accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Denis Petrov (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (ACCUMULO-3485) accumulo client tries to connect only to the first ZK in the list, making it SPOF
Date Fri, 16 Jan 2015 02:34:35 GMT

     [ https://issues.apache.org/jira/browse/ACCUMULO-3485?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Denis Petrov updated ACCUMULO-3485:
-----------------------------------
    Description: 
Accumulo client tries to connect only to the first ZK server.
if that single ZK server is down, the accumulo client does not try other ZK servers.
Thus the first ZK is single point of failure, if it is down, accumulo client cannot work

{code:java}
  val instance = new ZooKeeperInstance("zzz", "no-zk-here.example.com,zk1.example.com,zk2.example.com,zk3.example.com")
  val conn = instance.getConnector("root", "pass".getBytes)
  println("db connected, tables=" + conn.tableOperations.list)
{code}

  was:
Accumulo client libraries tries to connect only to the first ZK server.
if that single ZK server is down, the accumulo client does not try other ZK servers.
Thus the first ZK is single point of failure, if it is down, accumulo client cannot work

{code:java}
  val instance = new ZooKeeperInstance("zzz", "no-zk-here.example.com,zk1.example.com,zk2.example.com,zk3.example.com")
  val conn = instance.getConnector("root", "pass".getBytes)
  println("db connected, tables=" + conn.tableOperations.list)
{code}


> accumulo client tries to connect only to the first ZK in the list, making it SPOF
> ---------------------------------------------------------------------------------
>
>                 Key: ACCUMULO-3485
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3485
>             Project: Accumulo
>          Issue Type: Bug
>    Affects Versions: 1.6.1
>            Reporter: Denis Petrov
>            Priority: Critical
>
> Accumulo client tries to connect only to the first ZK server.
> if that single ZK server is down, the accumulo client does not try other ZK servers.
> Thus the first ZK is single point of failure, if it is down, accumulo client cannot work
> {code:java}
>   val instance = new ZooKeeperInstance("zzz", "no-zk-here.example.com,zk1.example.com,zk2.example.com,zk3.example.com")
>   val conn = instance.getConnector("root", "pass".getBytes)
>   println("db connected, tables=" + conn.tableOperations.list)
> {code}



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

Mime
View raw message