jael: re-enable ship-to-ship communication

Also fix bug in eth-watcher that didn't cancel outstanding threads when
config changes.

And set default rift for ourselves to 0.

pkg/arvo/app/eth-watcher.hoon

@@ -149,7 +149,7 @@
     =/  already  (~(has by dogs.state) path.poke)
     ~?  &(already restart)
       [dap.bowl 'overwriting existing watchdog on' path.poke]
-    =/  wait-cards
+    =/  wait-cards=(list card)
       ?:  already
         ~
       [(wait-shortcut path.poke now.bowl) ~]
@@ -162,7 +162,8 @@
           ==
         ~
       =/  =cage  [%spider-stop !>([u.running.u.dog &])]
-      [%pass [%starting path] %agent [our.bowl %spider] %poke cage]~
+      :_  ~
+      `card`[%pass [%starting path.poke] %agent [our.bowl %spider] %poke cage]
     =/  new-dog
       =/  dog=watchdog
         ?:  restart  *watchdog
@@ -172,11 +173,11 @@
         number  from.config.poke
       ==
     =.  dogs.state  (~(put by dogs.state) path.poke new-dog)
-    [wait-cards this]
+    [(weld wait-cards restart-cards) this]
   ::
       %clear
     =.  dogs.state  (~(del by dogs.state) path.poke)
-    [(weld wait-cards restart-cards) this]
+    [~ this]
   ==
 ::
 ::  +on-watch: subscribe & get initial subscription data

pkg/arvo/sys/vane/jael.hoon

@@ -257,14 +257,14 @@
       =.  sig.own.pki  sig.seed.tac
       ::  load our initial public key
       ::
+      =/  spon-ship=(unit ship)
+        =/  flopped-spon  (flop spon.tac)
+        ?~(flopped-spon ~ `ship.i.flopped-spon)
       =.  pos.zim.pki
         =/  cub  (nol:nu:crub:crypto key.seed.tac)
         %+  ~(put by pos.zim.pki)
           our
-        =/  spon-ship
-          =/  flopped-spon  (flop spon.tac)
-          ?~(flopped-spon ~ `ship.i.flopped-spon)
-        [1 lyf.seed.tac (my [lyf.seed.tac [1 pub:ex:cub]] ~) spon-ship]
+        [0 lyf.seed.tac (my [lyf.seed.tac [1 pub:ex:cub]] ~) spon-ship]
       ::  our initial private key
       ::
       =.  lyf.own.pki  lyf.seed.tac
@@ -307,8 +307,20 @@
       ::
       =.  +>.$  (poke-watch hen %azimuth-tracker nod.own.pki)
       =.  +>.$
-        %-  curd  =<  abet
-        (sources:~(feel su hen our pki etn) ~ [%| %azimuth-tracker])
+        ?-    (clan:title our)
+            %czar
+          %-  curd  =<  abet
+          (sources:~(feel su hen our pki etn) ~ [%| %azimuth-tracker])
+        ::
+            *
+          =.  +>.$
+            %-  curd  =<  abet
+            %+  sources:~(feel su hen our pki etn)
+              (silt (turn spon-points head))
+            [%| %azimuth-tracker]
+          %-  curd  =<  abet
+          (sources:~(feel su hen our pki etn) ~ [%& (need spon-ship)])
+        ==
       ::
       =.  moz
         %+  weld  moz
@@ -508,7 +520,7 @@
     ^+  +>
     ?-  hin
         [%a %done *]
-      ?~  error.hin  ~&(%ares-fine +>.$)
+      ?~  error.hin  +>.$
       ~&  [%done-bad tag.u.error.hin]
       %-  (slog tang.u.error.hin)
       ::TODO  fail:et

pkg/arvo/ted/ph/all.hoon

@@ -14,6 +14,7 @@
         %ph-child-sync
         %ph-breach-multiple
         %ph-breach-sudden
+        %ph-breach-hi-cousin
         %ph-hi-linnup-az
         %ph-moon-az
     ==

pkg/arvo/ted/ph/breach-hi-cousin.hoon

@@ -1,3 +1,7 @@
+::  This is useful to test that we properly hear about breaches outside
+::  our sponsorship tree.  We usually hear about these via ship-to-ship
+::  communication.
+::
 /-  spider
 /+  *ph-io
 =,  strand=strand:spider

pkg/arvo/ted/ph/breach-multiple.hoon

@@ -1,3 +1,5 @@
+::  This tests breaches of both parent and child in succession.
+::
 /-  spider
 /+  *ph-io
 =,  strand=strand:spider

pkg/arvo/ted/ph/breach-sudden.hoon

@@ -1,3 +1,7 @@
+::  This tests whether we can handle the case where our immediate
+::  sponsor dies without telling us about its breach, so we must hear
+::  about it from somewhere else.
+::
 /-  spider
 /+  *ph-io
 =,  strand=strand:spider

pkg/arvo/ted/ph/breach-sync.hoon

@@ -1,3 +1,5 @@
+::  This tests that syncs are correctly restarted after a breach
+::
 /-  spider
 /+  *ph-io
 =,  strand=strand:spider