转载:http://www.cnblogs.com/gordonchao/archive/2011/01/06.html
1.net_adm:names()
2.record_info(fields, Tab)
3.Mod:module_info(attributes)
4.beam_lib:chunks("Mod.beam", [attributes]).
5. -vsn(1234).
-author({gordon, chao}).
6. mnesia
erl -mnesia dir '"/home/gordon/mnesia"' %启动erlang时指定目录
mnesia:system_info(directory).%得到mnesia表存放的目录
mnesia:info().
mnesia:change_table_frag(dictionary, {activate, []}).%激活mnesia分片
激活表格的分片:(a@localhost)25> mnesia:change_table_frag(dictionary, {activate, []}).
{atomic,ok}
查看分辨是否激活了:
(a@localhost)26> mnesia:table_info(dictionary, frag_properties). [{base_table,dictionary}, {foreign_key,undefined}, {hash_module,mnesia_frag_hash}, {hash_state,{hash_state,1,1,0,phash2}}, {n_fragments,1}, {node_pool,[a@localhost,b@localhost,c@localhost]}]
现在我们可以添加表的分片了我为了方便,先创建一个函数,这个函数的作用是获得表的frag_dist
(a@localhost)28> GetTableInfo = fun(Item) -> mnesia:table_info(dictionary, Item) end.
#Fun<erl_eval.6.35866844>
(a@localhost)29> GetFragNodes = fun()-> mnesia:activity(sync_dirty, GetTableInfo, [frag_dist], mnesia_frag) end.#Fun<erl_eval.20.67289768>
测试一下:
(a@localhost)30> GetFragNodes().
[{b@localhost,0},{c@localhost,0},{a@localhost,1}]
现在可以添加一个分片了,通过GetFragNodes返回的节点列表,mnesia可以负载均衡的把新的分片添加到相对空闲的节点上:
(a@localhost)35> mnesia:change_table_frag(dictionary, {add_frag, GetFragNodes()}).{atomic,ok}
(a@localhost)36> mnesia:table_info(dictionary, frag_properties). [{base_table,dictionary}, {foreign_key,undefined}, {hash_module,mnesia_frag_hash}, {hash_state,{hash_state,2,1,1,phash2}}, {n_fragments,2}, {node_pool,[a@localhost,b@localhost,c@localhost]}]
看到了 {n_fragments,2},增加到2个了。看看新的分片分派到哪个节点了:
(a@localhost)37> GetFragNodes().
[{c@localhost,0},{a@localhost,1},{b@localhost,1}]
看到了a节点和b节点各有一个分片,b节点的分片是新的。
7.erl -man erl
erl -man lists
8.工具类命令
tv:start()
debugger:start()
appmon:start()
code:clash()
pman:start()
9.whereis(Mod)
10.process_info(Pid, message_queue_len).