perfSONAR User Q&A and Other Discussion

[Roman Lapacz <>]

Nicolas Simar wrote:
> Hi Roman,
>
> can you explain me the difference between the two? In particular in 
> which case you would go for one or the other.
>
> How does it solve the issue raised by Chris and Warren? Does it 
> install all with a single click?
I'm not sure about single click but a user could have all required 
things in one package and at least some of them could be configured (for 
example eXist could have an account and needed collection, Axis could be 
put in the application server).

The second package is what we have now. Just pS services.

Roman

> Cheers,
> Nicolas
>
> Roman Lapacz wrote:
> > Reading recent comments I believe that the next release should be 
> > represented by two packages:
> > - one which will contain pS services, application server (tomcat or 
> > jetty), eXist (it might be run in the application server without the 
> > need of GUI installation) or even JRE
> > - one which will contain only pS services as it is now
> >
> >
> > A user could choose which package is better for her/him.
> >
> >
> >
> > Chris Welti wrote:
> >
> > > Hi all,
> > >
> > > I'm afraid I have to agree with Warren here.
> > > I tried to get the java perfSONAR suite 1.0 running investing more 
> > > than a
> > >  day and I couldn't even get axis and exist running properly.
> > > Unless there is a really simple way of installing perfSONAR, you're
> > > not going to get this tool widely used.
> > > And by simple, i mean _really_ simple, like a debian package that can
> > > be installed via apt-get install, or something like make / make 
> > > install,
> > > or just a script that does ALL the installation tasks for you.
> > > The point here is that I want to be able to just execute ONE programm
> > > that does the installation for me. No worries about which paths to use
> > > for the different tools, version numbers of programs, setting 
> > > environment
> > > variables, configuring port numbers to use, and so on. There are way 
> > > to many steps involved in the installation process that can
> > > go wrong.
> > > The problem is of course that you are depending on 
> > > java/tomcat/axis/exist
> > > and you don't have control about it's configuration/installation.
> > > Well, that's the downside of not having a stand-alone application.
> > > The perl installation was much less of a hassle and is still running
> > >   
> >
> > probably you are talking about python  RRD MA :)
> >
> >
> >
> > Roman
> >
> >
> > > fine here at SWITCH.
> > >
> > > Regards,
> > > Chris
> > >
> > > Warren Matthews wrote:
> > >  
> > >
> > > > Hi All
> > > >
> > > > I hate to whine but Eric and Jeff say I should call it feedback and 
> > > > send
> > > > it to the list.
> > > >
> > > > We've had a terrible time with perfSONAR. I want to give up. 
> > > > Actually we
> > > > haven't even made it to trying to install perfSONAR. We haven't 
> > > > even got
> > > > the prerequisites installed yet.
> > > >
> > > > The real problem is we have no java expertise. I'm trying to get 
> > > > another
> > > > student but that'll be at least a few more weeks. The other problem is
> > > > we have to run RHEL. It is mandated by our director; Unfortunately our
> > > > site license does not include the application server channel, so we
> > > > don't get tomcat. Frankly, we're also not really interested in
> > > > understanding the packages and being experts on various server 
> > > > software
> > > > tools. We're network support staff, we want to measure and 
> > > > trouble-shoot
> > > > the network. Trying to get perfSONAR set up has been a real pain.
> > > >
> > > > Perhaps you will say we should wait for better documentation. But we
> > > > would like to be an early adopter because we need to offer monitoring
> > > > services for our global campus. I would like to use perfSONAR so in
> > > > future we can use data from other networks. Furthermore I claim most
> > > > network admins are not going to try as hard as we did. If it doesn't
> > > > just work they won't use it.
> > > >
> > > >
> > > > Warren Matthews, International Research Network Initiatives,
> > > > Georgia Institute of Technology.
> > > >
> > > > Adventure is a sign of incompetence. If you have an adventure, you're
> > > > doing something wrong. If you really plan things out then you don't 
> > > > have
> > > > adventures - The great polar explorer Vilhjalmur Stefanson.
> > > >
> > > >
> > > >
> > > >
> > > >     
> > >
> > >
> > >   


--

//  PSNC, Poland
//  phone: (+48 61) 858 20 24
//  http://www.man.poznan.pl