[jira] [Comment Edited] (MESOS-2707) Incorrect zh:// URI scheme causes Slave to SegFault

["haosdent (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MESOS-2707?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14536861#comment-14536861 ] 

haosdent edited comment on MESOS-2707 at 5/9/15 8:13 PM:
---------------------------------------------------------

you should replace zh to zk [~oliverpp]


was (Author: haosdent@gmail.com):
>From document http://mesos.apache.org/documentation/latest/high-availability/ , mesos-slave should start with zk.

And from code, 
{code}
  if (mechanism == "") {
    return new StandaloneMasterDetector();
  } else if (strings::startsWith(mechanism, "zk://")) {
    Try<zookeeper::URL> url = zookeeper::URL::parse(mechanism);
    if (url.isError()) {
      return Error(url.error());
    }
    if (url.get().path == "/") {
      return Error(
          "Expecting a (chroot) path for ZooKeeper ('/' is not supported)");
    }
    return new ZooKeeperMasterDetector(url.get());
  } else if (strings::startsWith(mechanism, "file://")) {
{code}

mesos-slave also should start with zk. So I think the real cause of segfault should be other reason.

> Incorrect zh:// URI scheme causes Slave to SegFault
> ---------------------------------------------------
>
>                 Key: MESOS-2707
>                 URL: https://issues.apache.org/jira/browse/MESOS-2707
>             Project: Mesos
>          Issue Type: Bug
>    Affects Versions: 0.22.0
>         Environment: Linux iZ25to7d407Z 2.6.32-504.16.2.el6.x86_64 #1 SMP Wed Apr 22 06:48:29 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
> Mesos 0.22.0，built from sources
> Zookeeper 3.4.6
>            Reporter: Shengwu Jiang
>            Assignee: Marco Massenzio
>
> I have 4 slave nodes with the same hardware, operating system and mesos configuration. 
> Few minutes ago, all 4 nodes were functioning well. I tried to change the config of *master* from _10.172.230.69:5050_ to _zh://10.172.230.69:2181/mesos_ and restarted them in turn. The other three had started normally but the last one got a segmentation fault as you can see below.
> {code}
> [root@iZ25to7d407Z ~]# mesos-slave --master=zh://10.172.230.69:2181/mesos --hostname=123.57.42.237 --containerizers=docker,mesos --quiet &
> [1] 1216
> [root@iZ25to7d407Z ~]# *** Aborted at 1431085131 (unix time) try "date -d @1431085131" if you are using GNU date ***
> PC: @       0x3aede7b53c (unknown)
> *** SIGSEGV (@0x0) received by PID 1216 (TID 0x7f12f984b820) from PID 0; stack trace: ***
>     @       0x3aee20f710 (unknown)
>     @       0x3aede7b53c (unknown)
>     @       0x3aedecf630 (unknown)
>     @     0x7f12fce1593f net::getIP()
>     @     0x7f12fce507ae process::operator>>()
>     @     0x7f12fce50107 process::UPID::UPID()
>     @     0x7f12fc52af71 mesos::internal::MasterDetector::create()
>     @           0x4b1290 main
>     @       0x3aede1ed5d (unknown)
>     @           0x4b00b9 (unknown)
> [1]+  Segmentation fault      mesos-slave --master=zh://10.172.230.69:2181/mesos --hostname=123.57.42.237 --containerizers=docker,mesos --quiet
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)