leave the pc19 in the list for longer

[spider.git] / Changes

diff --git a/Changes b/Changes
index 1c057be0a07f9641fec8cc7edcbd8abd25894409..5e3e1c049de97fe24518ac26b60d6b79cdb0d1d0 100644 (file)
--- a/Changes
+++ b/Changes
@@ -1,3 +1,26 @@
+19Jan03=======================================================================
+1. I have made a fundemental change to the way PC19s are dealt with as a 
+result of the discussions on the mailing list. From now on, only nodes that
+issue PC16s will have PC19s propagated. If they *ever* issue a PC16 then from
+then on they are considered as being 'connected' and thus routable. No PC16
+and you have no route. This should stop all passive links being routed. For
+the few cases where you know that that is a very small node and it rarely has
+users, then you can do a set/routepc19 <node> to enable PC19 propagation. In
+the even more rare case of you wanting to prevent PC19s being routed thru 
+your node, you can use unset/routepc19. THIS IS NOT A SUBSTITUTE FOR ROUTE
+FILTERING!!!!! [translators: I have added wpc16s, wpc16u, wpc19s, wpc19u]
+
+BTW the first time you connect again you will notice a separate PC19 go out
+for each 'new' PC16 that the node sees. This is normal and will go away the
+second time the system needs to issue a PC19.
+2. Alter the export_user command to remove users that have no useful info
+and have not been seen for more than (default) 1 year.
+17Jan03=======================================================================
+1. Fix problem with lines being left in the main input queue for a 
+disconnected/ing node.
+14Jan03=======================================================================
+1. Tidy up AUTOLOAD functions
+2. Make sure that empty PC16s are not sent on startup to a node.
 12Jan03=======================================================================
 1. add Czech language courtesy of Milan OK1XH
 06Jan03=======================================================================