https://jira.opendaylight.org This file is an XML representation of an issue en-us 8.20.10 820010 22-06-2022 https://jira.opendaylight.org/browse/OPNFLWPLUG-1047 OpenFlowPlugin <p>I'm looking at a Java Flight Recording obtained from (internal) scale lab testing, , based on Oxygen SR3 code, and am surprised to find that the biggest "TLAB allocation", with a huge factor of several magnitudes (4.97 TiB vs. next issue at 86/57/16/15 ... GiB!), is this:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> Iterator java.util.concurrent.ConcurrentHashMap$EntrySetView.iterator() 127113 void org.opendaylight.openflowplugin.applications.frm.nodeconfigurator.NodeConfiguratorImpl$JobQueueHandler.run() 127113 void java.lang.<span class="code-object">Thread</span>.run() 127113</pre> </div></div> <p>I'm not clear how object creation churn, and thus required GC, for a simple iterator() could be reduced, but it seems worth having a closer look at this code.</p> OPNFLWPLUG-1047

NodeConfiguratorImpl JobQueueHandler iterator() causing extensive object allocation

Bug Highest Resolved Done Michael Vorburger Michael Vorburger Tue, 13 Nov 2018 13:15:42 +0000 Wed, 5 Dec 2018 16:33:45 +0000 Wed, 5 Dec 2018 16:33:45 +0000 Oxygen-SR3 Oxygen-SR4 Fluorine-SR2 Neon 0 2 <p>That code needs to be refactored, because:</p> <ol> <li>it is a huge busy loop</li> <li>if (shutdown) does not actually shutdown</li> </ol> <p> </p> <p>Actually, this should be reworked to just use org.opendaylight.yangtools.util.concurrent.QueuedNotificationManager, which deals with the same issue and is known to actually work.</p> <p><a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=rovarga" class="user-hover" rel="rovarga">rovarga</a> Thank You very much for having jumped right on to contributing a fix for this problem!!</p> <p>No problem, I just didn't want you to go into a rabiit hole. The patch still needs to be verified to behave as needed – I won't have time to do that, anybody feel free to pick it up.</p> <p><a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=Avishnoi" class="user-hover" rel="Avishnoi">Avishnoi</a>, or any other active OFP commiters <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=d.arunprakash%40ericsson.com" class="user-hover" rel="d.arunprakash@ericsson.com">d.arunprakash@ericsson.com</a> (<a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=Arunprakash" class="user-hover" rel="Arunprakash">Arunprakash</a> ?), <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=gobinath%40ericsson.com" class="user-hover" rel="gobinath@ericsson.com">gobinath@ericsson.com</a> (<a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=gobinath" class="user-hover" rel="gobinath">gobinath</a> ?), <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=hema.gopalkrishnan%40ericsson.com" class="user-hover" rel="hema.gopalkrishnan@ericsson.com">hema.gopalkrishnan@ericsson.com</a> (<a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=Hematg" class="user-hover" rel="Hematg">Hematg</a> ?), <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=pp.prasana%40gmail.com" class="user-hover" rel="pp.prasana@gmail.com">pp.prasana@gmail.com</a>, <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=shuva.jyoti.kar.87%40gmail.com" class="user-hover" rel="shuva.jyoti.kar.87@gmail.com">shuva.jyoti.kar.87@gmail.com</a> (<a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=SHUVA+JYOTI+KAR" class="user-hover" rel="SHUVA JYOTI KAR">SHUVA JYOTI KAR</a> ?), or perhaps <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=jluhrsen" class="user-hover" rel="jluhrsen">jluhrsen</a>, will you run a suitable CSIT on <a href="https://git.opendaylight.org/gerrit/#/c/77732" class="external-link" target="_blank" rel="nofollow noopener">c/77732</a> and get it merged soon-ish? We'll also need a back-port. (<a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=tpantelis" class="user-hover" rel="tpantelis">tpantelis</a> already did a code review of <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=rovarga" class="user-hover" rel="rovarga">rovarga</a>'s contribution and +1 it.)</p> <p>&gt; we'll also need a back-port.</p> <p>may be safer if only to Fluorine but not to Oxygen anymore.. just to be extra cautious.</p> <p>&gt; may be safer if only to Fluorine but not to Oxygen anymore.. just to be extra cautious.</p> <p>Let's actually (propose to) get this into Oxygen as well; because <a href="https://jira.opendaylight.org/secure/ViewProfile.jspa?name=skitt" class="user-hover" rel="skitt">skitt</a> has seen this one again as the major allocator in a JFR (the one from <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1651668" class="external-link" target="_blank" rel="nofollow noopener">https://bugzilla.redhat.com/show_bug.cgi?id=1651668</a>).</p> Development Rank 0|i03ken: