James Hartig all time


 1 Collaborator
Mick Semb Wever

 7 Patch  1 Review
6b2b5101be5dfcf6803cc519da24ce9e77d4aac9, e46c11a3bc1e8c167f7ed8a6e645367906cfba31, f482cf60ff5892e4b25ae55a4659a3861343ae4b, 56d43d598ab6e840b2a49c9e1bedb1c8210f3472, d73e6b1002a78700df43099dd3de4f8c4e4d1ea9, c80d1b32903033df8db60e88d144b5638799b35a, 2274c8a9fab68a2be1c558b32eab48cb853163aa 685ed423ce2280b57407997573e260980264351e

685ed423ce2280b57407997573e260980264351e | Author: mck <mck@apache.org>
 | 2024-09-11 14:16:30+02:00

    Mark Matt Heath as agreed to donate contributions to ASF
    
     ref: #1751
    
     patch by Mick Semb Wever; reviewed by James Hartig for CASSANDRA-19723

6b2b5101be5dfcf6803cc519da24ce9e77d4aac9 | Author: James Hartig <me@jameshartig.com>
 | 2023-04-20 11:33:17-04:00

    Enable materialized views in cassandra and force proto=4

e46c11a3bc1e8c167f7ed8a6e645367906cfba31 | Author: James Hartig <me@jameshartig.com>
 | 2023-04-20 11:10:21-04:00

    Remove rpc_* from config

f482cf60ff5892e4b25ae55a4659a3861343ae4b | Author: James Hartig <me@jameshartig.com>
 | 2022-04-07 13:56:50-04:00

    Mark node as down if pool filling failed and pool is empty
    
    Fixes #1614

56d43d598ab6e840b2a49c9e1bedb1c8210f3472 | Author: James Hartig <me@jameshartig.com>
 | 2022-03-18 13:58:24-04:00

    Check HostFilter for control connection
    
    Fixes #1608

d73e6b1002a78700df43099dd3de4f8c4e4d1ea9 | Author: James Hartig <fastest963@gmail.com>
 | 2021-12-22 12:37:05-05:00

    Ignore events if they're disabled
    
    Fixes #1591
    Related to yugabyte/yugabyte-db#3090

c80d1b32903033df8db60e88d144b5638799b35a | Author: James Hartig <fastest963@gmail.com>
 | 2019-10-22 11:29:36-05:00

    Connect to peers synchronously only until ReadyPolicy is ready
    
    This checks if the HostSelectionPolicy implements ReadyPolicy and will stop
    blocking NewSession once Ready() returns true. If no ReadyPolicy was
    configured, the previous behavior will stay the same and it will block
    until all peers are connected. You can use SingleHostReadyPolicy to wrap an
    existing HostSelectionPolicy and return ready once a single host has
    connected.
    
    This also cleans up the behavior around reconnecting and adding new nodes.
    A node will be added as DOWN until it's connected then switch to UP.
    Additionally, connecting the control connection does not count as the node being UP since
    there's only the control connection.
    
    When a query is run on a specific connection (rather than the session) it
    will be pinned to that connection. This prevents inconsistencies or fetch
    to be called on the session and get a different connection. This also
    solves issues where the GetHosts call ends up trying to use the session
    before it's ready.
    
    Fixes #1365

2274c8a9fab68a2be1c558b32eab48cb853163aa | Author: James Hartig <fastest963@gmail.com>
 | 2019-10-15 11:20:34-04:00

    Don't send connectAddress when looping over peers (#1362)
    
    This ends up associating the wrong connectAddress with each peer since we
    don't overwrite connectAddress inside hostInfoFromMap.
    
    Fixes #1361