greg/machi
1
0
Fork 0
Code Issues 11 Pull requests 1 Projects Releases Packages Wiki Activity Actions

WIP: bugfix in transition, just-in-case commit

Browse source
This commit is contained in:
Scott Lystig Fritchie 2015-07-06 23:56:29 +09:00
parent 9d4b4b1df6
commit 54b5014446
1 changed files with 61 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

74
src/machi_chain_manager1.erl
View file

@ -748,9 +748,9 @@ rank_projections(Projs, CurrentProj) ->
rank_projection(#projection_v1{upi=[]}, _MemberRank, _N) ->
-100;
rank_projection(#projection_v1{author_server=Author,
rank_projection(#projection_v1{author_server=_Author,
upi=UPI_list,
repairing=Repairing_list}, MemberRank, N) ->
repairing=Repairing_list}, _MemberRank, N) ->
%% It's possible that there's "cross-talk" across projection
%% stores. For example, we were a chain of [a,b], then the
%% administrator sets a's members_dict to include only a.
@ -1054,6 +1054,34 @@ react_to_env_A30(Retries, P_latest, LatestUnanimousP, _ReadExtra,
%% guarantee (yet?) that the [c,d] chain will be the UPI basis
%% for repairs when the partition is healed: the quickest author
%% after the healing will make that choice for everyone.
%%
%% 2015-07-06: Ha! This works, almost all of the time. But there
%% is a bug.
%%
%% The bug: if a repair has finished near the time that we fall
%% out of flapping mode and back to normal (one of the reasons
%% that we are here), then it's possible to have a situation like
%% this:
%% outer: {epoch,4638},{author,c},{upi,[e,c]},{repair,[d,a,b]}
%% inner: {epoch,4539},{author,e},{upi,[e,c,d]},{repair,[]}
%%
%% Code prior to today would simply use the inner projection and
%% only keep the outer's epoch number. However, if we do that,
%% then C100 will fail a sanity check: author e cannot add d to
%% the end of the UPI, only C is allowed to do that.
%%
%% After checking all 5 participants, they all agree with the
%% outer and inner shown above. But all 5 fail their C100
%% transition safety check, and so all 5 spin in an infinite loop,
%% cool!
%%
%% Fix for today: Send a signal (through a new func arg) to C100
%% that we're moving from inner to outer. If the
%% 'expected_author2' error is the only sanity check that fails at
%% C100, then that's OK, because: 1. We've lost track of the
%% author, so we can't satisfy the check 100% of the time. (We
%% have the option of picking the
%%
%% TODO: Perhaps that quickest author should consult all of the
%% other private stores, check their inner, and if there is a
@ -1063,12 +1091,19 @@ react_to_env_A30(Retries, P_latest, LatestUnanimousP, _ReadExtra,
%% in the down list, quite odd! Go investigate that.
P_inner2A = inner_projection_or_self(P_current),
ResetEpoch = P_newprop10#projection_v1.epoch_number,
ResetAuthor = P_current#projection_v1.author_server,
ClauseInfo2 = [{old_author, P_inner2A#projection_v1.author_server},
{reset_author, ResetAuthor},
{reset_epoch, ResetEpoch}],
P_inner2B =
P_inner2A#projection_v1{epoch_number=
P_newprop10#projection_v1.epoch_number,
dbg=ClauseInfo},
react_to_env_C100(P_inner2B, P_latest, S);
P_inner2A#projection_v1{epoch_number=ResetEpoch,
author_server=ResetAuthor,
dbg=ClauseInfo++ClauseInfo2},
ReactI = [{inner2b,machi_projection:make_summary(P_inner2B)}],
?REACT({a30, ?LINE, ReactI}),
io:format(user, "HEE30 ~w ~w ~w\n", [S#ch_mgr.name, self(), lists:reverse(get(react))]), timer:sleep(100),
react_to_env_C100(P_inner2B, P_latest, reset_loop_maybe, S);
true ->
?REACT({a30, ?LINE, []}),
react_to_env_A40(Retries, P_newprop10, P_latest,
@ -1246,7 +1281,7 @@ react_to_env_B10(Retries, P_newprop, P_latest, LatestUnanimousP,
]}),
put(b10_hack, false),
react_to_env_C100(P_newprop, P_latest, S);
react_to_env_C100(P_newprop, P_latest, undefined, S);
P_newprop_flap_count >= FlapLimit ->
%% I am flapping ... what else do I do?
@ -1339,13 +1374,10 @@ react_to_env_B10(Retries, P_newprop, P_latest, LatestUnanimousP,
react_to_env_C300(P_newprop, P_latest, S)
end.
react_to_env_C100(P_newprop, P_latest,
react_to_env_C100(P_newprop, P_latest, PerhapsReset,
#ch_mgr{name=MyName, proj=P_current}=S) ->
?REACT(c100),
I_am_UPI_in_newprop_p = lists:member(MyName, P_newprop#projection_v1.upi),
I_am_Repairing_in_latest_p = lists:member(MyName,
P_latest#projection_v1.repairing),
Sane = projection_transition_is_sane(P_current, P_latest, MyName),
%% put(xxx_hack, [{p_current, machi_projection:make_summary(P_current)},
%% {epoch_compare, P_latest#projection_v1.epoch_number > P_current#projection_v1.epoch_number},
@ -1356,11 +1388,28 @@ react_to_env_C100(P_newprop, P_latest,
_ when P_current#projection_v1.epoch_number == 0 ->
%% Epoch == 0 is reserved for first-time, just booting conditions.
?REACT({c100, ?LINE, [first_write]}),
erase(perhaps_reset_loop),
react_to_env_C110(P_latest, S);
true ->
?REACT({c100, ?LINE, [sane]}),
erase(perhaps_reset_loop),
react_to_env_C110(P_latest, S);
_AnyOtherReturnValue ->
io:format(user, "RESETLOOP: ~p ~w ~P", [MyName, get(perhaps_reset_loop), get(react), 70]),
if PerhapsReset == reset_loop_maybe ->
case get(perhaps_reset_loop) of
undefined ->
put(perhaps_reset_loop, 1);
X when X > 10 ->
Msg = lists:flatten(
io_lib:format("~P", [get(react), 200])),
exit({not_supposed_to_happen, ?MODULE, ?LINE, Msg});
X ->
put(perhaps_reset_loop, X+1)
end;
PerhapsReset == undefined ->
ok
end,
%% P_latest is not sane.
%% By process of elimination, P_newprop is best,
%% so let's write it.
@ -1389,7 +1438,6 @@ react_to_env_C110(P_latest, #ch_mgr{name=MyName} = S) ->
{_,_,C} = os:timestamp(),
MSec = trunc(C / 1000),
{HH,MM,SS} = time(),
io:format(user, "HEE120 ~w ~w ~P\n", [S#ch_mgr.name, self(), get(react), 150]),
case inner_projection_exists(P_latest2) of
false ->
case proplists:get_value(private_write_verbose, S#ch_mgr.opts) of