The abort is a bug of its own. Filed with a patch here: https://issues.apache.org/jira/browse/THRIFT-780
Jonathan Ellis wrote: > Is this related to https://issues.apache.org/jira/browse/THRIFT-638 or > something else? > > On Mon, Apr 26, 2010 at 10:44 PM, David Reiss <[email protected]> wrote: >> The author of the php module sent me a patch for this. I'll try to get >> it up on JIRA and into trunk tomorrow. >> >> Ken Sandney wrote: >>> Hi >>> >>> I am using Cassandra PHP client and if I enable thrift_protocol >>> extension I always got some time-out error. Here is the message I sent >>> to cassandra user mailing list: >>> >>> Hi, >>> >>> The attachment is a package for testing time-outs. In my situation I >>> have 4 nodes cluster with latest 0.6.1. I am doing this test in >>> command line and only for 1 thread. With php thrift_protocol enabled >>> it seems more chances to get time-out and sometimes crashed with >>> this error: >>> >>> ...terminate called after throwing an instance of >>> 'PHPExceptionWrapper' >>> what(): PHP exception zval=0x10075508 >>> Aborted >>> >>> >>> Without the extension and with this patch on >>> https://issues.apache.org/jira/browse/THRIFT-347 , the test result >>> seems better but still got time-outs >>> >>> To do the test, just unpack the package and run >>> >>> $php test.php >>> >>> >>> all exception and time-outs will be written to log.txt. Before test >>> starts, you have to modify test.php for Cassandra connection settings. >>> >>> I am new to Cassandra and have some RDBMS background. Remind >>> me/correct me if I missed something/have made any mistakes. >>> >>> Thank you >>> >>> >>> And the attachment is the package I made for test >>> >>> Thanks for any feedbacks > > >
