The University of California, San Diego
UCSD

{5} Assigned, Active Tickets by Owner (Full Description) (8 matches)

List tickets assigned, group by ticket owner. This report demonstrates the use of full-row display.

mmcnett

Ticket Summary Component Milestone Type Created
Description
#28 Additional API/ush functionality client_api enhancement 01/09/07

This may be an ongoing ticket for a while. Here's the current list:

  • hibernate Hibernate a VM. This was working, but needs a slight tweak now after some code refactoring. Should be done soon.
  • hibdel Delete a hibernated VM image
  • list -l Detailed list of VMs. This should also replace node_status
  • vmm_list -l Detailed list of VMMs. This should also replace vmm_status
#34 lost (invalid) domain check global Release-0.4 enhancement 04/06/07

When the Controller receives an error that a domain it's attempting to manipulate is not on the lnm where it's records indicate, a check should be performed to see if it's actually running somewhere else. Hopefully, this will fix a rare inconsistency problem where VMs with the same name have been running on different lnms (where one of the VMs is not being reported). Hopefully, better LNM state checking/reporting will help with this as well.

#31 Cluster based callback scheme global refactor 04/05/07

The current callback scheme for the client api is VM based. It should be cluster based since clients won't always be interested in events for all of a users VMs. This will also be necessary to support Cacti.

#18 Basic Usher decision engine module plugin request 08/26/06

Basic decision engine for initial Usher release.

#19 Monitoring plugin request 08/26/06

Monitoring for input into Usher decision engine.

#36 epydoc friendly docstrings global task 04/07/07

The docstrings need to be brought up to date and made more epydoc friendly. With lots of code changes (and a bit of laziness), it's been difficult to keep up with this. This does, however, need to be remedied.

#38 dns and other cleanups udns task 04/29/07

We need automatic cleanup of stale dns and rrdns entries at usherctrl startup. It's possible that VM crashes not properly handled by the controller will leave these hanging around. The controller can and should check for these and clean up dns when it starts.

#39 start placement option plugin request 05/04/07

Many users have asked for an option to specify initial placement. E.g. least loaded, evenly spread (1 per physical node, 2 per node, etc), random, etc. This should be an option to start and the controller can choose to honor or disregard the request as it sees fit.

Note: See TracReports for help on using and creating reports.