This isn't expected behavior and I haven't been able to duplicate the behavior in my 0.9.1 install with an ETS backend. What version of Riak are you running and were there any errors in your sas-error.log? I would try restarting your node and seeing if the same behavior persists. If it does please send us your app.config and vm.args files.
Grant Schofield Developer Advocate Basho Technologies, Inc. On Apr 13, 2010, at 2:08 PM, Alexander Zhuravlev wrote: > On Tue, Apr 13, 2010 at 11:03 PM, Dan Reverri <[email protected]> wrote: >> This list_keys method is updated asynchronously so it may not immediately >> reflect the addition or removal of keys. The list_keys method should >> eventually reflect the deletion of the key. Can you try running list_keys >> after a delay? > > We have just run the following command: > > (r...@macbook-siden)9> C:list_keys(<<"aaa">>). > {ok,[<<"key1">>]} > > it still shows the same information (it has been about 5 hours since > we run the first C:list_keys command). > >> On Tue, Apr 13, 2010 at 11:13 AM, Alexander Zhuravlev >> <[email protected]> wrote: >>> >>> Hello, >>> >>> We are having an issue with riak_client:list_keys function. >>> The thing is that the function returns information about the keys which >>> were successfully deleted from a bucket. For example, please check >>> output from our tests below: >>> >>> (r...@macbook-siden)1> {ok, C} = >>> riak:client_connect('r...@macbook-siden'). >>> {ok,{riak_client,'r...@macbook-siden',<<7,122,127,92>>}} >>> (r...@macbook-siden)2> O0 = riak_object:new(<<"aaa">>, <<"key1">>, >>> "val1"). >>> {r_object,<<"aaa">>,<<"key1">>, >>> [{r_content,{dict,0,16,16,8,80,48, >>> {[],[],[],[],[],[],[],[],[],[],[],[],[],[],...}, >>> {{[],[],[],[],[],[],[],[],[],[],[],[],...}}}, >>> "val1"}], >>> [], >>> {dict,1,16,16,8,80,48, >>> {[],[],[],[],[],[],[],[],[],[],[],[],[],[],[],...}, >>> {{[],[],[],[],[],[],[],[],[],[],[],[],[],...}}}, >>> undefined} >>> (r...@macbook-siden)3> C:put(O0, 2). >>> ok >>> (r...@macbook-siden)4> C:list_keys(<<"aaa">>). >>> {ok,[<<"key1">>]} >>> (r...@macbook-siden)5> C:delete(<<"aaa">>, <<"key1">>, 2). >>> ok >>> (r...@macbook-siden)6> C:get(<<"aaa">>, <<"key1">>, 2). >>> {error,notfound} >>> (r...@macbook-siden)7> C:list_keys(<<"aaa">>). >>> {ok,[<<"key1">>]} >>> >>> backend = ets >>> riak 0.9.1 >>> >>> As you can see the key was successfully deleted (as it can be confirmed by >>> the output from riak_client:get call), but riak_client:list_keys still >>> reports that the key >>> exists in the bucket. Is this the intended behavior of did we hit a bug in >>> riak? >>> Please advise. >>> >>> Thanks. >>> >>> -- >>> Alexander Zhuravlev >>> >>> _______________________________________________ >>> riak-users mailing list >>> [email protected] >>> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com >> >> > > > > -- > Alexander Zhuravlev > > _______________________________________________ > riak-users mailing list > [email protected] > http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com _______________________________________________ riak-users mailing list [email protected] http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
