|
User
Joined: 24 Dec 2009
Posts: 10
Location: New Jersey, U.S.A.
|
Apologies for the lengthy lead-in. The 3 sample modules (a.erl, b.erl,
and d.erl) will highlight my observations. My questions are at the end of the post.
-module (a).
-export ([start/0]).
start()->
mnesia:create_schema ([node()]),
mnesia:start().
-module (b).
-export ([start/0]).
-record (dummy, {d1,d2}).
start()->
net_adm:ping (n1@T110),
mnesia:start(),
mnesia:change_config(extra_db_nodes, nodes()),
mnesia:add_table_copy(schema, node(), ram_copies),
mnesia:create_table (dummy,[{ram_copies, [node()]}, {attributes,
record_info (fields, dummy)}]).
-module (d).
-export ([start/0]).
-record (dummy, {d1,d2}).
start()->
net_adm:ping (n1@T110),
mnesia:start(),
mnesia:change_config(extra_db_nodes, nodes()),
mnesia:add_table_copy(schema, node(), ram_copies),
Result1 = mnesia:delete_table (dummy),
io:format ("Return val from mnesia:delete_table~p~n", [Result1]),
Result2 = mnesia:create_table (dummy, [{ram_copies, [node()]},
{attributes, record_info (fields, dummy)}]),
io:format ("Return val from mnesia:create_table~p~n", [Result2]).
I ran a.beam on a node named n1@T110, which created a schema. Then I
ran b.beam on a node named n2@T110, which created a ram copy of the
schema, and a table named dummy on n2@T110. Nothing strange so far. I
exited n2@T110 with q(), then started n3@T110 and ran d.beam. I got
this:
pm@T110:~/t1$ erl -sname n3
Erlang R15B (erts-5.9) [source] [64-bit] [smp:4:4] [async-threads:0]
[hipe] [kernel-poll:false]
Eshell V5.9 (abort with ^G)
(n3@T110)1> d:start().
Return val from mnesia:delete_table{aborted,{no_exists,dummy}}
Return val from mnesia:create_table{aborted,{already_exists,dummy}}
ok
(n3@T110)2>
Going back to n1@T110 and doing an mnesia:table_info (dummy, all)
resulted in this:
(n1@T110)2> mnesia:table_info (dummy, all).
[{access_mode,read_write},
{active_replicas,[]},
{all_nodes,[n2@T110]},
{arity,3},
{attributes,[d1,d2]},
{checkpoints,[]},
{commit_work,[]},
{cookie,{{1328,757112,486107},n2@T110}},
{cstruct,{cstruct,dummy,set,
[n2@T110],
[],[],0,read_write,false,[],[],false,dummy,
[d1,d2],
[],[],[],{...},...}},
{disc_copies,[]},
{disc_only_copies,[]},
{frag_properties,[]},
{index,[]},
{load_by_force,false},
{load_node,unknown},
{load_order,0},
{load_reason,unknown},
{local_content,false},
{majority,false},
{master_nodes,[]},
{memory,0},(n1@T110)3>
{ram_copies,[n2@T110]},
{record_name,dummy},
{record_validation,{dummy,3,set}},
{type,set},
{size,0},
{snmp,[]},
{storage_properties,...},
{...}|...]
(n1@T110)6> nodes().
[n3@T110]
(n1@T110)7>
Here are my questions:
1) From n3@T110, d.erl wasn't able to delete the "dummy" table because
mnesia doesn't think it exists. But I can't create a table name "dummy"
because Mnesia thinks it already exists. What am I doing wrong here?
2) Even though n2@T110 is already dead. mnesia:table_info still shows
n2@T110 having a ram copy of "dummy." Why is this?
Thanks in advance for your help.
Regards,
Patrick Moy |
|
|
Digg It
Del.icio.us
Reddit
Facebook
Stumble Upon
Technorati
