| Age | Commit message (Expand) | Author |
|---|
| 2010-08-12 | clean up membership application callback mod | Adam Kocoloski |
| 2010-08-12 | add distinct membership http req handler to membership app, not showroom | Brad Anderson |
| 2010-08-12 | patch up bugzid 10333, a cast on nodeup now sends gossip as well | Brad Anderson |
| 2010-08-12 | add membership events back into mem3, and add dbs handlers | Brad Anderson |
| 2010-08-12 | skip non-membership docs in dbs.couch | Adam Kocoloski |
| 2010-08-12 | atoms are not ejson keys, use bins for http compatibility | Adam Kocoloski |
| 2010-08-12 | better handling of repeated create_db attempts. BugzID 10300 | Adam Kocoloski |
| 2010-08-12 | mem3:nodes() does the trick here | Adam Kocoloski |
| 2010-08-12 | remove hash_int(), just convert hash() to int inline | Adam Kocoloski |
| 2010-08-12 | remove unused code | Adam Kocoloski |
| 2010-08-12 | more detailed report for mem3:states() | Adam Kocoloski |
| 2010-08-12 | faster+simpler get_pingnode_state | Adam Kocoloski |
| 2010-08-12 | missed a lib_misc function that vector clocks used, during dynomite -> member... | Brad Anderson |
| 2010-08-12 | dbs database creation moved to membership app | Brad Anderson |
| 2010-08-12 | throw an error in all_parts if DB does not exist | Adam Kocoloski |
| 2010-08-12 | forgot to start the db children | Brad Anderson |
| 2010-05-28 | begin move of dynomite to membership | Brad Anderson |
| 2010-05-28 | remove dynomite cruft | Brad Anderson |
| 2010-05-27 | open_doc call in fabric | Adam Kocoloski |
| 2010-05-27 | allow for single-node cluster 'init' | Brad Anderson |
| 2010-05-27 | replace #part with #shard | Adam Kocoloski |
| 2010-05-26 | change {N,P} fullmap over to #part{} record for future partition-splitting ho... | Brad Anderson |
| 2010-05-25 | all_databases now working with/without Customer param, types moved to mem.hrl... | Brad Anderson |
| 2010-05-24 | http admin call for join 'replace' | Brad Anderson |
| 2010-05-24 | comments | Brad Anderson |
| 2010-05-24 | some mem3 docs/specs | Brad Anderson |
| 2010-05-20 | add another clause for join-replace | Brad Anderson |
| 2010-05-19 | mem3 code for node replacement | Brad Anderson |
| 2010-05-19 | remove all vector clocks from #doc, and code to go with it. Not sure if thin... | Brad Anderson |
| 2010-05-14 | removing more of dynomite deps from mem3 | Brad Anderson |
| 2010-05-12 | create_db quorum check done, as well as initial install_fullmap function | Brad Anderson |
| 2010-05-11 | move partitions into showroom, now that there's not much left of dynomite code | Brad Anderson |
| 2010-05-09 | undo last commit, let's leave cluster_ops alone, make a new showroom module | Brad Anderson |
| 2010-05-09 | change cluster_ops:all_parts to accept the list of NodeParts | Brad Anderson |
| 2010-05-09 | work on create_db functionality, can now generate fullmap for a db based on i... | Brad Anderson |
| 2010-05-09 | disable some partition calls, and results from http admin layer | Brad Anderson |
| 2010-05-09 | add 0.5sec sleep to ping_all, seems to have solved 'init' bug picked up by in... | Brad Anderson |
| 2010-05-09 | switch join types to init,join,replace,leave and add first json-rpc handler f... | Brad Anderson |
| 2010-05-09 | fix bug in get_remote_states | Brad Anderson |
| 2010-05-09 | dang, large commit. | Brad Anderson |
| 2010-05-09 | reworking gossip, BugzID 10069 | Brad Anderson |
| 2010-05-09 | remote node should receive gossip if their clock was less than local | Brad Anderson |
| 2010-05-09 | dbg and doc | Brad Anderson |
| 2010-05-09 | gossip handling revamped, BugzID 10068 | Brad Anderson |
| 2010-05-09 | writing state to disk now, and handle_init installs disk state, if vector clo... | Brad Anderson |
| 2010-05-09 | tested gossip with two nodes and a jointype of 'first', added pinging of node... | Brad Anderson |
| 2010-05-09 | removed pinging of nodes based on _cluster endpoint | Brad Anderson |
| 2010-05-09 | added mem3:nodes() back in, and fixed some gen_server calls to return {ok, Va... | Brad Anderson |
| 2010-05-09 | changed test in Args to have a test node name for executing node. First cut ... | Brad Anderson |
| 2010-05-09 | not using 'cache' field in #mem, now that we're using mochiglobal | Brad Anderson |
