Confused Confused Confused ...How to start with Fedora 4, instructions for Fedora 3 users?

[Jeevan Patnaik]

Hmm...I had used Fedora 3 which was already installed and setup. 

Now, my responsibility is to evaluate Fedora 4 for upgradation. I have installed. But couldn't understand single this how it is supposed to work. It's completely different.

The documentation for Fedora 3 was very clear where the architecture is explained and slowly the concepts. 

But for Fedora 4, there seems to be no head and tail. I couldn't get anything. 

Someone care to explain please. What I need to know and where shall I start?

At the end of the day, I need this application to create repository for my archived data.

[Julie Allinson]

Hi Jeevan,

This Fedora 4 training video might help? https://www.youtube.com/watch?v=9wbt0sPR-4E

Julie

--
You received this message because you are subscribed to the Google Groups "Fedora Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fedora-communi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Julie Allinson <julie.a...@york.ac.uk>
Technology Development Manager
Collections Services

Library & Archives
University of York
Heslington, York, YO10 5DD, UK
tel: ++44 (0) 1904 324083 

skype: j.allinson | gtalk: julie.a...@york.ac.uk

orcid: http://orcid.org/0000-0001-6459-6227
calendar: http://tinyurl.com/jal-gcal

email disclaimer: http://www.york.ac.uk/docs/disclaimer/email.htm

[David Wilcox]

Hi Jeevan,

Thanks for your message. If the Fedora 4 documentation is not clear then we certainly want to improve it! It would be helpful to know if you have specific questions that you could not find answers to, or types documentation you found lacking. Are you using (or planning to use) Fedora within a larger application framework (e.g. Hydra, Islandora, something custom) or on its own? 

In terms of where to start, I would begin with the current version of the Fedora 4 documentation [1]. The landing page contains an architecture diagram, and there are descriptions of the features, including the main differences between Fedora 3 and 4 [2]. If you are interested in migrating from Fedora 3 to 4, please check out the migration utility [3] and associated documentation.

Hopefully that helps - please let us know if further resources and/or explanations would be useful.

Regards,

David

[1] https://wiki.duraspace.org/display/FEDORA45/Fedora+4.5+Documentation

[2] https://wiki.duraspace.org/display/FEDORA45/Concept+Mapping+-+Fedora+3+to+4 

[3] https://github.com/fcrepo4-exts/migration-utils

--

David Wilcox
Fedora Product Manager
DuraSpace
dwi...@duraspace.org

[Jeevan Patnaik]

Hi Allison,

Thank you.

I haven't gone through the whole documentation yet. I only tried to understood from some pages. I had gone through the complete Fedora 3 documentation. So, when I go through this, there is no intimation in the starting pages that the architecture is changed when compared to Fedora 3 or any page like trying to compare both the versions.

My personal feeling is the documentation is not really good for novice. Little difficult to understand, not in very simple words or may be the explanation seems to be not continuous like a story ( I personally prefer that), and is very broken.

And thanks for the links and the video, I will go though those. I need to check how the framework of our old built is, I am not sure if it uses any external frameworks. But, I will try to go through those also once.

BTW, one quick question, so this migration from Fedora 3 to Fedora 4, is all about changing the existing metadata objects to Fedora 4 compatible objects right...and any challenges we may go through. The reason I am asking is I am less on time to have and troubleshoot any challenges now. 

Thank you,

Jeevan,

[Jeevan Patnaik]

Hi David,

I am trying to understand if there is any external framework used in my previous installation. Can you please help me with that.

For example, from the following admin console, can you say if it's using any external framework?

I have installed the version: Fedora 4.5.0 Web Application with Authorization and Audit with tomcat.

So, I noticed the following directories:

data:

com.arjuna.ats.arjuna.common.ObjectStoreEnvironmentBean.default.objectStoreDir  fcrepo.activemq.directory

com.arjuna.ats.arjuna.objectstore.objectStoreDir                                fcrepo.ispn.repo.cache

tomcat:

bin  conf  lib  LICENSE  logs  NOTICE  RELEASE-NOTES  RUNNING.txt  temp  webapps  work

I compared with the older fedora 3 installation (installed by someone else).

client:

bin  demo  fcrepo-client-admin-3.4.1.jar  lib  truststore

data:

activemq-data  datastreams  fedora-xacml-policies  objects  resourceIndex

docs:

license  userdocs

install: 

fedorawar  fedora.war  install.properties

pdp:

conf  doc  policies  policies-optional  xsd

server:

access  bin  config  fedora-internal-use  logs  management  schematron  status  truststore  xsd

tomcat:

bin  conf  lib  LICENSE  logs  NOTICE  RELEASE-NOTES  RUNNING.txt  temp  webapps  work

So, as you can see some directories are missing in my new installation: client utilities, server utilities and  pdp. install directory can be ignored, because it contains webapps for tomcat. pdp seems to be about policies, about which I have read a long time ago, now I forgot. 

So, no utilities are being given with the new fedora installation?

Anyhow, we have custom python client scripts for creating metadata and for uploading data, we use the webinterface. So, we haven't used client scripts before and also haven't used server scripts too.

Now, I am getting some idea, atleast that I need not worry about the installation.

But, the new interface, it looks very odd and confusing. This makes me think that may be there is external framework used for my old installation, because it was very better and clear, as you can see in the above picture. But see the new interface below:

I have compared the WEB-INF/web.xml in both Fedora 3 and Fedora 4, the web.xml

I still need to start reading the links that you have given. Will start today, I hope it should clear some confusions.

Thank you!

[Jeevan Patnaik]

Ok, correction!

The picture of old fedora webapp I have attached previously is admin console, it's not default homepage for the webapp.

The following is the default homepage. 

So, I've tried finding admin page for the new installation, but not there. 

Fedora 3:

$ls fedora

admin  images  index.html  jsp  META-INF  ri  WEB-INF

Fedora 4:

$ls fcrepo-webapp-plus-audit-4.5.0

favicon.ico  images  index.html  META-INF  static  WEB-INF

So, no separate admin page for fedora 4?

Thanks.

Unfortunately, this is my way of knowing things. First, I will try to explore and understand and later I go through documentation, if things got too complicated to understand :(

[Jeevan Patnaik]

Also, I have just found that the server directory that is under fedora 3 installation,also contains the authentication settings that are required to open the fedora app or the fedora Admin app.

Where were these managed in the new fedora version 4. 

I think in new version, it is managed by the tomcat authentication only. But I couldn't understand the rest of the documentation. 

[David Wilcox]

Hi Jeevan,

The old Fedora Web Administration client has been replaced by a more modern, responsive HTML interface. It uses content negotiation to present the resources in your repository in this manner, but you are effectively interacting with the REST-API when you manage your content using this interface. If you navigate to the top level of your repository you should be able to browse through all of your resources. Incidentally, if you want to view the main repository landing page you can do that by navigating to http://your.website/fcrepo/ (or whatever directory you installed Fedora in). You should see a page that tells you what version of Fedora 4 you’re using and a link to the REST-API endpoint.

As was mentioned, resources in Fedora 4 are no longer constructed using FOXML - instead, they are web resources with URIs and any number of RDF properties (stored as key-value pairs), which is what you see when you navigate to a resource as you have done in your screenshots. If you create a new resource you will see that it has many of the same kinds of properties that it would have had in Fedora 3 (e.g. who created the resource, when it was created, etc.). Since these properties are represented as RDF triples, you can create/edit/delete them using SPARQL-Update. The web interface has an ‘Update Properties’ box that serves this purpose. You can find some examples in the documentation [1]. However, if you have metadata in XML format you can still upload the XML file as a binary (what we used to call a Datastream).

Regarding authorization, this is fully pluggable. You can run no authorization, basic role-based authorization, XACML, or Web Access Control. Since this is a pluggable module it is not included in the basic Fedora 4 WAR file, but it has been packaged as fcrepo4-webapp-plus [2]. You can find more information on the different authorization options in the documentation [3].

Hopefully that helps - thanks for posting your questions to the mailing list and for proving feedback on the documentation. As you may know, Fedora is a community project that we are all a part of - you are welcome to not only ask questions but to edit or comment on the documentation directly. We’re always looking for ways to improve!

Regards,

David

[1] https://wiki.duraspace.org/display/FEDORA45/SPARQL+Recipes

[2] https://github.com/fcrepo4-exts/fcrepo-webapp-plus

[3] https://wiki.duraspace.org/display/FEDORA45/Authorization+Delegates

[Jeevan Patnaik]

Hi David,

Thank you very much for the detailed response. I think I need to do lot of home work. This is new to me.

I would really help in the documentation part when I start knowing these things. Now I get some idea of how to upload archive data (by changing type to 'binary'). I will take my time in exploring and going through the documentation. 

Thanks again.

Regards,

Jeevan.

[Joshua Westgard]

Jeevan, there is indeed a lot of homework involved in transitioning from previous Fedoras to Fedora 4, but in my opinion/experience the work is well worth it.  The new system is very powerful and well designed.  And there is no better way to learn than by tinkering.

Josh Westgard

University of Maryland

[Jeevan Patnaik]

Hi Josh,

yes, thanks for the heads up! I have started with this already: https://www.w3.org/TR/ldp/#ldpc

[Jeevan Patnaik]

Hi, 

So far good. I am at least now able to understand the rest, ldp and container concepts and the hierarchical metadata storage structure being used in the new fedora. 

For a while relaxing, I was tinkering again and what is the file extension for the version file?

I tried opening with notepad, it seems to be a binary file. So, how to make use of it and where it is helpful?

Example in the below image:

And adding to that why the URI of the child resources are like so random, I mean http:/example:8080/fcrepo/rest being the root container, the URI of it's child is somehow turned to be http:/example:8080/fcrepo/rest/ca/9f/62/f9/ca9f62f9-252c-4b79-a38a-0a2a356fce53

Thank you!

[Jeevan Patnaik]

Hi,

Ok, its clear. It seems we need to give our own id instead of auto generated id. 

But, Normal file system low level storage in fedora 3 used to be date based and it used to save data in the sub directories according to date and time.

But in this, I am seeing some thing like this:

 fcrepo.binary.directory]$ ls -l *

37:

total 4

drwxr-xr-x. 3 fedora fedora 4096 Mar 18 19:04 13

3b:

total 4

drwxr-xr-x. 3 fedora fedora 4096 Mar 18 19:04 e2

c0:

total 4

drwxr-xr-x. 3 fedora fedora 4096 Mar 17 15:04 c9

ce:

total 4

drwxr-xr-x. 3 fedora fedora 4096 Mar 17 15:04 85

trash:

total 8

drwxr-xr-x. 3 fedora fedora 4096 Mar 18 19:04 3b

drwxr-xr-x. 3 fedora fedora 4096 Mar 17 15:04 ce

In the old fedora, it used to be like this:

data]$ tree datastreams/

datastreams/

|-- 2010

|   |-- 1117

|   |   `-- 08

|   |       `-- 53

|   |           `-- test-ee_1+sig06-010_mac.jpg+sig06-010_mac.jpg.0

|   |-- 1129

|   |   |-- 13

|   |   |   |-- 58

|   |   |   |   |-- test-ee_64+metadata_creator.py+metadata_creator.py.0

|   |   |   |   `-- test-ee_65+metadata_creator.py+metadata_creator.py.0

|   |   |   `-- 59

|   |   |       `-- test-ee_66+metadata_creator.py+metadata_creator.py.0

How to get back to that data based storage structure. Any extensions we should use?

[Joshua Westgard]

Jeevan,

I had many of the same questions when first getting to know Fedora 4.  The auto-generated URI you are seeing is constructed from the UUID of the object and while you *could* override this by POSTing to containers of your own choosing directly, I would instead recommend that you just use the auto-generated URI.  This is in order to avoid performance problems and potential significant issues when moving content or switching out components from your application stack in the future.  

To put it simply, by avoiding introducing any semantics into the low level storage path, keeping the opaque identifier and building your semantics on top of that, you will give your future self the freedom to modify the low-level architecture much more easily.  This assumes that performance and future-proofing are things that matter for your use case (there are plenty of use cases where that might not matter).

There's a good explanation of the origin of these URIs in this thread:

https://groups.google.com/forum/#!topic/fedora-tech/2mCP7CqOGDw

For additional reading you might also search for "Pairtree" on the fedora-tech mailing list and the Duraspace wiki.

Josh

[Jeevan Patnaik]

Hi Joshua,

Thanks for the link. Also, one more thing I am worried about is that no search functionality for the metadata I can find in this new version. Any extension we need for this?

Regards,

Jeevan.

[Joshua Westgard]

Hi Jeevan,

That's right -- you need to set up one of the variety of external search integrations.  We are using Solr as our primary administrative index.  We also plan to expose a SPARQL endpoint.  

See https://wiki.duraspace.org/display/FEDORA45/External+Search for more information on your options.

Josh

[Jeevan Patnaik]

Hi Joshua,

Thanks again. I could see that in fedora downloads, there is one click application claiming that it's fully functional but with Jetty servlet container. I am also trying it in parallel hoping the search functionality is included. I will explore the Fesuki and Solari in parallel. The thing is I should complete this with in 3 days. So, that's the hurry. I should have been completed reading the documentation by now. But bloody bed bugs prevented me from doing any work this Sat and Sun. Anyhow, they are gone now after a big trojan war. Now, I find time again to sit with RESTful fedora, that made me restless..:P

Regards,

Jeevan.

[Stefano Cossu]

Hi Jeevan,
I am afraid that bed bugs are the only ones that the Fedora community is not able to help with... Jokes aside, if you are evaluating Fedora, you can try out a Vagrant build with all the components built in instead of figuring out the nitty-gritty details from the outset:

https://github.com/fcrepo4-exts/fcrepo4-vagrant

This build spins up a virtual machine with Fedora, auth, audit and triplestore index components enabled.

Best,
Stefano

> >>>>> <https://lh3.googleusercontent.com/-AgUkpAxdjgs/Vuvc72kwFoI/AAAAAAAAEvY/ZbbN3vUAE-wu3vc07zEifN5AGL2aTKwyQ/s1600/121.JPG>

> >>>>>>>>>> <https://lh3.googleusercontent.com/-QNL_9i3EAoA/Vul2LlsqejI/AAAAAAAAEuc/aXOkZXMBuN4x2JH0a0ij9z827PiCXYFKw/s1600/capture2.JPG>

> >>>>>>>>>> So, I've tried finding admin page for the new
> >>>>>>>>>> installation, but not there.
> >>>>>>>>>>
> >>>>>>>>>> Fedora 3:
> >>>>>>>>>> $ls fedora
> >>>>>>>>>> admin images index.html jsp META-INF ri WEB-INF
> >>>>>>>>>>
> >>>>>>>>>> Fedora 4:
> >>>>>>>>>> $ls fcrepo-webapp-plus-audit-4.5.0
> >>>>>>>>>> favicon.ico images index.html META-INF static WEB-INF
> >>>>>>>>>>
> >>>>>>>>>> So, no separate admin page for fedora 4?
> >>>>>>>>>>
> >>>>>>>>>> Thanks.
> >>>>>>>>>>
> >>>>>>>>>> Unfortunately, this is my way of knowing things. First, I
> >>>>>>>>>> will try to explore and understand and later I go through
> >>>>>>>>>> documentation, if things got too complicated to
> >>>>>>>>>> understand :(
> >>>>>>>>>>
> >>>>>>>>>> On Wednesday, March 16, 2016 at 6:17:13 PM UTC+5:30,
> >>>>>>>>>> Jeevan Patnaik wrote:
> >>>>>>>>>>>
> >>>>>>>>>>> Hi David,
> >>>>>>>>>>>
> >>>>>>>>>>> I am trying to understand if there is any external
> >>>>>>>>>>> framework used in my previous installation. Can you
> >>>>>>>>>>> please help me with that.
> >>>>>>>>>>>
> >>>>>>>>>>> For example, from the following admin console, can you
> >>>>>>>>>>> say if it's using any external framework?
> >>>>>>>>>>>
> >>>>>>>>>>>

> >>>>>>>>>>> <https://lh3.googleusercontent.com/-uWzD-n1u86I/Vuk2jNAMzZI/AAAAAAAAEt4/8ey3uAWWCWwLns7wHWA8sFbsIw5-6mK1w/s1600/Capture.JPG>

> >>>>>>>>>>> <https://lh3.googleusercontent.com/-jsrK_nnEUbk/VulTKux3XzI/AAAAAAAAEuI/bt2xeAD9bXAVVeGacUM8nHghkDfOoxeTg/s1600/Capture1.JPG>

--
Stefano Cossu
Director of Application Services, Collections

The Art Institute of Chicago
116 S. Michigan Avenue
Chicago, IL 60603

[Jeevan Patnaik]

Thanks Stefano,

I will start testing with it now.

And also thanks everyone for all the guidance. Really appreciate it! :)

Regards,

Jeevan,

[Jeevan Patnaik]

Hi Joshua,

I tried solr as per the instructions in the documentation.But getting the following error. I have no idea what it is. It seems I need to comment or uncomment some lines that are not mentioned in the page. Please help.

4931 [coreLoadExecutor-3-thread-1] ERROR org.apache.solr.core.CoreContainer  â Failed to load file /opt/HWEEArchive/solr/solr-4.6.0/example/solr/collection1/solrconfig.xml

4932 [coreLoadExecutor-3-thread-1] ERROR org.apache.solr.core.CoreContainer  â Unable to create core: collection1

org.apache.solr.common.SolrException: Could not load config file /opt/HWEEArchive/solr/solr-4.6.0/example/solr/collection1/solrconfig.xml

        at org.apache.solr.core.CoreContainer.createFromLocal(CoreContainer.java:525)

        at org.apache.solr.core.CoreContainer.create(CoreContainer.java:592)

        at org.apache.solr.core.CoreContainer$1.call(CoreContainer.java:271)

        at org.apache.solr.core.CoreContainer$1.call(CoreContainer.java:263)

        at java.util.concurrent.FutureTask.run(FutureTask.java:266)

        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)

        at java.util.concurrent.FutureTask.run(FutureTask.java:266)

        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

        at java.lang.Thread.run(Thread.java:745)

Caused by: org.apache.solr.common.SolrException: Found 2 configuration sections when at most 1 is allowed matching expression: schemaFactory

        at org.apache.solr.core.SolrConfig.loadPluginInfo(SolrConfig.java:284)

        at org.apache.solr.core.SolrConfig.<init>(SolrConfig.java:257)

        at org.apache.solr.core.CoreContainer.createFromLocal(CoreContainer.java:522)

        ... 9 more

4936 [coreLoadExecutor-3-thread-1] ERROR org.apache.solr.core.CoreContainer  â null:org.apache.solr.common.SolrException: Unable to create core: collection1

        at org.apache.solr.core.CoreContainer.recordAndThrow(CoreContainer.java:977)

        at org.apache.solr.core.CoreContainer.create(CoreContainer.java:601)

        at org.apache.solr.core.CoreContainer$1.call(CoreContainer.java:271)

        at org.apache.solr.core.CoreContainer$1.call(CoreContainer.java:263)

        at java.util.concurrent.FutureTask.run(FutureTask.java:266)

        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)

        at java.util.concurrent.FutureTask.run(FutureTask.java:266)

        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

        at java.lang.Thread.run(Thread.java:745)

Caused by: org.apache.solr.common.SolrException: Could not load config file /opt/HWEEArchive/solr/solr-4.6.0/example/solr/collection1/solrconfig.xml

        at org.apache.solr.core.CoreContainer.createFromLocal(CoreContainer.java:525)

        at org.apache.solr.core.CoreContainer.create(CoreContainer.java:592)

        ... 8 more

Caused by: org.apache.solr.common.SolrException: Found 2 configuration sections when at most 1 is allowed matching expression: schemaFactory

        at org.apache.solr.core.SolrConfig.loadPluginInfo(SolrConfig.java:284)

        at org.apache.solr.core.SolrConfig.<init>(SolrConfig.java:257)

        at org.apache.solr.core.CoreContainer.createFromLocal(CoreContainer.java:522)

        ... 9 more

4938 [main] INFO  org.apache.solr.servlet.SolrDispatchFilter  â user.dir=/opt/HWEEArchive/solr/solr-4.6.0/example

4939 [main] INFO  org.apache.solr.servlet.SolrDispatchFilter  â SolrDispatchFilter.init() done

5027 [main] INFO  org.eclipse.jetty.server.AbstractConnector  â Started SocketC...@0.0.0.0:8983

Regards,

Jeevan.

[Jeevan Patnaik]

I get this error, after I run this command:

java -jar start.jar

Instructions are given here: https://wiki.duraspace.org/display/FEDORA45/Solr+Indexing+Quick+Guide

[Joshua Westgard]

Jeevan,

Thanks for pointing out this issue.  There is an ambiguity in the instructions on the wiki.  When you uncomment the schemaFactory on lines 134-37, you must also comment out line 151 as follows: 

<!-- <schemaFactory class="ClassicIndexSchemaFactory"/> -->

This is indicated in the instructions if you look carefully, but would be easy to overlook. I'll add a note for future users.

If you modify your solrconfig.xml and stop/restart jetty, it should work.

Josh

...

[Jeevan Patnaik]

Hey many many thanks, I will check it out. That leaves only one pending task. The data Migration - don't know if it's going to be difficult. Need to check it.

Regards,
Jeevan.

[Jeevan Patnaik]

Hi Joshua,

After adding the comment, it worked. However, I think it also needs to be run in background.

After that, I do not understand from the documentation what to do?

Do I need to run that curl command. But there is no file solr-fields.json. 

example]# curl -X POST -H "Content-Type: application/json" --data-binary "@solr-fields.json" "http://localhost:8983/solr/schema/fields"

Warning: Couldn't read data from file "solr-fields.json", this makes an empty

Warning: POST.

341107 [qtp1725097945-18] ERROR org.apache.solr.rest.schema.FieldCollectionResource  â null:java.lang.NullPointerException

        at org.noggit.JSONParser.<init>(JSONParser.java:119)

Ofcourse, as the file is not there.

I have checked the compete solr directory:

[solr-4.6.0]# find . -name "solr-fields.json"

[solr-4.6.0]#

Also, the webpage comes like this: localhost:8983

Thank you!.

[Jeevan Patnaik]

I think, by now, you might have got idea that I am not aware of these concepts. So, I don't know what and why I am running these commands. So, I can't troubleshoot it myself.

But, I will definitely learn these after the implementation is done.

Regards,

Jeevan.

...

[Joshua Westgard]

Hi Jeevan, on the bright side, these all appear to be easy problems to fix:

On Thursday, March 24, 2016 at 11:36:16 PM UTC-4, Jeevan Patnaik wrote:

Hi Joshua,

After adding the comment, it worked. However, I think it also needs to be run in background.

I suppose you could do that.  In my case, I simply open another terminal for the curl commands.  Keeping each component running  in the front of a terminal allows you to see the communication between components as it is happening.

After that, I do not understand from the documentation what to do?

Do I need to run that curl command. But there is no file solr-fields.json. 

example]# curl -X POST -H "Content-Type: application/json" --data-binary "@solr-fields.json" "http://localhost:8983/solr/schema/fields"

Warning: Couldn't read data from file "solr-fields.json", this makes an empty

Warning: POST.

341107 [qtp1725097945-18] ERROR org.apache.solr.rest.schema.FieldCollectionResource  â null:java.lang.NullPointerException

        at org.noggit.JSONParser.<init>(JSONParser.java:119)

Ofcourse, as the file is not there.

 

You must create this file.  It is a simple text file with the contents given on the wiki:

[{"name":"uuid","type":"text_general","stored":"true","indexed":"true"}]

 

I have checked the compete solr directory:

[solr-4.6.0]# find . -name "solr-fields.json"

[solr-4.6.0]#

Also, the webpage comes like this: localhost:8983

The correct URL is localhost:8983/solr as indicated in the instructions.  You can also just click the link in the error page you posted.

[Jeevan Patnaik]

Oh yes, 

I forgot that curl is mentioned in this post. I have learnt it just after this post for a different reason. (For manipulating the repo from command line).

Oh my goodness. Thank you! :)

Can you help me for this question too if you can?: https://groups.google.com/forum/#!topic/fedora-tech/yt9fq-8oqbM

[Jeevan Patnaik]

Hi Joshua, 

Fails again.

error while performing maven clean install -DskipTests:

[root@eslxarv-test fcrepo-message-consumer]# mvn clean install -DskipTests

[INFO] Scanning for projects...

Downloading: http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/maven-metadata.xml

[WARNING] Could not transfer metadata org.fcrepo:fcrepo-parent:4.3.1-SNAPSHOT/maven-metadata.xml from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out

Downloading: http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/fcrepo-parent-4.3.1-SNAPSHOT.pom

and it's still running. I haven't stopped it. Hopefully more errors.

I often see this in most of the compilations. There seems to be issues in Repo URLS. :(

Any idea how to fix this?

Thanks!

Regards,

Jeevan.

[Jeevan Patnaik]

Hi,

here is the complete errors:

[fcrepo-message-consumer]# mvn clean install -DskipTests

[INFO] Scanning for projects...

Downloading: http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/maven-metadata.xml

[WARNING] Could not transfer metadata org.fcrepo:fcrepo-parent:4.3.1-SNAPSHOT/maven-metadata.xml from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out

Downloading: http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/fcrepo-parent-4.3.1-SNAPSHOT.pom

[ERROR] [ERROR] Some problems were encountered while processing the POMs:

[FATAL] Non-resolvable parent POM for org.fcrepo:fcrepo-message-consumer:[unknown-version]: Could not transfer artifact org.fcrepo:fcrepo-parent:pom:4.3.1-SNAPSHOT from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out and 'parent.relativePath' points at wrong local POM @ line 5, column 11

 @

[ERROR] The build could not read 1 project -> [Help 1]

[ERROR]

[ERROR]   The project org.fcrepo:fcrepo-message-consumer:[unknown-version] (/opt/HWEEArchive/fcrepo4-exts/fcrepo-message-consumer/pom.xml) has 1 error

[ERROR]     Non-resolvable parent POM for org.fcrepo:fcrepo-message-consumer:[unknown-version]: Could not transfer artifact org.fcrepo:fcrepo-parent:pom:4.3.1-SNAPSHOT from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out and 'parent.relativePath' points at wrong local POM @ line 5, column 11 -> [Help 2]

[ERROR]

[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.

[ERROR] Re-run Maven using the -X switch to enable full debug logging.

[ERROR]

[ERROR] For more information about the errors and possible solutions, please read the following articles:

[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/ProjectBuildingException

[ERROR] [Help 2] http://cwiki.apache.org/confluence/display/MAVEN/UnresolvableModelException

And the repo I have used is this: https://github.com/fcrepo4-exts/fcrepo-message-consumer.git

[Jeevan Patnaik]

Hi,

I think the error is because of http redirection to https. Because that's what I see when I open it in webpage.

Hmm..I have only one day now for deadline. Still the new implementation is suspense for me. But the good side is I learned a lot in this 1 week, since I started setting up fedora.

Regards,

Jeevan.

[Joshua Westgard]

Jeevan,

I'm afraid these errors are a little trickier to diagnose.  It looks like there are some network issues with your maven install command. It downloads a lot of files, and doesn't seem to be getting them all.  Beyond that it is difficult for me to say much.

As for the other error on the Tech list, I'm afraid I don't know the answer there either.  Hopefully someone else can chime in with better advice.

On the bright side, if you truly manage to get up and running with a Fedora 4 repository in production in the space of a single week, you will undoubtedly have set some sort of speed record.

Josh

[Jeevan Patnaik]

Hi Joshua,

Thanks for the reply.

I have tried adding -X for debugging at the end of maven install command. What I have understood is it is trying to find the xml locally first and if it doesn't get it it's trying to download remotely. Well then, I would download all and put it in the local repo. I believe that it will work. Lets see. Thanks.

[DEBUG] Reading user toolchains from /root/.m2/toolchains.xml

[DEBUG] Using local repository at /root/.m2/repository

[DEBUG] Using manager EnhancedLocalRepositoryManager with priority 10.0 for /root/.m2/repository

[INFO] Scanning for projects...

[DEBUG] Could not find metadata org.fcrepo:fcrepo-parent:4.3.1-SNAPSHOT/maven-metadata.xml in local (/root/.m2/repository)

[DEBUG] Using transporter WagonTransporter with priority -1.0 for http://oss.sonatype.org/content/repositories/snapshots

[DEBUG] Using connector BasicRepositoryConnector with priority 0.0 for http://oss.sonatype.org/content/repositories/snapshots

Downloading: http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/maven-metadata.xml

[DEBUG] Writing tracking file /root/.m2/repository/org/fcrepo/fcrepo-parent/4.3.1-SNAPSHOT/resolver-status.properties

[WARNING] Could not transfer metadata org.fcrepo:fcrepo-parent:4.3.1-SNAPSHOT/maven-metadata.xml from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out

org.eclipse.aether.transfer.MetadataTransferException: Could not transfer metadata org.fcrepo:fcrepo-parent:4.3.1-SNAPSHOT/maven-metadata.xml from/to snapshots-repo (http://oss.sonatype.org/content/repositories/snapshots): Connect to oss.sonatype.org:80 [oss.sonatype.org/107.23.166.173, oss.sonatype.org/52.22.249.229] failed: Connection timed out

Regards,

Jeevan.