805Notes for Development Block 4.2
===============================
Previous epad notes: http://epad.dataone.org/2014-24-Block-3-4
G+ URL: https://plus.google.com/hangouts/_/nceas.ucsb.edu/d1-standup
Etherpad for V1, V2 API naming convention: http://epad.dataone.org/20140701-API-V2
Sprint Planning
---------------
Topics for this block:
- CN upgrades
- CCI 1.2.8 release
- UCSB first, then ORC, UNM
- CCI 1.3.0 release
- Migrate to new version of redmine: https://redmine.dataone.org/issues/5569
- libclient refactor and review (trunk only)
- v2 planning/implementation
Topics for CCI 1.4 - https://redmine.dataone.org/issues/5533
- ONEMercury UI changes for screencasts
- Refactor d1_processing/d1_cn_rest to minimize Hazelcast usage
- log aggregation
- will continue to use hazelcast for sysmeta map listener
synchronizationreplicationcn-rest webapp must communicate node changes/additions to d1_processing components
Incorporate JibX extensions into d1_common_java, release it(at CCI SB in July we discussed this as a part of v2 due to scope of changes that would be neededin v1 stack)Refactor CN components to use new d1_common_java changes- Schedule Meeting to determine the scope of changes to apply or add. (review empty vs null List/Map variables in JibX, add in Builder Methods for complex objects)
- Incorporate HTTPClient 4.2 into libclient (affects CN stack - focus on testing affected components)
- Replication Auditing w/splunk logging
David D
-------
20140721
- STAGE CN SYNCREPL:
- Sync replication working across all stage CNs
- Tested Friday - cn-stage-unm-1 not accepting or pushing replications; restarted slapd, replication activity began across all stage CNs
- Tested over the weekend - replication working across all stage CNs without any further prodding on my end
- TO DO: Ask coredev - is this good to call fixed, or do we need any further tests?
- SPLUNK CLUSTER
- worked on the cluster peers over the weekend, replication activity is now happening across all peer indexers at ORC, and replicated data is searchable by search head
- TO DO: Full list of to-dos can be found at https://redmine.dataone.org/issues/5521
need to bug Chris, Dave, et al re: whether we want to get further cluster VMs set up offsite (UNM/UCSB) VMs to be set up at UNM for offsite cluster peering
- REPLACEMENT TRAINING
- Hopefully should have a name for replacement by Tuesday; if not, Bruce will begin making further inquiries
- Sending out more documentation to ORNL JICS contact
20140718
- VM upgrades and migrations are complete on all but mn-orc-1
- script that modifies node status page is returning errors when I try to run it
- Splunk cluster work
Need to build my keys into subversion done- Syncrepl on stage CNs
- Now that stage CNs are on 12.04, need to check and see if this is still failing
20140716
- Working on Splunk cluster buildout
- Still need to bring ORC VMs down for various upgrades
- nearly all machines need to have their network adapters updated, the virtual hardware they're running on is no longer supported
- need to bring prod and stage MNs over to new host hardware
- some VMs need to be placed in the correct virtual network
20140714
- Brought down cn-orc-1 Friday, migrated to new HW, removed old network adapter/added new network adapter, tested good, brought back up Saturday
- Gathered up several ORC sysadmin docs and links to send off to the team lead at ORNL who will be in charge of my replacement
- needed to update several documents to reflect new hardware and VM management tools
- TO DO:
- work on Splunk cluster buildout
- several VMs at ORC are running on outdated VMware versions - need to make a list of affected VMs, test upgrade on a nonessential VM, schedule some downtime and roll out upgrade on remaining VMs
- this isn't strictly necessary in a mission-critical sense, but probably good to do this while we still have a sysadmin working full-time
Dave V
------
20140718
- Need to ensure that all VMs have correct entries in resolv.conf
- MN-UNM-1 is using an NFS mount which is *very* slow. Need to plan how to move away from that storage.
- Need to complete setup of the maven repo. Requesting that Rob work with Mark Servilla to get this sorted out.
- Seems likely that InCommon will move away from supporting ProtectNetwork as an identity provider due to a significant increase in membership fees. Need to evaluate whether DataONE will be negatively impacted by this change.
20140716
- Suggest adding ssh public keys to svn at:
/documents/Management/Users/${USER}/sshpublickey/
and moving gpg public keys to:
/documents/Management/Users/${USER}/gpgpublickey/
to help simplify server etc setup
20140714
- Setting up new version of redmine (2.4.2) + backlogs
- Configuring new document server for Phase II (Alfresco community edition)
- Scheduling work for Phase II project execution plan
Chris
-----
- 20140718
- continued work on v2 libclient refactor with Ben, Skye, Rob
- EDORA and RGD MN meeting
- having problems with dataone-approve-node in stage
- cjones@cn-stage-ucsb-1:bin$ sudo ./dataone-approve-node
- [sudo] password for cjones:
- Choose the number of the Certificate to use
- 0) urn:node:cnStageUCSB1.pem
- 1) urn_node_cnStageUCSB1.pem
- 0
- Jul 18, 2014 4:40:18 PM com.hazelcast.config.ClasspathXmlConfig
- INFO: Configuring Hazelcast from 'org/dataone/configuration/hazelcastClientConf.xml'.
- Exception in thread "main" java.lang.NoSuchMethodError: com.hazelcast.config.NetworkConfig.getPort()I
- at org.dataone.cn.hazelcast.ClientConfiguration.getPort(ClientConfiguration.java:68)
- at org.dataone.cn.hazelcast.ClientConfiguration.getLocalhost(ClientConfiguration.java:72)
- at org.dataone.cn.client.NodeApproval.<init>(NodeApproval.java:58)
- at org.dataone.cn.client.Main.main(Main.java:93)
- TODO: contact Pat @ urn:node:SC
- 20140716
- continued v2 work - Ben, Rob, Skye
- working on Metacat v2 impl
- IOE production cert
- 20140714
- Continued work on v2 impl with Ben
- Trimmed back /var/postgres-bak in production
Skye
-----
- 201400714
- Libclient http client work branch review.
- OneMercury UI changes as requested by Heather/Amber.
- 20140716
- Small cross browser fixing for onemercury
- Safari 'processing' spinner broken
- Paging control rendering in chrome does not handle page zoom
- Strategies for supporting differening API versions
- Support 1.3.0 testing
Peter
-----
- 20140714
- working on CCI 1.4
- help with testing CCI 1.3
Robert
------
- 20140714
- create doodle pool for refactoring
- work on 1.2.8 release
- branch 1.3.0 release for testing
- 20140716
- UNM and ORC are upgraded to 1.2.8
- Upgrade UCSB to 1.2.8
- look at tidy again
- 20140718
- UCSB upgraded to 1.2.8 and places as Active Master in DNS RR
- Looking at 1.3.0
Rob
---
- 20140716
- libclient_java:
- compiling comments from yesterday's review
- 20140718
- libclient_java
- committed changes to d1_common_java and d1_libclient_java branches to remove httpclient version restriction
- v2 refactoring design discussions and work
Roger
-----
- 20140718
- Continued work on ONEDrive Zotero integration.
- 20140716
- Adding install docs for ticket generator for Windows
- Fixing a small issue for the ticket generator for Windows
- Helping Laura with generator installation and use
- Started on Zotero integration for ONEDrive
Topics
------
- Scope of libclient software
- DECISION: libclient v1.x changes need to be backwards-compatible
- DECISION: libclient is primarily an external product for use by DataONE users/MN developers, etc.
- Is libclient and external or internal product?
- Dave: Everything that we do should be considered external
- Scope includes low-level and high-level functionality
- This refactor includes backwards incompatible changes
- Suggestion: break libclient into two products: one for external clients, one for the CN
- Against: we should be using th same codebase for the same functions
- Against: more code to maintain, duplication is possible
- For: more flexible code for client-side vs server-side software
- For backwards compatibility:
- Perhaps move D1Client back to org.dataone.client from org.dataone.client.itk
- move concrete MNode and CNode back to org.dataone.client, and change the interface locations
- v2 libclient will exist independently of v1, and will not be backwards-compatible
- D1RestClient is no longer in v1.4
- ObjectFormatCache has been moved
- We probably should incrementally make these changes: first, HTTPClient changes (v1.4), then refactor the API (v2.0)
- This may sacrifice some testing features in v1.4
DECISION: roll back HTTPClient to the 4.1.3 series in d1_common_java trunk- DECISION: move the libclient trunk pom to version 2.0-SNAPSHOT
- DECISION: move d1_common_java Mime/Multipart support into d1_libclient_java in v2
- remove/delete MultipartRequestHandler which is a deprecated mini HttpClient and causes the dependency in d1_common_java
- TODO: Rob - change the pom of d1_common_java v1.2 branch such that it doesn't restrict HttpClient version (v1.2.1) (resulting in new tag 1.2.x)
- TODO: Rob - change the pom d1_libclient_java from v1.3 branch that points to new d1_common_java version v1.2.x (resulting in new tag 1.3.x)