Metacat

handle "BIN" objects so as to avoid repeated calls to lookup the non-existent ObjectFormat

do not wait for SM generation to complete during the upgrade -- this way the web UI wont hang for days. the process sets a metacat property when it is complete.

Fixed a bug the a hyper-link included the username/password input fields.

use RC-2 DataONE jars -- these are built from trunk still, but include the next tag naming convention

do not shutdown hazelcast -- it needs to be running after the upgrade process so that Metacat actually works.
I think the newer version of HZ makes it so the threads are all released as needed.
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5572

Commenting out the parts of the upgrade script that started to refer to EXID. At this point, the registration of EZID identifiers will be done out-of-band with respect to the upgrade.

use plain String parameter for {pid} instead of XML serialization of it.

remove {pid} from POST URL on CN.registerSystemMetadata()
https://redmine.dataone.org/issues/2284

remove {pid} from POST URL on CN.create()
https://redmine.dataone.org/issues/2284

remove {pid} from POST URL on MN.create()
https://redmine.dataone.org/issues/2284

catch cases where the previous/next revision of objects have not had system metadata generated yet

create system metadata object if it wasn't found in HZ

adjust the width of the label suffix.

process systemMetadata from the docInfo string before writing to the database so that we guarantee guid-docid mapping exists before attempting to look it up.

Adjust the column width of the search result.

upgrade to hazelcast 1.9.4.6 so that threadpools are released when not needed (http://code.google.com/p/hazelcast/issues/detail?id=765).
include ant target to run a specific main class (mostly for debugging)

use File.deleteOnExit() not a half hour timer thread to do it.

multithreaded implementation for processing docids for system metadata generation.
need to investigate ant/junit running that deadlocks hazelcast (config?)

additional logging of the config file being used - seem to have thread locking on the xmlConfig use when running under ant/junit

calculate object size using the size on the file system rather than re-reading as an input stream.
Now only EML document bytes will be read twice: once for the checksum and again for parsing out datapackage details

system metadata generation optionally skips entries that have already been generated (data size, checksum) but allows the latest EML that describes them to have the last word on object format

remove DML for parsing -- the D1 EML parser still uses DOM, so this may not be too big of a perfromance improvement

test harness for running system metadata generation outside of the upgrade process

include comment about KNB estimated time to run during upgrade:
Total time: 20 minutes 58 seconds

only attempt to update date-like nodedata values.

use "test" to exercise upgrade code on staging DB.

include generate system metadata upgrade in the success flag

more clean up - reuse prepared statement for data update

look up nodedata values first, then update each one - trying to avoid out of memory exception.

eliminate the cross product that occurred when updating xml_access with a join

rollback processing Error change -- creates a loop on error. ugh

report processing errors after exceptions have been caught and recorded, otherwise the web UI is blank and there is no clue what happened unless you look in the logs.

semicolons!

fix a bug in MNodeService.replicate() where the checksum value was being compared to the computed checksum object, not its value.

use a temporary table to calculate the maximum revision for a given docid and use that when setting the accessfileid during upgrade. the query plan for the all-in-one statement must be brutal as it's been running for 4 hours at this point....

do not insert duplicate GUID entries when adding rows from the xml_revisions table

add "IF EXISTS" clause to identifier table drop in case it does not exist on the given deployment (as is the case on the KNB)

use UTC serialization for log entries so that the timestamp, not just the date, is preserved
https://redmine.dataone.org/issues/2257

edited the configuration docs, swapped in new fully-configured screen shot
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5516

Update the MNodeServiceTest to test the validity of the node document returned by getCapabilities() by parsing it with the TypeMarshaller.

Update the D1Admin class to set the dataone.contactSubject property. I've added the property to the http request to be added to the JSP form, but for now am setting the property using the dataone.subject field value. Not sure if we want to expose the contact subject in the form yet or not.

marked the properties that are only editable manually in metacat.properties
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5516

Add the dataone.contactSubject property to be used when getCapabilities() returns the MN node instance.

In MN.getCapabilities(), the required contact subject was not being added to the node instance from the dataone properties. Add it in.

format xml/eml table -- added missing row delimiter
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5516

some minimal editing on replication configuration -- bulk of it had been done in the Word doc before rst conversion
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5516

generate ORE maps only once -- and persist the flag to the main backup properties so that subsequent Metacat upgrades remember this value.

use RC-1 Dataone jars

Added DOI generation to the 2.0.0 upgrade process. To succeed, this script must be run on a fresh 2.0.0 database, or on a 1.9.5 version database, as those are the only ways to get the needed foreign keys to be marked as deferrable. The identifier conversion must be turned on by setting correct properties in metacat.properties. See the comments in GenerateGlobalIdentifiers for details. By default, conversion is set to false in the properties file. If you want to convert an instance to use DOIs, be sure to set metacat.properties up BEFORE running through the Metacat configuration and database upgrade.

Refactoring classes that throw generic Exception class to throw their more specific subclasses so that new exceptions are not hidden behind generic messages. Makes debugging easier.

try to read the local document before making the localid->guid mapping (in cases where we fail to read the data locally like if it is referenced in an EML file but does not exist on this Metacat instance)

add note to enable ssl mod

-include instructions about enabling AJP connector in Tomcat config
-use postgres 8.4
-include hint to disable default Apache site (otherwise the knb config never gets used!)

include instructions for installing Sun's Java6 JDK since it is no longer in apt-get repos

Ensure we have the object and sysmeta params for MN.create(). We were getting a fatal SAX parsing error encapsulated in a ServiceFailure when a science metadata object param was null. Cut it off at the pass after parsing the MMP entity.

An example python script that uses the python client to loop through a list of
files, read them from disk, and insert them into metacat.

make it clear that the Apache config files are samples and may need to be modified for different servers
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5516

update system requirements to be more reasonable

use larger ("text") db field for guid in the xml_access.accessfileid column

use EML 2.1.1 tag as final tag for the schema
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5532

use RELEASE_EML_UTILS_1_0_0 for EML style sheets
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5532

comment our the Demis world map layer -- it was prompting for username/password to use the WMS

Use the Collections class from java.util.

Remove null field tests in the IdentifierManager class. Schema-level required fields are checked on serialization/deserialization using JibX during the REST resource handler classes. Other required fields are checked in MNodeService and CNodeService, higher in the stack.

For MNs that haven't set the archived flag to false on create(), set it here. Also, ensure that the CN sync code sets the authoritative and origin member node fields.

On MN.create(), set the archived flag to be false. This field isn't required in the schema, but is needed by the DataONE indexer once objects are sync'd.

use EML 2.1.1 RC4 tag before final tag (schema)
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5532

use final tag for building with utilities (tags/UTILITIES_1_1_0)
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5532

use final tag for building with ecogrid
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5532

-generate system meta for all docids, even those not originating on the server (replicas from the past)
-generate ORE docs and download remote data only for those documents that originated on this server being upgraded.
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5522

refactor generate system meta loop to the factory class -- to be reused in sysmeta and ORE generation
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5522

When managing obsoletes/obsoletedBy system metadata fields, set the archived flag to false initially, and set it to true on system metadata for objects that a revision obsoletes.

do NOT generate ORE maps or download data when we do the initial System Metadata generation -- this is deferred until D1 registration.

make more generic so that a custom list of IDs can be passed in.

check that the resourceMap (based on Id only) does not currently exist in the local metacat when generating OREs

insert OR update system metadata -- no need to do an update right after initial insert...

call the System Metadata generator during upgrade to 2.0.0

In IdentifierManager.updateSystemMetadata(), add a check for invalid system metadata (fields that throw a NullPointerException on access) to ensure that system metadata is populated correctly. Updated calling classes to handle the exception.

Fixed formatting to make fixed-width line much shorter so boxes fit on the page.

Removing obsoleted Admin Guide -- this guide is now maintained using Sphinx in docs/user/metacat -- the word document and PDF file are now obsolete.

Properly initialize the servlet context when starting alternate servlets, which makes sure that the configuration files have been loaded and config properties are available.

Removed link to obsolete PDF file.

Modifications to make the build more readily find locally installed versions of
sphinx-build, and to remove the unneccary 'html' directory; output of
admindoc is now written directly into build/docs.

Modified build to include documentation in the war file that is generated as
part of the build, if the 'documentation' target has been run before the
war target. This is not done automatically because its not clear if all
people will have the proper sphinx environment set up to build the...

Adapted the build to be able to generate the Sphinx Admin Guide, and to
better handle the copyright for Javadoc generation.

Added links to Javadoc. Tuned the layout a bit.

Completed first draft of Admin guide chapter on DataONE.

Handle SQLExceptions when trying to save system metadata locally.

Convert SQLExceptions to RuntimeExceptions for Hazelcast MapStore operations.

In IdentifierManager, throw SQLExceptions rather than just logging them, and let them be handled higher up in the stack.

use new endpoint/method:
http://mule1.dataone.org/ArchitectureDocs-current/apis/CN_APIs.html#CNReplication.deleteReplicationMetadata

use PUT /obsoletedBy/{pid} for CNCore.setObsoletedBy per our discussion today

Keep the hzIdentifiers set in sync with the Metacat systemmetadata table. If entries are added/updated in the hzSystemMetadata map, make sure the identifier is in the set. If (for some administrative reason) the entry is removed, remove the identifier from the set. This usually doesn't happen.

When loading all keys from Metacat into the hzSystemMetadata map, also load identifiers into the hzIdentifiers set if they are not already there. Although entries may be evicted from the map, the list of identifiers will remain. The list will have a fairly small memory footprint since it's just identifiers.

Add support for the distributed Set of unique identifiers in the storage cluster called 'hzIdentifiers'. This set is a persistent total list of all identifiers (even when entries in the hzSystemMetadata map are evicted). It reflects the state of the identifiers in the postgresql systemmetadata table, but is distributed across the cluster. Add the getIdentifiers() method, which returns the ISet of identifiers.

Add the dataone.hazelcast.storageCluster.identifiersSet property that defines the name of the distributed set of unique DataONE identifiers (called 'hzIdentifiers'). This ISet can't be configured in the hazelcast.xml file (only maps and queues can).

Continued authoring the description of DataONE in Metacat. More to come.

include new methods needed for replication (in new d1 jars)
https://redmine.dataone.org/issues/2203

add method: setObsoletedBy (https://redmine.dataone.org/issues/2185)
augement new method: deleteReplicationMetadata