Return-Path: X-Original-To: apmail-cassandra-commits-archive@www.apache.org Delivered-To: apmail-cassandra-commits-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id CFD1E10B22 for ; Wed, 19 Mar 2014 00:32:52 +0000 (UTC) Received: (qmail 13270 invoked by uid 500); 19 Mar 2014 00:32:47 -0000 Delivered-To: apmail-cassandra-commits-archive@cassandra.apache.org Received: (qmail 13102 invoked by uid 500); 19 Mar 2014 00:32:45 -0000 Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@cassandra.apache.org Delivered-To: mailing list commits@cassandra.apache.org Received: (qmail 13051 invoked by uid 99); 19 Mar 2014 00:32:44 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 19 Mar 2014 00:32:44 +0000 Date: Wed, 19 Mar 2014 00:32:44 +0000 (UTC) From: "Russ Hatch (JIRA)" To: commits@cassandra.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (CASSANDRA-6883) stress read fails with IOException "Data returned was not validated" MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/CASSANDRA-6883?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13940008#comment-13940008 ] Russ Hatch commented on CASSANDRA-6883: --------------------------------------- Ok, it makes sense that the schema/data has to be present before the reads will work. But given the fact that before the read, I did a million row write with uniform distribution (over only 100 cells) it seems a strange that the read test would fail due to missing data. I ran my write as above in the description and noticed a weird coincidence. The write step creates a schema as expected: {noformat} cqlsh> desc TABLE "Keyspace1"."Standard1"; CREATE TABLE "Keyspace1"."Standard1" ( key blob PRIMARY KEY, "C0" blob, ....................................snip.................................... "C99" blob {noformat} but the last cell never gets populated in that table, it remains null. So I think maybe there's some kind of mismatch between how the uniform distribution is written vs how it is read back (off by one or something). > stress read fails with IOException "Data returned was not validated" > -------------------------------------------------------------------- > > Key: CASSANDRA-6883 > URL: https://issues.apache.org/jira/browse/CASSANDRA-6883 > Project: Cassandra > Issue Type: Bug > Components: Tools > Environment: ccm 3 node cluster, java 1.7.0_51 > cassandra 2.1 branch -- 615612f61566 > Reporter: Russ Hatch > > I'm working to do some basic testing of read/write with the new stress. First, I populate data using write like so: > {noformat} > rhatch@whatup:~/git/cstar/cassandra$ tools/bin/cassandra-stress write n=1000000 CL=ONE -key dist=FIXED\(1\) -col n=UNIFORM\(1..100\) -schema replication\(factor=3\) -rate threads=50 > {noformat} > Then I attempt a read test, and this happens: > {noformat} > rhatch@whatup:~/git/cstar/cassandra$ tools/bin/cassandra-stress read n=1000000 CL=ONE -key dist=FIXED\(1\) -col n=UNIFORM\(1..100\) -schema replication\(factor=3\) -rate threads=50 > Warming up READ with 50000 iterations... > INFO 19:38:14 New Cassandra host /127.0.0.3 added > INFO 19:38:14 New Cassandra host /127.0.0.2 added > Connected to cluster: test_stress > Datatacenter: datacenter1; Host: localhost/127.0.0.1; Rack: rack1 > Datatacenter: datacenter1; Host: /127.0.0.2; Rack: rack1 > Datatacenter: datacenter1; Host: /127.0.0.3; Rack: rack1 > java.io.IOException: Operation [11055] x0 key 0000000001 (0x30303030303030303031) Data returned was not validated > at org.apache.cassandra.stress.Operation.error(Operation.java:298) > at org.apache.cassandra.stress.Operation.timeWithRetry(Operation.java:276) > at org.apache.cassandra.stress.operations.ThriftReader.run(ThriftReader.java:46) > at org.apache.cassandra.stress.StressAction$Consumer.run(StressAction.java:304) > Sleeping 2s... > Running READ with 50 threads for 1000000 iterations > ops , op/s,adj op/s, key/s, mean, med, .95, .99, .999, max, time, stderr > 11287 , 11286, 11286, 11286, 4.7, 3.8, 11.9, 21.8, 34.0, 52.9, 1.0, 0.00000 > 23263 , 11764, 11764, 11764, 4.3, 3.5, 10.4, 18.1, 29.8, 51.8, 2.0, 0.00000 > 35300 , 11889, 11889, 11889, 4.2, 3.7, 9.6, 15.3, 27.3, 40.0, 3.0, 0.01467 > 47239 , 11737, 11737, 11737, 4.3, 3.7, 10.1, 17.0, 26.9, 47.7, 4.0, 0.01289 > 59140 , 11729, 11729, 11729, 4.3, 3.8, 9.6, 15.1, 25.7, 47.9, 5.1, 0.00979 > java.io.IOException: Operation [66434] x0 key 0000000001 (0x30303030303030303031) Data returned was not validated > at org.apache.cassandra.stress.Operation.error(Operation.java:298) > at org.apache.cassandra.stress.Operation.timeWithRetry(Operation.java:276) > at org.apache.cassandra.stress.operations.ThriftReader.run(ThriftReader.java:46) > at org.apache.cassandra.stress.StressAction$Consumer.run(StressAction.java:304) > 66584 , 11952, 11952, 11952, 4.2, 3.6, 9.9, 15.7, 24.7, 64.9, 5.7, 0.00788 > FAILURE > {noformat} -- This message was sent by Atlassian JIRA (v6.2#6252)