[jira] [Resolved] (KUDU-1374) Operations triggered by TS heartbeats may go unperformed

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

     [ https://issues.apache.org/jira/browse/KUDU-1374?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Adar Dembo resolved KUDU-1374.
------------------------------
       Resolution: Fixed
    Fix Version/s: 0.10.0

Fixed in commit d648141.

> Operations triggered by TS heartbeats may go unperformed
> --------------------------------------------------------
>
>                 Key: KUDU-1374
>                 URL: https://issues.apache.org/jira/browse/KUDU-1374
>             Project: Kudu
>          Issue Type: Sub-task
>          Components: master
>    Affects Versions: 0.7.1
>            Reporter: Adar Dembo
>            Assignee: Adar Dembo
>            Priority: Critical
>             Fix For: 0.10.0
>
>
> (copying this from my multi-master design doc)
> The inclusion or exclusion of a tablet in an incremental tablet report is edge-triggered, and may result in a state changing operation on the tserver, communicated via out-of-band RPC. This RPC is retried until it is successful. However, if the leader master dies *after* it is able to respond to the tserver's heartbeat but *before* the out-of-band RPC is sent, the edge-triggered tablet report may be missed, and the state changing operation will not be performed until the next time the tablet is included in a tablet report. As tablet report inclusion criteria is narrow, operations may be "missed" for quite some time.
> These operations include:
> # Some tablet deletions, such as tablets belonging to orphaned tables, or tablets whose deletion RPCs were sent and failed during an earlier *DeleteTable()* request.
> # Some tablet alters, such as tablets whose alter RPCs were sent and failed during an earlier *AlterTable()* request.
> # Config changes sent due to under-replicated tablets.
> A simple fix is to require that tservers send a full tablet report when they detect that a new leader master was elected.



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