Member Nodes

Hi David,

Thanks for the update on the Israel node. I will remove it from our tracking system. Other answers to your questions are inline below...

On Wed, Jul 17, 2013 at 6:10 AM, David Blankman dblankman1@gmail.com wrote:
Matt,

First, the Israel metacat is no longer in existence.

We will be discussing DataOne at the ILTER meeting in Korea in October and will come to a decision as to how we are going to work, e.g. will Japan, Malaysia, Australia (all have metacats) join independently or will
they replicate to a regional metacat hosted in Taiwan. Is there an advantage of one strategy versus another?

Yes, there are pros and cons of each. Advantages of running multiple nodes include 1) ability to share costs for storage, 2) ability to replicate data at multiple local nodes, and 3) keeping data closer to the users of the node, which can be important on some networks. Disadvantages include 1) multiple systems have to be created and maintained.

Am I correct that LTER-Europe can start out as just a member node and then later on replicate?

Yes. You can start by just providing metadata and exposing your data for download, and then later turn on replication services. Metacat has a switch in the configuration that allows you to do this.

Is DataOne at the point where we can replicate only LTER networks?

Yes, DataONE is, but there are issues. Each MN can determine which data sets they accept for replication. The process is that the Source MN flags a data set for replication, then the CN looks for a Target MN that can host the dataset, and then asks the Target MN to replicate the data, which the Target MN can either accept or decline. However, Metacat doesn't currently provide a mechanism for this -- it accepts all replication requests if it has space available. So you would have to make modifications to support this kind of a filter. We want to add more features for selecting which data to replicate and how much space to make available, but we haven't had the resources to implement this yet.

Also am I correct that we can replicate metadata only and be selective about the data we would be willing to replicate?

Yes, replication policies (as well as access policies) are done on a per-object basis, and can be separately specified for metadata and data objects; That said, all metadata gets replicated to all of the Coordinating Nodes, even though it may be access restricted.

For now, I will be the technical contact for the LTER-Europe DataOne node. We now have the metacat working along with Harvesting, so we are ready to start participating in DataOne. The other key person from
the system administration perspective is Mike Wilson (mw@ceh.ac.uk). He is probably the best person to contact about coordinating certificates and similar system administration issues.

Great, thanks. I will pass Mike's name on to our development team and we'll follow up.

Matt

David

David Blankman
Chair, ILTER Information Management Committee
Director, Information Management, Israel LTER

Formally announced 6/18/14.

9/4/18: from Christoph Wohner, Ecosystem Research & Environmental Information Management Christoph.Wohner@umweltbundesamt.at re: LTER-europe status (-1) as of 8/20/18

DEIMS is currently migrated to a new server and is undergoing some changes in terms of the software stack in use.

For the time being we turned off Metacat as we have deployed a new service that incorporates more and nodes and systems that DEIMS had.
However, we are currently still setting up a few things and are in touch with Wim Hugo wim@saeon.ac.za, who is responsible for the new data strategy of ILTER.
Therefore it might take a couple of weeks until it can be harvested again.