Permalink
Commits on May 14, 2013
  1. Coalesce multiple handoff completion events to reduce ring churn

    Introduces the new handoff helper gen_server that buffers completion
    events for up to 100ms, updating the ring once with all events that
    happened during the interval.
    jtuple committed May 12, 2013
  2. Randomize vnode inactivity timeout

    The goal is to prevent multiple vnodes timing out simultaneously
    and overloading the vnode manager with tons of messages.
    jtuple committed May 12, 2013
  3. Optimize riak_core_vnode_manager:compute_forwarding

    Change from O(n*m) to O(m) + O(n)
    jtuple committed May 12, 2013
  4. Optimize vnode manager

    Change certain uses of orddict to dict.
    
    Change update_never_started to only query ring indices when there
    are actually unknown modules.
    jtuple committed May 12, 2013
  5. Optimize the "up nodes" logic of riak_core_apl

    Change riak_core_apl to use an unordered list and lists:member rather
    than ordsets and ordsets:is_element for determining up nodes. This
    approach is much faster in practice. Both ordsets:is_element and
    lists:member are O(N); however, lists:member is a BIF and has a
    significantly lower constant factor. Likewise, this approach avoids
    the lists:usort performed by ordsets:from_list.
    jtuple committed May 14, 2013
  6. Add chashbin and bucket metadata

    jtuple committed May 10, 2013
Commits on May 10, 2013
  1. Hybrid ring/mochiglobal approach

    jtuple committed May 10, 2013
Commits on Jan 29, 2013
Commits on Jan 23, 2013
  1. Merge pull request #268 from basho/sdc-fix-urlmap

    Fix riak_core_wm_urlmap
    seancribbs committed Jan 23, 2013
  2. Fix riak_core_wm_urlmap.

    When the webmachine dispatch list was moved to ETS, this resource was
    not changed to reflect that, and so it still hit the application env
    for the dispatch list, resulting in a badmatch error because the key
    was not set.
    seancribbs committed Jan 23, 2013
Commits on Jan 3, 2013
  1. Fix bug in riak_core_util:rpc_every_member_ann

    The utility fun rpc_every_member_ann, uses by several status/console
    commands, was written assuming rpc:multicall returned results in the
    same order as the provided node list. This is an invalid assumption.
    This commit fixes rpc_every_member_ann as well as adds several
    additional utility functions to riak_core_util: pmap, multi_rpc,
    multi_rpc_ann, and multicall_ann.
    jtuple committed Jan 3, 2013
  2. Merge pull request #264 from basho/eas-fix-vnode-terminate-abnormal-r…

    …eason
    
    Make vnode terminate backend for any exit reason
    engelsanchez committed Jan 3, 2013
  3. Merge pull request #251 from basho/readme-rewrite

    rewriting revised readme in .md and removing .org version.
    seancribbs committed Jan 3, 2013
  4. Line wrap stuff.

    seancribbs committed Jan 3, 2013
Commits on Jan 2, 2013
  1. Wrapping pool shutdown in try/catch

    Making sure that we really always call terminate to avoid other issues
    like basho/riak_test#137
    engelsanchez committed Jan 2, 2013
  2. Make vnode terminate backend for any exit reason

    This should fix issue basho/riak_test#137, where partition repair was
    killing vnodes with reason kill_for_test and experiencing sporading
    bitcask data corruption.
    engelsanchez committed Jan 2, 2013
Commits on Dec 20, 2012
  1. Merge pull request #259 from basho/jdb-supervisor-order

    Adjust riak_core_sup child order for cleaner shutdown
    jtuple committed Dec 20, 2012