Java tutorial
/* * RHQ Management Platform * Copyright (C) 2005-2008 Red Hat, Inc. * All rights reserved. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation version 2 of the License. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ package org.rhq.enterprise.server.scheduler.jobs; import org.apache.commons.logging.Log; import org.apache.commons.logging.LogFactory; import org.quartz.Job; import org.quartz.JobDataMap; import org.quartz.JobDetail; import org.quartz.JobExecutionContext; import org.quartz.JobExecutionException; import org.quartz.StatefulJob; import org.rhq.core.domain.auth.Subject; import org.rhq.core.domain.content.ContentSource; import org.rhq.core.domain.content.Repo; import org.rhq.core.domain.criteria.RepoCriteria; import org.rhq.core.domain.util.PageList; import org.rhq.enterprise.server.auth.SubjectManagerLocal; import org.rhq.enterprise.server.content.ContentSourceManagerLocal; import org.rhq.enterprise.server.content.RepoManagerLocal; import org.rhq.enterprise.server.util.LookupUtil; /** * This is a Quartz scheduler job whose job is to synchronize one particular {@link ContentSource}. After synchronizing * the metadata, this will also attempt to load the content if the content source is not configured for * {@link ContentSource#isLazyLoad() lazy loading}. * * <p>This implements {@link StatefulJob} (as opposed to {@link Job}) because we do not need nor want this job triggered * concurrently. That is, we don't need multiple instances of this job running at the same time.</p> * * @author John Mazzitelli * @author Jason Dobies */ public class ContentProviderSyncJob implements StatefulJob { private static final String DATAMAP_CONTENT_SOURCE_NAME = "contentSourceName"; private static final String DATAMAP_CONTENT_SOURCE_TYPE_NAME = "contentSourceTypeName"; public static final String DATAMAP_SYNC_IMPORTED_REPOS = "syncImportedRepos"; private static final Log log = LogFactory.getLog(ContentProviderSyncJob.class); private static final String SEPARATOR = "--"; public int getPercentComplete() { int retval = (int) (Math.random() * 100); return retval; } public void execute(JobExecutionContext context) throws JobExecutionException { try { JobDetail jobDetail = context.getJobDetail(); if (jobDetail == null) { throw new IllegalStateException("The job does not have any details"); } JobDataMap dataMap = jobDetail.getJobDataMap(); if (dataMap == null) { throw new IllegalStateException("The job does not have any data in its details"); } String name = dataMap.getString(DATAMAP_CONTENT_SOURCE_NAME); String typeName = dataMap.getString(DATAMAP_CONTENT_SOURCE_TYPE_NAME); boolean syncImportedRepos = dataMap.getBoolean(DATAMAP_SYNC_IMPORTED_REPOS); if (name == null) { throw new IllegalStateException("Missing the content source name in details data"); } if (typeName == null) { throw new IllegalStateException("Missing the content source type name in details data"); } ContentSource contentSource = synchronizeAndLoad(name, typeName); if (syncImportedRepos) { // Synchronize all repos associated with the content source - these may be repos we just // discovered during this sync or ones that were already associated. syncImportedRepos(contentSource); } } catch (Exception e) { String errorMsg = "Failed to sync content source in job [" + context.getJobDetail() + "]"; log.error(errorMsg, e); JobExecutionException jobExecutionException = new JobExecutionException(errorMsg, e, false); // should we unschedule so we never attempt to sync again? // That would mean any error will cause this sync to never occur again automatically until // we restart the server, restart the server-side content plugin container or somehow manually create // the schedule again. I will assume we will allow this schedule to trigger again, not sure if // that is what we want, but we can flip this to true if we want the other behavior. // NOTE: we do NOT retrigger if we threw IllegalStateException because we know it'll never work anyway if (!(e instanceof IllegalStateException)) { jobExecutionException.setUnscheduleAllTriggers(false); } throw jobExecutionException; } } private void syncImportedRepos(ContentSource contentSource) throws InterruptedException { SubjectManagerLocal subjectManager = LookupUtil.getSubjectManager(); Subject overlord = subjectManager.getOverlord(); RepoCriteria repoCriteria = new RepoCriteria(); repoCriteria.addFilterContentSourceIds(contentSource.getId()); repoCriteria.addFilterCandidate(false); RepoManagerLocal repoManager = LookupUtil.getRepoManagerLocal(); PageList<Repo> repos = repoManager.findReposByCriteria(overlord, repoCriteria); Integer[] repoIds = new Integer[repos.size()]; for (int i = 0; i < repos.size(); i++) { Repo repo = repos.get(i); repoIds[i] = repo.getId(); } repoManager.internalSynchronizeRepos(overlord, repoIds); } /** * This will synchronize the identified content source such that its package version information is updated and, if * not lazy-loading, its package bits are downloaded. * * <p>Note that this method executes outside of any transaction. This is very important since this job is * potentially very long running (on the order of hours potentially). We do our processing in here with this in * mind. We make sure we never do any one thing that potentially could timeout a transaction.</p> * * @param contentSourceName name of the {@link org.rhq.core.domain.content.ContentSource} * @param contentSourceTypeName name of the {@link org.rhq.core.domain.content.ContentSourceType} * * @throws Exception if either the sync failed or one of the packages failed to download */ private ContentSource synchronizeAndLoad(String contentSourceName, String contentSourceTypeName) throws Exception { // note that we will keep calling getOverlord on this subject manager - the overlord // has a very short session lifespan so we need to keep asking for a new one, due to the possibility // that some of the methods we call here take longer than the overlord's lifespan SubjectManagerLocal subjectManager = LookupUtil.getSubjectManager(); Subject overlord; ContentSourceManagerLocal contentManager = LookupUtil.getContentSourceManager(); ContentSource contentSource; overlord = subjectManager.getOverlord(); contentSource = contentManager.getContentSourceByNameAndType(overlord, contentSourceName, contentSourceTypeName); if (contentSource == null) { throw new Exception("Sync job was asked to sync an unknown content source: " + contentSourceName + "|" + contentSourceTypeName); } int contentSourceId = contentSource.getId(); // If 'completed' is false, there was already a synchronization taking place, // so we should abort and let that already running sync take care of everything. boolean completed = contentManager.internalSynchronizeContentSource(contentSourceId); if (!completed) { log.info("Content source [" + contentSourceName + "] is currently being synchronized already. " + "Please wait for the current sync job to finish."); } return contentSource; } /** * All content source sync jobs must have specified data prepared * in their job details data map. This creates that data map. You must * call this method everytime you schedule a content source sync job. * If the given details is not <code>null</code>, this will place the * created data map in the details for you. Otherwise, you must ensure * the returned data map gets associated with the job when it is created. * * @param contentSource the content source whose sync job's details is being prepared * @param details where the job's data map will be stored (may be <code>null</code>) * * @return the data map with the data necessary to execute a content sync job */ public static JobDataMap createJobDataMap(ContentSource contentSource, JobDetail details) { JobDataMap dataMap; if (details != null) { dataMap = details.getJobDataMap(); } else { dataMap = new JobDataMap(); } dataMap.put(DATAMAP_CONTENT_SOURCE_NAME, contentSource.getName()); dataMap.put(DATAMAP_CONTENT_SOURCE_TYPE_NAME, contentSource.getContentSourceType().getName()); return dataMap; } /** * Creates the name for the scheduled content source's sync job. Calling this * method multiple times with the same content source always produces the same name. * * @param cs the content source whose scheduled job name is to be returned * * @return the scheduled job name for the given content source */ public static String createJobName(ContentSource cs) { // the quartz table has a limited column width of 80 - but we need to use the names to make jobs unique // so encode the names' hashcodes to ensure we fix into the quartz job name column. String nameEncoded = Integer.toHexString(cs.getName().hashCode()); String typeNameEncoded = Integer.toHexString(cs.getContentSourceType().getName().hashCode()); String jobName = nameEncoded + SEPARATOR + typeNameEncoded; if (jobName.length() > 80) { throw new IllegalArgumentException( "Job names max size is 80 chars due to DB column size restrictions: " + jobName); } return jobName; } /** * Creates a unique name for a new content source sync job. Calling this * method multiple times with the same content source always produces a different name * which is useful if you want to schedule an new job that is separate and distinct * from any other job in the system. * * @param cs the content source * * @return a unique job name that can be used for a new job to sync a given content source */ public static String createUniqueJobName(ContentSource cs) { // the quartz table has a limited column width of 80 - but we need to use the names to make jobs unique // so encode the names' hashcodes to ensure we fix into the quartz job name column. // appendStr is used to make the job unique among others for the same content source. String nameEncoded = Integer.toHexString(cs.getName().hashCode()); String typeNameEncoded = Integer.toHexString(cs.getContentSourceType().getName().hashCode()); String appendStr = Long.toHexString(System.currentTimeMillis()); String jobName = nameEncoded + SEPARATOR + typeNameEncoded + SEPARATOR + appendStr; if (jobName.length() > 80) { throw new IllegalArgumentException( "Job names max size is 80 chars due to DB column size restrictions: " + jobName); } return jobName; } }