<div dir="ltr">On Thu, May 11, 2017 at 1:54 PM, Katia Aresti <span dir="ltr"><<a href="mailto:karesti@redhat.com" target="_blank">karesti@redhat.com</a>></span> wrote:<br><div><div><div><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hi Gustavo, thanks for the help !<div><br><div>Indeed, I stopped using the rv variable because I realised it was not always the compute command result, and somehow I decided I could not take for granted in the code that this is the compute command result value. But I think this the case only when the PrepareCommand is called on tx mode for this particular case.</div></div><div><br></div><div>What confused me a lot in this matter are commands like RemoveCommand, for example.</div><div><span style="font-family:menlo;font-size:9pt;color:rgb(0,0,128);font-weight:bold"><br></span></div><div><span style="font-family:menlo;font-size:9pt;color:rgb(0,0,128);font-weight:bold">private void </span><span style="color:rgb(0,0,0);font-family:menlo;font-size:9pt">processRemoveCommand(</span><span style="font-family:menlo;font-size:9pt;color:rgb(0,0,128);font-weight:bold">final </span><span style="color:rgb(0,0,0);font-family:menlo;font-size:9pt">RemoveCommand command, </span><span style="font-family:menlo;font-size:9pt;color:rgb(0,0,128);font-weight:bold">final </span><span style="color:rgb(0,0,0);font-family:menlo;font-size:9pt">InvocationContext ctx, </span><span style="font-family:menlo;font-size:9pt;color:rgb(0,0,128);font-weight:bold">final </span><span style="color:rgb(0,0,0);font-family:menlo;font-size:9pt">Object valueRemoved, TransactionContext transactionContext)</span><br></div><div><pre><font face="arial, helvetica, sans-serif">When this method is called from the visitPrepareCommand method, we indeed pass the previous value found just before the call in the cache. </font></pre><pre><font face="arial, helvetica, sans-serif">B</font><span style="font-family:arial,helvetica,sans-serif">ut after, when the same method is called from the visitRemoveCommand we passe the rv parameter. Which is the Remove Command perfom method answer.</span></pre><pre><font face="arial, helvetica, sans-serif"><br>So, having a look to the RemoveCommand, I can see sometimes we indeed return the prev value. But it might happen depending to return a boolean instead. </font></pre><pre><font face="arial, helvetica, sans-serif"><br></font></pre><pre><font face="monospace, monospace">remove("key", "value")cache2<span style="font-size:9pt;color:rgb(0,0,0)">.remove(</span><span style="font-size:9pt;color:rgb(0,128,0);font-weight:bold">"<wbr>newGoat"</span><span style="font-size:9pt;color:rgb(0,0,0)">);</span><br></font></pre><pre><font face="menlo" color="#000000"><b>LOGS</b></font></pre><pre><font face="menlo" color="#000000">Thread => 79
ctx.getOrigin() => ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeA-15125
valueRemoved => Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}
Thread => 38
ctx.getOrigin() => ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeB-23459
valueRemoved => Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}
Thread => 47
ctx.getOrigin() => null
valueRemoved => Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}
</font><font face="menlo"><b><font color="#cc0000">removeFromIndexes method is called !!!</font><span style="color:rgb(0,0,0);font-size:9pt"><br></span></b></font></pre><pre><font face="menlo" color="#000000"><br></font></pre><pre><font face="arial, helvetica, sans-serif"><font color="#000000">But if we call remove with specific value, the method </font>removeFromIndexes is never called.</font></pre><pre><pre style="color:rgb(0,0,0);font-size:9pt"><font face="monospace, monospace"><span style="color:rgb(102,14,122);font-weight:bold">cache2</span>.remove(<span style="color:rgb(0,128,0);font-weight:bold">"newGoat"</span>, person4);</font></pre></pre><pre><font face="menlo" color="#000000"><b>LOGS</b></font></pre><pre><font face="menlo" color="#000000">Thread => 79
ctx.getOrigin() => ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeA-22063
valueRemoved => true
Thread => 38
ctx.getOrigin() => ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeB-47249
valueRemoved => true
Thread => 47
ctx.getOrigin() => null
valueRemoved => true<br></font></pre><pre><font face="arial, helvetica, sans-serif">But in both cases the remove seems to be working, because this assertions work.</font></pre><span class="gmail-"><pre><font face="monospace, monospace">found = searchManager.<Person><wbr>getQuery(allQuery, Person.<span style="font-size:9pt;color:rgb(0,0,128);font-weight:bold">class</span><span style="color:rgb(0,0,0);font-size:9pt">).list();</span><br></font></pre></span><pre><pre style="color:rgb(0,0,0);font-size:9pt"><font face="monospace, monospace"><span style="font-style:italic">assertEquals</span>(<span style="color:rgb(0,0,255)">0</span>, found.size());</font></pre></pre><pre style="color:rgb(0,0,0);font-family:menlo;font-size:9pt"><br></pre><pre style="color:rgb(0,0,0);font-size:9pt"><font face="arial, helvetica, sans-serif">I'm a bit confused about all this, but we can chat o IRC or bluejeans. It might be some bugs concerning this interceptor, Radim has already opened an issue on this matter.</font></pre></div></div></blockquote><div><br></div><div>It seems to be working, since queries results are correct, but the value is not actually removed from the indexes.<br><br><a href="https://issues.jboss.org/browse/ISPN-7825">https://issues.jboss.org/browse/ISPN-7825</a><br><br></div><div>Gustavo<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><span class="gmail-HOEnZb"><font color="#888888"><pre style="color:rgb(0,0,0);font-size:9pt"><font face="arial, helvetica, sans-serif">Katia</font></pre></font></span></div></div><div class="gmail-HOEnZb"><div class="gmail-h5"><div class="gmail_extra"><br><div class="gmail_quote">On Wed, May 10, 2017 at 3:33 PM, Gustavo Fernandes <span dir="ltr"><<a href="mailto:gustavo@infinispan.org" target="_blank">gustavo@infinispan.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>The test fails every time NodeB (cache2) happens to <b>not be</b> the primary owner of 'newGoat', and needs to forward the command to NodeA. Sequence of events:<br><br>1) [NodeB] Compute is called, key is 'newGoat'<br>2) [NodeB] Command gets visited by the QueryInterceptor, that suspends the execution<br>3) [NodeB] Owner for 'newGoat' is nodeA, so nodeB forwards the command to nodeA<br><br>4) [NodeA] Command gets visited by the QueryInterceptor, that suspends the execution<br>5) [NodeA] perform() is called in the Compute command<br>6) [NodeA] Command is then sent to NodeB, which is a backup owner<br><br>7) [NodeB] Command gets visited by the QueryInterceptor, that suspends the execution<br>8) [NodeB] perform() is called in the compute command<br>9) [NodeB] QueryInterceptor resumes execution. Since command was originated remotely, no indexing is done (due to Index.LOCAL)<br><br>9) [NodeA] Receive response from the call done on 6)<br>10)[NodeA] resumes execution from the QueryInterceptor from 4)<br>11)[NodeA] Since command was originated remotely, no indexing is done (due to Index.LOCAL)<br><br>12)[NodeB] receives response from 3). At this point <b>the computed value is available</b> as the return type of the remote invocation<br>13)[NodeB] resumes the QueryInterceptor invocation from 2)<br>14)[NodeB] processComputes is then executed, but since the computedValue is not available in the command itself nor in the context, indexing is skipped since there is no value to index or remove<br><br><br></div><div>Looking at the method visitComputCommand, the variable "rv" stores the return value from the command, but it's not being used, instead the stateBeforeCompute is used which is always null in this scenario,<br></div><div>because it is evaluated on 2) which is before the the key exists in the data container:<br></div><div><pre style="background-color:rgb(255,255,255);color:rgb(0,0,0);font-family:"dejavu sans mono";font-size:9pt"><span style="color:rgb(0,0,128);font-weight:bold">return </span>invokeNextThenAccept(ctx, command, (rCtx, rCommand, rv) -> processComputeCommand(((Comput<wbr>eCommand) rCommand), rCtx, <span style="background-color:rgb(228,228,255)">stateBeforeCompute</span>, <span style="color:rgb(0,0,128);font-weight:bold">null</span>));<br></pre></div><div><div><div><div><pre style="background-color:rgb(255,255,255);color:rgb(0,0,0);font-family:"dejavu sans mono";font-size:9pt"><br>Gustavo</pre><div><div class="gmail_extra"><br><div class="gmail_quote"><div><div class="gmail-m_-1959327242177314279h5">On Tue, May 9, 2017 at 2:21 PM, Katia Aresti <span dir="ltr"><<a href="mailto:karesti@redhat.com" target="_blank">karesti@redhat.com</a>></span> wrote:<br></div></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><div class="gmail-m_-1959327242177314279h5"><div dir="ltr"><div>Hi all,</div><div><br></div><div>I'm really struggling with something in order to finish the compute methods.</div><div><br></div><div>I added a test in <b>ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT</b></div><div><br></div><div><pre style="color:rgb(0,0,0);font-family:menlo;font-size:9pt"><pre style="font-family:menlo;font-size:9pt"><span style="color:rgb(0,0,128);font-weight:bold">public void </span>testToto() <span style="color:rgb(0,0,128);font-weight:bold">throws </span>Exception {<br> SearchManager searchManager = Search.<span style="font-style:italic">getSearchManager</span>(<span style="color:rgb(102,14,122);font-weight:bold">cache2</span><wbr>);<br> QueryBuilder queryBuilder = searchManager<br> .buildQueryBuilderForClass(Per<wbr>son.<span style="color:rgb(0,0,128);font-weight:bold">class</span>)<br> .get();<br> Query allQuery = queryBuilder.all().createQuery<wbr>();<br><br> String key = <span style="color:rgb(0,128,0);font-weight:bold">"newGoat"</span>;<br> Person person4 = <span style="color:rgb(0,0,128);font-weight:bold">new </span>Person(key, <span style="color:rgb(0,128,0);font-weight:bold">"eats something"</span>, <span style="color:rgb(0,0,255)">42</span>);<br><br> <span style="color:rgb(102,14,122);font-weight:bold">cache2</span>.putIfAbsent(key, person4);<br> StaticTestingErrorHandler.<span style="font-style:italic">asse<wbr>rtAllGood</span>(<span style="color:rgb(102,14,122);font-weight:bold">cache1</span>, <span style="color:rgb(102,14,122);font-weight:bold">cache2</span>);<br><br> List<Person> found = searchManager.<Person>getQuery<wbr>(allQuery, Person.<span style="color:rgb(0,0,128);font-weight:bold">class</span>).list();<br> <span style="font-style:italic">assertEquals</span>(<span style="color:rgb(0,0,255)">1</span>, found.size());<br> <span style="font-style:italic">assertTrue</span>(found.contains(pers<wbr>on4));<br>}</pre></pre></div><div><div>I put some logs in the <span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-family:menlo;font-size:9pt">processPutKeyValueCommand </span><wbr>method in the <b>QueryInterceptor</b> to explain what is happening.<br></div><div><br></div><div><b>2 threads</b></div><div>Sometimes two threads get involved. </div><div><br></div><div>= Thread 72 First (or second) call </div><div>It happens from a non local Node. The so the <span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-family:menlo;font-size:9pt">shouldModifyIndexes </span>says "no, you should not modify any index" because the <span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-family:menlo;font-size:9pt">IndexModificationStrategy</span> <wbr>is set to "LOCAL ONLY". [1]<br></div><div><br></div><div><div><font face="monospace, monospace">72 ctx.getOrigin() = ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeB-19565</font></div><div><font face="monospace, monospace">72 should modify false</font></div><div><font face="monospace, monospace">72 previousValue null</font></div><div><font face="monospace, monospace">72 putValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null} // value in the command</font></div><div><font face="monospace, monospace">72 contextValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null} //value in the invocation context</font></div></div><div><br></div><div>= Thread 48 Second (or first) call</div><div>the origin is null, and this is considered as a LOCAL in the <span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-family:menlo;font-size:9pt">SingleKeyNonTxInvocationCo<wbr>ntext. [2]</span> In this case, the index is modified correctly, the value in the context has already been set up by the PutKeyValueCommand and the index get's correctly updated.</div><div><br></div><div><div><font face="monospace, monospace">48 ctx.getOrigin() = null</font></div><div><font face="monospace, monospace">48 should modify true</font></div><div><font face="monospace, monospace">48 previousValue null</font></div><div><font face="monospace, monospace">48 putValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace">48 contextValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div></div><div><br></div><div>And everything is ok. Everything is fine too in the case of a compute method instead of the put method.</div><div><br></div><div>But sometimes, this is not executed like that.</div></div><div><br></div><div><div><b>3 threads</b></div></div><div><b><br></b></div><div><div>What is a bit more weird to me is this second scenario where the commands are executed both from non local nodes (A and B). And so the index is not updated.</div><div>But just later, another thread get's involved and calls the QueryInterceptor with a invocation context where the command has not been executed (the value is not inside the context and the debugger does not enter in the perform method, this has happened just twice before). This call is coming like from a callback? in the <span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-family:menlo;font-size:9pt">QueueAsyncInvocationStage.</span></div></div><div><span style="background-color:rgb(228,228,255);color:rgb(0,0,0);font-size:9pt"><font face="monospace, monospace"><br></font></span></div><div><div><font face="monospace, monospace">80 ctx.getOrigin() = ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeA-65110</font></div><div><font face="monospace, monospace">80 should modify false</font></div><div><font face="monospace, monospace">80 prev null</font></div><div><font face="monospace, monospace">80 putValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace">80 contextValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">38 ctx.getOrigin() = ClusteredCacheWithElasticsearc<wbr>hIndexManagerIT-NodeB-35919</font></div><div><font face="monospace, monospace">38 should modify false</font></div><div><font face="monospace, monospace">38 prev null</font></div><div><font face="monospace, monospace">38 putValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace">38 contextValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">48 ctx.getOrigin() = null</font></div><div><font face="monospace, monospace">48 should modify true</font></div><div><font face="monospace, monospace">48 prev null</font></div><div><font face="monospace, monospace">48 putValue Person{name='newGoat', blurb='eats something', age=42, dateOfGraduation=null}</font></div><div><font face="monospace, monospace" color="#ff0000">48 contextValue null</font></div></div><div><br></div><div><br></div><div>This execution works perfectly with PutKeyValueCommand. But don't work wth compute.</div><div><br></div><div>The "computed value" is not inside the Command like put, replace or others. It is computed in the perform method (if needed). So, the first time the command is executed in A, the computed value is in the context, but the index is not updated. Second call, executed in B, value in context, but the index is not updated. The magic callback is executed, but the computed value is nowhere because the command is not executed a third time, so the context is null.<br></div><div><br></div><div>Can somebody please give me some light on this and explain to me what am I missing ? Other tests are failing for the same problem, like org.infinispan.query.blac<wbr>kbox.ClusteredCacheWithInfinis<wbr>panDirectoryTest</div><div><br></div><div>Thank you very much for your help !</div><div><br></div><div>Katia</div><div><br></div><div>[1] <a href="https://github.com/infinispan/infinispan/blob/master/query/src/main/java/org/infinispan/query/backend/IndexModificationStrategy.java#L50" target="_blank">https://github.com/infinis<wbr>pan/infinispan/blob/master/que<wbr>ry/src/main/java/org/infinispa<wbr>n/query/backend/IndexModificat<wbr>ionStrategy.java#L50</a></div><div>[2] <a href="https://github.com/infinispan/infinispan/blob/master/core/src/main/java/org/infinispan/context/SingleKeyNonTxInvocationContext.java#L39" target="_blank">https://github.com/infinis<wbr>pan/infinispan/blob/master/cor<wbr>e/src/main/java/org/infinispan<wbr>/context/SingleKeyNonTxInvocat<wbr>ionContext.java#L39</a></div><div><br></div></div>
<br></div></div>______________________________<wbr>_________________<br>
infinispan-dev mailing list<br>
<a href="mailto:infinispan-dev@lists.jboss.org" target="_blank">infinispan-dev@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/infinispan-dev" rel="noreferrer" target="_blank">https://lists.jboss.org/mailma<wbr>n/listinfo/infinispan-dev</a><br></blockquote></div><br></div></div></div></div></div></div></div>
<br>______________________________<wbr>_________________<br>
infinispan-dev mailing list<br>
<a href="mailto:infinispan-dev@lists.jboss.org" target="_blank">infinispan-dev@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/infinispan-dev" rel="noreferrer" target="_blank">https://lists.jboss.org/mailma<wbr>n/listinfo/infinispan-dev</a><br></blockquote></div><br></div>
</div></div><br>______________________________<wbr>_________________<br>
infinispan-dev mailing list<br>
<a href="mailto:infinispan-dev@lists.jboss.org">infinispan-dev@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/infinispan-dev" rel="noreferrer" target="_blank">https://lists.jboss.org/<wbr>mailman/listinfo/infinispan-<wbr>dev</a><br></blockquote></div><br></div></div></div></div></div>