[ https://issues.apache.org/jira/browse/YARN-2637?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231832#comment-14231832 ]
Junping Du commented on YARN-2637: ---------------------------------- Hi [~cwelch], sorry for coming late and thank you for updating the patch. A couple of comments: {code} Resource amLimit = Resources.multiply( lastClusterResource, maxAMResourcePerQueuePercent); {code} Looks like maxAMResourcePerQueuePercent is a allowed percent for AM resource in each queue. So we may should calculate amLimit per queue rather than aggregate all applications together. {code} + private int maxActiveApplicationsForQueue = -1; // To allow manualy setting {code} typo for *manualy* {code} + protected final Resource usedAMResources; {code} usedAMResources is not used by sub-class, so suggest to replace it with private {code} + if (application.getAMResource() == null) throw new RuntimeException("c1"); + if (usedAMResources == null) throw new RuntimeException("c2"); {code} Exception messages here should be more meaningful than "c1", or "c2". {code} + if (!Resources.fitsIn(amIfStarted, amLimit)) { + LOG.debug("not starting application as amIfStarted exceeds amLimit"); + continue; + } {code} The log level here should be info or warn level rather than debug level. In most cases, LOG.debug() should be under block of LOG.isDebugEnabled(). More comments may come later. > maximum-am-resource-percent could be violated when resource of AM is > > minimumAllocation > ---------------------------------------------------------------------------------------- > > Key: YARN-2637 > URL: https://issues.apache.org/jira/browse/YARN-2637 > Project: Hadoop YARN > Issue Type: Bug > Components: resourcemanager > Affects Versions: 2.6.0 > Reporter: Wangda Tan > Assignee: Craig Welch > Priority: Critical > Attachments: YARN-2637.0.patch, YARN-2637.1.patch, > YARN-2637.12.patch, YARN-2637.13.patch, YARN-2637.2.patch, YARN-2637.6.patch, > YARN-2637.7.patch, YARN-2637.9.patch > > > Currently, number of AM in leaf queue will be calculated in following way: > {code} > max_am_resource = queue_max_capacity * maximum_am_resource_percent > #max_am_number = max_am_resource / minimum_allocation > #max_am_number_for_each_user = #max_am_number * userlimit * userlimit_factor > {code} > And when submit new application to RM, it will check if an app can be > activated in following way: > {code} > for (Iterator<FiCaSchedulerApp> i=pendingApplications.iterator(); > i.hasNext(); ) { > FiCaSchedulerApp application = i.next(); > > // Check queue limit > if (getNumActiveApplications() >= getMaximumActiveApplications()) { > break; > } > > // Check user limit > User user = getUser(application.getUser()); > if (user.getActiveApplications() < > getMaximumActiveApplicationsPerUser()) { > user.activateApplication(); > activeApplications.add(application); > i.remove(); > LOG.info("Application " + application.getApplicationId() + > " from user: " + application.getUser() + > " activated in queue: " + getQueueName()); > } > } > {code} > An example is, > If a queue has capacity = 1G, max_am_resource_percent = 0.2, the maximum > resource that AM can use is 200M, assuming minimum_allocation=1M, #am can be > launched is 200, and if user uses 5M for each AM (> minimum_allocation). All > apps can still be activated, and it will occupy all resource of a queue > instead of only a max_am_resource_percent of a queue. -- This message was sent by Atlassian JIRA (v6.3.4#6332)