| Jim Archer 2005-09-07, 3:25 am |
| Hi All...
Well I have gone and screwed up my test cluster. I had a cluster running
that consisted of one origin feeding two subscribers, all on seperate
machines. One was connected with a private wire using fake net IPs while
the other was accross the Internet.
I issued a slonik script with this in it:
lock set (id =3D 1, origin =3D 1);
wait for event (origin =3D 1, confirmed =3D 2);
wait for event (origin =3D 1, confirmed =3D 3);
move set (id =3D 1, old origin =3D 1, new origin =3D 3);
wait for event (origin =3D 1, confirmed =3D 2);
wait for event (origin =3D 1, confirmed =3D 3);
This hung for quite a while so I aborted it. Now, none of the 3 nodes
thinks it is the origin.
Two questions. First, what is wrong with my script and second, is there =
a
way to gracefully recover from this condition? If this were to happen to
my production database I would be up the creek...
Thanks...
|