• Francesc Guasch's avatar
    #984 nodes (#1001) · f4cc0072
    Francesc Guasch authored
    * wip(cli): use faster machine listing
    
    issue #984
    
    * feature(backend); set process priorities
    
    issue #984
    
    * wip(backend): cache connections and check free memory
    
    Now Virtual Manager connections are cached.
    When balancing VMs, check if machine memory fits
    
    issue #984
    
    * wip(backend): ignores disabled nodes, disable when one fails
    
    When a node is disabled, try not to use it.
    When a node fails, set disabled.
    Use process priorities.
    Improve MAC address generation on KVM
    
    * fix(KVM): don't reset the password if machine up
    
    issue #955
    
    * wip(backend): remove forkmanager
    
    We handle forks ourselves so we can tweak this more.
    ie we can allow more process with more nodes.
    
    issue #984
    
    * test(nodes): check remote remove base
    
    issue #984
    
    * test(nodes): migrate without CD and fix volume names
    
    issue #984
    
    * test(download): give time process to start
    
    * test(nodes): check base and volatile clones
    
    issue #984
    
    * wip(nodes): cached nodes and allow clone when preparing base
    
    Check the cached nodes are there and not renamed
    When preparing a remote base allow to clone the machine
    
    issue #984
    
    * refactor(KVM): disable nodes and better UUID
    
    creating random uuids sometimes failed
    
    issue #984
    
    * wip(requests): kill request if pid not found
    
    issue #984
    
    * test(volumes): create mock volumes with different names
    
    issue #984
    
    * refactor(nodes): clean remote bases
    
    issue #984
    
    * test(nodes): check volatile clones on remote
    
    * test(backend): properly create mock domains
    
    issue #984
    
    * refactor(KVM): cope with duplicated uuids
    
    issue #984
    
    * refactor(nodes): disable failed remote domains
    
    also cope with duplicated uuids on migration
    
    issue #984
    
    * refactor(nodes): improve balancing failover and volatile
    
    issue #984
    
    * test(pools): check unique uuid on creation
    
    issue #984
    
    * refactor(nodes): fixed remove base on remote down node
    
    issue #984
    
    * refactor(nodes): do not check active if node down
    
    issue #984
    
    * refactor(requests): allow shutdown timeout on refreshing vms
    
    May come useful one day, handy for testing by now
    
    issue #984
    
    * refactor(tests): removed debug
    
    * refactor(test): check down domain on disabled/down nodes
    
    issue #984
    
    * refactor(backend): return all known remote ips
    
    In some cases may be more than 1 remote ip, probably
    when accessing from localhost
    
    issue #984
    
    * refactor(test): wait more for remote machines down
    
    issue #984
    
    * refactor(backend): use timeout shutdown only if defined
    
    issue #984
    
    * refactor(nodes): deal with down node on refresh
    
    issue #984
    
    * refactor(nodes): open machine on requested node
    
    issue #984
    
    * refactor(test): properly test down node
    
    issue #984
    
    * refactor(test): new request ordering makes this test obsolete
    
    issue #984
    
    * refactor(test): make sure test node starts ok
    
    issue #984
    
    * refactor(backend): deal with is_active failing
    
    issue #984
    
    * refactor(nodes): fallback on local machine if remote fails
    
    issue #984
    
    * fix(requests): create instead create_domain
    
    issue #984
    
    * refactor(nodes): connect on is_alive cold deep recurse
    
    Also private method to clean cache
    
    issue #984
    
    * refactor(kvm): removed annoying debug message
    
    issue #984
    f4cc0072