> On March 28, 2014, 2:53 p.m., kturner wrote: > > docs/src/main/resources/design/ACCUMULO-378-design.mdtext, line 157 > > <https://reviews.apache.org/r/19790/diff/1/?file=539855#file539855line157> > > > > Are you thinking a FATE operation per file? FATE uses zookeeper, and > > zookeeper keeps everything in memory. > > Josh Elser wrote: > Not sure. Using FATE when appropriate is mostly what I was thinking of > right now - I don't have explicit examples of where we would want to use > FATE. The obvious place is that we don't want multiple hosts sending the same > data more than once, but we also want to make sure we re-send data that > failed to send the first time around. > > Some more thought is needed here, I believe.
The ongoing work on replication reminds me that an upgrade can't happen while a FATE operation is in flight. So, as the design here develops, that's something to keep in mind. - Bill ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/19790/#review38927 ----------------------------------------------------------- On March 28, 2014, 1:54 p.m., kturner wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/19790/ > ----------------------------------------------------------- > > (Updated March 28, 2014, 1:54 p.m.) > > > Review request for accumulo. > > > Bugs: ACCUMULO-378 > https://issues.apache.org/jira/browse/ACCUMULO-378 > > > Repository: accumulo > > > Description > ------- > > ACCUMULO-378 Design document. Posting for review here, not meant for commit. > Final version of document should be posted on issue. > > > Diffs > ----- > > docs/src/main/resources/design/ACCUMULO-378-design.mdtext PRE-CREATION > > Diff: https://reviews.apache.org/r/19790/diff/ > > > Testing > ------- > > > Thanks, > > kturner > >
