Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 219014

Summary:	[irc] [bot] Support rebooting of KOS-MOS without SSH-ing into ecf.eclipse.org
Product:	[RT] ECF	Reporter:	Remy Suen <remy.suen>
Component:	ecf.examples	Assignee:	Remy Suen <remy.suen>
Status:	RESOLVED WONTFIX	QA Contact:
Severity:	enhancement
Priority:	P3	CC:	slewis
Version:	unspecified	Keywords:	helpwanted
Target Milestone:	---
Hardware:	All
OS:	All
Whiteboard:

Description Remy Suen

2008-02-14 14:52:05 EST

KOS-MOS goes down once in a while and I am not always available to reboot her. As SSH access to ecf.eclipse.org is being restricted by ip, when I am on the move in school or elsewhere, I have no means of restarting KOS-MOS. This is often a source of frustration for KOS-MOS fans everywhere.

What we need is:
a) some way to reboot her through a browser (or something else, I'm open to suggestions)
b) naturally, we need some sort of authentication mechanism for this

Comment 1 Scott Lewis

2008-02-14 15:31:20 EST

We are also now running ecftcp://ecf.eclipse.org:3282/server as a direct Equinox application, so we may want to combine these into one Equinox process...or we could keep them apart.

Question:  Are you seeing KOS-MOS go down frequently on its own for some reason?  Is this something we can potentially fix in our code?  

Some thoughts/possibilities for admin UI:

A tomcat (or other servlet container) running on 8080 (which is already open on ecf.eclipse.org).  We've already got tomcat installed on 8080.  We could then just piggy back on tomcat admin auth.  We've only been using http, so I guess there would be no harm in continuing that.

We could then either:

1) run equinox(s) in servlet container using servlet bridge:
http://www.eclipse.org/equinox/server/http_in_container.php

2) write little servlets that start/stop/configure external Equinox process(es)

Comment 2 Remy Suen

2008-02-14 16:36:47 EST

(In reply to comment #1)
> Question:  Are you seeing KOS-MOS go down frequently on its own for some
> reason?  Is this something we can potentially fix in our code?

It's quite possible that it's just been caused by code. I haven't been logging exceptions properly but I'll do that some time this weekend.

Comment 3 Scott Lewis

2014-05-09 13:12:15 EDT

Resolving as wontfix due to no resources.  If committer or contributor resources become available, then please reopen.