thrift-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James E. King, III (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (THRIFT-4340) Erlang fix a crash on client close
Date Fri, 22 Sep 2017 19:00:03 GMT

     [ https://issues.apache.org/jira/browse/THRIFT-4340?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

James E. King, III updated THRIFT-4340:
---------------------------------------
    Description: 
When a client closes a connection to a framed server the server was
crashing because the fact that the transport was framed was being
lost. Looking through the file I noticed that the block from lines
87-95, looked different from the one from 59-66. The culprit was
that when an error was occuring in the 59-66 block it was being
propagated up without rewrapping. That would cause a failure
much further up the chain.

Note: original author's PR is THRIFT-4339 but the github PR wasn't linked to THRIFT-4339.

  was:
When a client closes a connection to a framed server the server was
crashing because the fact that the transport was framed was being
lost. Looking through the file I noticed that the block from lines
87-95, looked different from the one from 59-66. The culprit was
that when an error was occuring in the 59-66 block it was being
propagated up without rewrapping. That would cause a failure
much further up the chain.


> Erlang fix a crash on client close
> ----------------------------------
>
>                 Key: THRIFT-4340
>                 URL: https://issues.apache.org/jira/browse/THRIFT-4340
>             Project: Thrift
>          Issue Type: Bug
>          Components: Erlang - Library
>    Affects Versions: 0.10.0
>            Reporter: James E. King, III
>            Assignee: James E. King, III
>             Fix For: 0.11.0
>
>
> When a client closes a connection to a framed server the server was
> crashing because the fact that the transport was framed was being
> lost. Looking through the file I noticed that the block from lines
> 87-95, looked different from the one from 59-66. The culprit was
> that when an error was occuring in the 59-66 block it was being
> propagated up without rewrapping. That would cause a failure
> much further up the chain.
> Note: original author's PR is THRIFT-4339 but the github PR wasn't linked to THRIFT-4339.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message