Ticket #697 (closed enhancement: fixed)

Opened 14 years ago

Last modified 14 years ago

Make installation on 10/half networks suck less

Reported by: jdreed Owned by:
Priority: low Milestone: IAP 2011
Component: -- Keywords:
Cc: Fixed in version:
Upstream bug:

Description

2-032 and 38-370 had > 50% failure rates for the upgrade, because they're still served by 2072s. The possibility of physical upgrades is being pursued, but assuming that fails, we should think about how to do something clever in those clusters. Perhaps special-casing the hostnames so that only =<3 machines upgrade per day?

Change History

comment:1 in reply to: ↑ description Changed 14 years ago by geofft

Replying to jdreed:

Perhaps special-casing the hostnames so that only =<3 machines upgrade per day?

Is it worth, independent of these particular mac having machines look at their own cluster in larvnet and not reinstall if too many other machines in the cluster are reinstalling, for the sake of not having a cluster go unavailable all at once?

comment:2 Changed 14 years ago by jdreed

  • Status changed from new to closed
  • Resolution set to fixed

I special-cased 38-370 in auto-upgrade, and 2-032 is now a 100Mbit network.

Note: See TracTickets for help on using tickets.