[
https://issues.apache.org/jira/browse/JCLOUDS-1192?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15632368#comment-15632368
]
Andrei commented on JCLOUDS-1192:
---------------------------------
Thanks [~nacx], currently I see the reason of usage mentioned parameter.
As for {{submitScriptOnNode}} or {{runScriptOnNode}}... well I still should
care about concurrency. And carry out force close {{ComputeServiceContext}} with
{noformat}
net.schmizz.sshj.common.SSHException: Connection reset
at net.schmizz.sshj.common.SSHException$1.chain(SSHException.java:36)
~[sshj-0.10.0.jar:na]
at net.schmizz.sshj.common.SSHException$1.chain(SSHException.java:29)
~[sshj-0.10.0.jar:na]
{noformat}
> TIMEOUT_SCRIPT_COMPLETE doesn't work.
> -------------------------------------
>
> Key: JCLOUDS-1192
> URL: https://issues.apache.org/jira/browse/JCLOUDS-1192
> Project: jclouds
> Issue Type: Bug
> Components: jclouds-scriptbuilder
> Affects Versions: 1.9.2
> Reporter: Andrei
>
> Provided part of source code performs running _host_update.sh_ scripts via
> *runScriptOnNode* method. Mentioned script contains at the end command that
> performs new release in DHCP, and as result *ComputeService* lost connection
> and cannot return the *ExecResponse*. It leads to hung execution app. It has
> been tried to configure any timeouts, but it doesn't interrupt
> *runScriptOnNode* method in configured time.
> Operating System Version - Windows 10
> Java version - 1.8.0_102
> Part of source code could be found
> [here|https://gist.github.com/borodatu/dcf9248786b16a509fbea476f7f01e82]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
