK_DELAY = 30L * 1000L; // Thirty seconds expressed in milliseconds. @@ -205,8 +202,9 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi String capacityCheckPeriodStr = configs.get("capacity.check.period"); if (capacityCheckPeriodStr != null) { _capacityCheckPeriod = Long.parseLong(capacityCheckPeriodStr); - if (_capacityCheckPeriod <= 0) + if (_capacityCheckPeriod <= 0) { _capacityCheckPeriod = Long.parseLong(Config.CapacityCheckPeriod.getDefaultValue()); + } } _timer = new Timer("CapacityChecker"); @@ -249,7 +247,7 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi if (_emailAlert != null) { _emailAlert.sendAlert(alertType, dataCenterId, podId, null, subject, body); } else { - s_alertsLogger.warn("AlertType:: " + alertType + " | dataCenterId:: " + dataCenterId + " | podId:: " + podId + + s_logger.warn("AlertType:: " + alertType + " | dataCenterId:: " + dataCenterId + " | podId:: " + podId + " | message:: " + subject + " | body:: " + body); } } catch (Exception ex) { @@ -589,7 +587,7 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi alertType = AlertManager.AlertType.ALERT_TYPE_LOCAL_STORAGE; break; - //Pod Level + //Pod Level case Capacity.CAPACITY_TYPE_PRIVATE_IP: msgSubject = "System Alert: Number of unallocated private IPs is low in pod " + pod.getName() + " of availability zone " + dc.getName(); totalStr = Double.toString(totalCapacity); @@ -598,7 +596,7 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi alertType = AlertManager.AlertType.ALERT_TYPE_PRIVATE_IP; break; - //Zone Level + //Zone Level case Capacity.CAPACITY_TYPE_SECONDARY_STORAGE: msgSubject = "System Alert: Low Available Secondary Storage in availability zone " + dc.getName(); totalStr = formatBytesToMegabytes(totalCapacity); @@ -746,22 +744,22 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi // TODO: make sure this handles SSL transport (useAuth is true) and regular public void sendAlert(AlertType alertType, long dataCenterId, Long podId, Long clusterId, String subject, String content) throws MessagingException, - UnsupportedEncodingException { - s_alertsLogger.warn("AlertType:: " + alertType + " | dataCenterId:: " + dataCenterId + " | podId:: " + + UnsupportedEncodingException { + s_logger.warn("AlertType:: " + alertType + " | dataCenterId:: " + dataCenterId + " | podId:: " + podId + " | clusterId:: " + clusterId + " | message:: " + subject); AlertVO alert = null; if ((alertType != AlertManager.AlertType.ALERT_TYPE_HOST) && - (alertType != AlertManager.AlertType.ALERT_TYPE_USERVM) && - (alertType != AlertManager.AlertType.ALERT_TYPE_DOMAIN_ROUTER) && - (alertType != AlertManager.AlertType.ALERT_TYPE_CONSOLE_PROXY) && - (alertType != AlertManager.AlertType.ALERT_TYPE_SSVM) && - (alertType != AlertManager.AlertType.ALERT_TYPE_STORAGE_MISC) && - (alertType != AlertManager.AlertType.ALERT_TYPE_MANAGMENT_NODE) && - (alertType != AlertManager.AlertType.ALERT_TYPE_RESOURCE_LIMIT_EXCEEDED) && - (alertType != AlertManager.AlertType.ALERT_TYPE_UPLOAD_FAILED) && - (alertType != AlertManager.AlertType.ALERT_TYPE_OOBM_AUTH_ERROR) && - (alertType != AlertManager.AlertType.ALERT_TYPE_HA_ACTION) && - (alertType != AlertManager.AlertType.ALERT_TYPE_CA_CERT)) { + (alertType != AlertManager.AlertType.ALERT_TYPE_USERVM) && + (alertType != AlertManager.AlertType.ALERT_TYPE_DOMAIN_ROUTER) && + (alertType != AlertManager.AlertType.ALERT_TYPE_CONSOLE_PROXY) && + (alertType != AlertManager.AlertType.ALERT_TYPE_SSVM) && + (alertType != AlertManager.AlertType.ALERT_TYPE_STORAGE_MISC) && + (alertType != AlertManager.AlertType.ALERT_TYPE_MANAGMENT_NODE) && + (alertType != AlertManager.AlertType.ALERT_TYPE_RESOURCE_LIMIT_EXCEEDED) && + (alertType != AlertManager.AlertType.ALERT_TYPE_UPLOAD_FAILED) && + (alertType != AlertManager.AlertType.ALERT_TYPE_OOBM_AUTH_ERROR) && + (alertType != AlertManager.AlertType.ALERT_TYPE_HA_ACTION) && + (alertType != AlertManager.AlertType.ALERT_TYPE_CA_CERT)) { alert = _alertDao.getLastAlert(alertType.getType(), dataCenterId, podId, clusterId); } @@ -770,6 +768,7 @@ public class AlertManagerImpl extends ManagerBase implements AlertManager, Confi AlertVO newAlert = new AlertVO(); newAlert.setType(alertType.getType()); newAlert.setSubject(subject); + newAlert.setContent(content); newAlert.setClusterId(clusterId); newAlert.setPodId(podId); newAlert.setDataCenterId(dataCenterId); diff --git a/server/src/main/java/com/cloud/resource/ResourceManagerImpl.java b/server/src/main/java/com/cloud/resource/ResourceManagerImpl.java index 2966d41d8bf..195543686cb 100755 --- a/server/src/main/java/com/cloud/resource/ResourceManagerImpl.java +++ b/server/src/main/java/com/cloud/resource/ResourceManagerImpl.java @@ -30,10 +30,6 @@ import java.util.Random; import javax.inject.Inject; import javax.naming.ConfigurationException; -import org.apache.commons.lang.ObjectUtils; -import org.apache.log4j.Logger; -import org.springframework.stereotype.Component; - import org.apache.cloudstack.api.ApiConstants; import org.apache.cloudstack.api.command.admin.cluster.AddClusterCmd; import org.apache.cloudstack.api.command.admin.cluster.DeleteClusterCmd; @@ -50,6 +46,9 @@ import org.apache.cloudstack.storage.datastore.db.PrimaryDataStoreDao; import org.apache.cloudstack.storage.datastore.db.StoragePoolVO; import org.apache.cloudstack.utils.identity.ManagementServerNode; import org.apache.commons.collections.CollectionUtils; +import org.apache.commons.lang.ObjectUtils; +import org.apache.log4j.Logger; +import org.springframework.stereotype.Component; import com.cloud.agent.AgentManager; import com.cloud.agent.api.Answer; @@ -75,7 +74,6 @@ import com.cloud.capacity.CapacityVO; import com.cloud.capacity.dao.CapacityDao; import com.cloud.cluster.ClusterManager; import com.cloud.configuration.Config; -import com.cloud.configuration.ConfigurationManager; import com.cloud.dc.ClusterDetailsDao; import com.cloud.dc.ClusterDetailsVO; import com.cloud.dc.ClusterVO; @@ -249,8 +247,6 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, @Inject private VMTemplateDao _templateDao; @Inject - private ConfigurationManager _configMgr; - @Inject private ClusterVSMMapDao _clusterVSMMapDao; private final long _nodeId = ManagementServerNode.getManagementServerId(); @@ -610,7 +606,7 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, private List discoverHostsFull(final Long dcId, final Long podId, Long clusterId, final String clusterName, String url, String username, String password, final String hypervisorType, final List hostTags, final Map params, final boolean deferAgentCreation) throws IllegalArgumentException, DiscoveryException, - InvalidParameterValueException { + InvalidParameterValueException { URI uri = null; // Check if the zone exists in the system @@ -836,7 +832,6 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, } // Get storage pool host mappings here because they can be removed as a // part of handleDisconnect later - // TODO: find out the bad boy, what's a buggy logic! final List pools = _storagePoolHostDao.listByHostIdIncludingRemoved(hostId); final ResourceStateAdapter.DeleteHostAnswer answer = @@ -1166,15 +1161,15 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, } @Override - public Host reconnectHost(final ReconnectHostCmd cmd) { - final Long hostId = cmd.getId(); + public Host reconnectHost(ReconnectHostCmd cmd) throws AgentUnavailableException { + Long hostId = cmd.getId(); - final HostVO host = _hostDao.findById(hostId); + HostVO host = _hostDao.findById(hostId); if (host == null) { throw new InvalidParameterValueException("Host with id " + hostId.toString() + " doesn't exist"); } - - return _agentMgr.reconnect(hostId) ? host : null; + _agentMgr.reconnect(hostId); + return host; } @Override @@ -2310,7 +2305,6 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, } else if (event == ResourceState.Event.AdminCancelMaintenance) { return doCancelMaintenance(hostId); } else if (event == ResourceState.Event.DeleteHost) { - /* TODO: Ask alex why we assume the last two parameters are false */ return doDeleteHost(hostId, false, false); } else if (event == ResourceState.Event.Unmanaged) { return doUmanageHost(hostId); @@ -2329,7 +2323,7 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, } if (host.getHypervisorType() == HypervisorType.KVM || host.getHypervisorType() == HypervisorType.LXC) { - final MaintainAnswer answer = (MaintainAnswer)_agentMgr.easySend(hostId, new MaintainCommand()); + _agentMgr.easySend(hostId, new MaintainCommand()); } _agentMgr.disconnectWithoutInvestigation(hostId, Event.ShutdownRequested); @@ -2379,10 +2373,6 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, final List hosts = listAllHostsInCluster(command.getClusterId()); for (final HostVO host : hosts) { try { - /* - * FIXME: this is a buggy logic, check with alex. Shouldn't - * return if propagation return non null - */ final Boolean result = propagateResourceEvent(host.getId(), ResourceState.Event.UpdatePassword); if (result != null) { return result; @@ -2852,11 +2842,4 @@ public class ResourceManagerImpl extends ManagerBase implements ResourceManager, return false; } } - - @Override - public boolean start() { - // TODO Auto-generated method stub - return super.start(); - } - } diff --git a/server/src/main/java/org/apache/cloudstack/ca/CAManagerImpl.java b/server/src/main/java/org/apache/cloudstack/ca/CAManagerImpl.java index 3a904315fe1..02e7a4231a4 100644 --- a/server/src/main/java/org/apache/cloudstack/ca/CAManagerImpl.java +++ b/server/src/main/java/org/apache/cloudstack/ca/CAManagerImpl.java @@ -191,7 +191,8 @@ public class CAManagerImpl extends ManagerBase implements CAManager { if (Strings.isNullOrEmpty(csr)) { return false; } - final Certificate certificate = issueCertificate(csr, Collections.singletonList(host.getName()), Arrays.asList(host.getPrivateIpAddress(), host.getPublicIpAddress(), host.getStorageIpAddress()), CAManager.CertValidityPeriod.value(), caProvider); + final Certificate certificate = issueCertificate(csr, Collections.singletonList(host.getName()), + Arrays.asList(host.getPrivateIpAddress(), host.getPublicIpAddress(), host.getStorageIpAddress()), CAManager.CertValidityPeriod.value(), caProvider); return deployCertificate(host, certificate, reconnect, null); } catch (final AgentUnavailableException | OperationTimedoutException e) { LOG.error("Host/agent is not available or operation timed out, failed to setup keystore and generate CSR for host/agent id=" + host.getId() + ", due to: ", e); @@ -206,18 +207,19 @@ public class CAManagerImpl extends ManagerBase implements CAManager { cmd.setAccessDetail(sshAccessDetails); } CallContext.current().setEventDetails("generating keystore and CSR for host id: " + host.getId()); - final SetupKeystoreAnswer answer = (SetupKeystoreAnswer) agentManager.send(host.getId(), cmd); + final SetupKeystoreAnswer answer = (SetupKeystoreAnswer)agentManager.send(host.getId(), cmd); return answer.getCsr(); } @Override - public boolean deployCertificate(final Host host, final Certificate certificate, final Boolean reconnect, final Map sshAccessDetails) throws AgentUnavailableException, OperationTimedoutException { + public boolean deployCertificate(final Host host, final Certificate certificate, final Boolean reconnect, final Map sshAccessDetails) + throws AgentUnavailableException, OperationTimedoutException { final SetupCertificateCommand cmd = new SetupCertificateCommand(certificate); if (sshAccessDetails != null && !sshAccessDetails.isEmpty()) { cmd.setAccessDetail(sshAccessDetails); } CallContext.current().setEventDetails("deploying certificate for host id: " + host.getId()); - final SetupCertificateAnswer answer = (SetupCertificateAnswer) agentManager.send(host.getId(), cmd); + final SetupCertificateAnswer answer = (SetupCertificateAnswer)agentManager.send(host.getId(), cmd); if (answer.getResult()) { CallContext.current().setEventDetails("successfully deployed certificate for host id: " + host.getId()); } else { @@ -227,9 +229,12 @@ public class CAManagerImpl extends ManagerBase implements CAManager { if (answer.getResult()) { getActiveCertificatesMap().put(host.getPrivateIpAddress(), certificate.getClientCertificate()); if (sshAccessDetails == null && reconnect != null && reconnect) { - LOG.info(String.format("Successfully setup certificate on host, reconnecting with agent with id=%d, name=%s, address=%s", - host.getId(), host.getName(), host.getPublicIpAddress())); - return agentManager.reconnect(host.getId()); + LOG.info(String.format("Successfully setup certificate on host, reconnecting with agent with id=%d, name=%s, address=%s", host.getId(), host.getName(), host.getPublicIpAddress())); + try { + agentManager.reconnect(host.getId()); + } catch (AgentUnavailableException | CloudRuntimeException e) { + LOG.debug("Error when reconnecting to host: " + host.getUuid(), e); + } } return true; } @@ -257,8 +262,7 @@ public class CAManagerImpl extends ManagerBase implements CAManager { if (host == null) { return; } - alertManager.sendAlert(AlertManager.AlertType.ALERT_TYPE_CA_CERT, - host.getDataCenterId(), host.getPodId(), subject, message); + alertManager.sendAlert(AlertManager.AlertType.ALERT_TYPE_CA_CERT, host.getDataCenterId(), host.getPodId(), subject, message); } @Override @@ -303,7 +307,7 @@ public class CAManagerImpl extends ManagerBase implements CAManager { } final DateTime now = DateTime.now(DateTimeZone.UTC); final Map certsMap = caManager.getActiveCertificatesMap(); - for (final Iterator> it = certsMap.entrySet().iterator(); it.hasNext(); ) { + for (final Iterator> it = certsMap.entrySet().iterator(); it.hasNext();) { final Map.Entry entry = it.next(); if (entry == null) { continue; @@ -315,19 +319,14 @@ public class CAManagerImpl extends ManagerBase implements CAManager { continue; } final Host host = hostDao.findByIp(hostIp); - if (host == null || host.getManagementServerId() == null || - host.getManagementServerId() != ManagementServerNode.getManagementServerId() || - host.getStatus() != Status.Up) { - if (host == null || - (host.getManagementServerId() != null && - host.getManagementServerId() != ManagementServerNode.getManagementServerId())) { + if (host == null || host.getManagementServerId() == null || host.getManagementServerId() != ManagementServerNode.getManagementServerId() || host.getStatus() != Status.Up) { + if (host == null || (host.getManagementServerId() != null && host.getManagementServerId() != ManagementServerNode.getManagementServerId())) { it.remove(); } continue; } - final String hostDescription = String.format("host id=%d, uuid=%s, name=%s, ip=%s, zone id=%d", - host.getId(), host.getUuid(), host.getName(), hostIp, host.getDataCenterId()); + final String hostDescription = String.format("host id=%d, uuid=%s, name=%s, ip=%s, zone id=%d", host.getId(), host.getUuid(), host.getName(), hostIp, host.getDataCenterId()); try { certificate.checkValidity(now.plusDays(CertExpiryAlertPeriod.valueIn(host.getClusterId())).toDate()); @@ -345,7 +344,8 @@ public class CAManagerImpl extends ManagerBase implements CAManager { } catch (final Throwable ex) { LOG.warn("Failed to auto-renew certificate for " + hostDescription + ", with error=", ex); caManager.sendAlert(host, "Certificate auto-renewal failed for " + hostDescription, - String.format("Certificate is going to expire for %s. Auto-renewal failed to renew the certificate, please renew it manually. It is not valid after %s.", hostDescription, certificate.getNotAfter())); + String.format("Certificate is going to expire for %s. Auto-renewal failed to renew the certificate, please renew it manually. It is not valid after %s.", + hostDescription, certificate.getNotAfter())); } } else { if (alertMap.containsKey(hostIp)) { @@ -355,8 +355,7 @@ public class CAManagerImpl extends ManagerBase implements CAManager { } } caManager.sendAlert(host, "Certificate expiring soon for " + hostDescription, - String.format("Certificate is going to expire for %s. Please renew it, it is not valid after %s.", - hostDescription, certificate.getNotAfter())); + String.format("Certificate is going to expire for %s. Please renew it, it is not valid after %s.", hostDescription, certificate.getNotAfter())); alertMap.put(hostIp, new Date()); } } @@ -427,14 +426,6 @@ public class CAManagerImpl extends ManagerBase implements CAManager { @Override public ConfigKey[] getConfigKeys() { - return new ConfigKey[]{ - CAProviderPlugin, - CertKeySize, - CertSignatureAlgorithm, - CertValidityPeriod, - AutomaticCertRenewal, - CABackgroundJobDelay, - CertExpiryAlertPeriod - }; + return new ConfigKey[] {CAProviderPlugin, CertKeySize, CertSignatureAlgorithm, CertValidityPeriod, AutomaticCertRenewal, CABackgroundJobDelay, CertExpiryAlertPeriod}; } } diff --git a/server/src/test/java/org/apache/cloudstack/ca/CAManagerImplTest.java b/server/src/test/java/org/apache/cloudstack/ca/CAManagerImplTest.java index 14ecc971d5f..64cbf205812 100644 --- a/server/src/test/java/org/apache/cloudstack/ca/CAManagerImplTest.java +++ b/server/src/test/java/org/apache/cloudstack/ca/CAManagerImplTest.java @@ -113,7 +113,7 @@ public class CAManagerImplTest { final X509Certificate certificate = CertUtils.generateV3Certificate(null, keyPair, keyPair.getPublic(), "CN=ca", "SHA256withRSA", 365, null, null); Mockito.when(caProvider.issueCertificate(Mockito.anyString(), Mockito.anyList(), Mockito.anyList(), Mockito.anyInt())).thenReturn(new Certificate(certificate, null, Collections.singletonList(certificate))); Mockito.when(agentManager.send(Mockito.anyLong(), Mockito.any(SetupKeyStoreCommand.class))).thenReturn(new SetupKeystoreAnswer("someCsr")); - Mockito.when(agentManager.reconnect(Mockito.anyLong())).thenReturn(true); + Mockito.doNothing().when(agentManager).reconnect(Mockito.anyLong()); Assert.assertTrue(caManager.provisionCertificate(host, true, null)); Mockito.verify(agentManager, Mockito.times(2)).send(Mockito.anyLong(), Mockito.any(Answer.class)); Mockito.verify(agentManager, Mockito.times(1)).reconnect(Mockito.anyLong());