Import couchdb setup #567
Import couchdb setup #567
Conversation
The error tuple returned is not tested by the caller so it serves only to crash the binary comprehension in an obscure way.
PR: #7 PR-URL: apache/couchdb-setup#7 Reviewed-By: Robert Newson <rnewson@apache.org>
Adding nodes to a cluster fails if the node names (the `name` of
`name@hostname` in vm.args) is different from "couchdb".
The code currently infers this name from the port: "node1" if 15984,
"node2" if 25984, "node3" if 35984, "couchdb" otherwise. No other
possibility.
This is not suited for a production set-up, where multiple servers could
have different names.
This patch fixes this problem by adding an optional "name" option to the
"add_node" command:
POST /_cluster_setup
{
"action": "add_node",
"username": "root",
"password": "******",
"host": "production-server.com",
"port": 5984,
"name": "node5"
}
This fixes: COUCHDB-3119
* robertkowalski/2594-2598-number-of-nodes: fix wording use config:setineger/3 require nodecount on setup
* adrienverge/COUCHDB-3119: add_node: Don't fail if node name != "couchdb" or "node1"
* asf/salt-distribution: fix cluster setup: use same admin pq salt on all nodes
|
not sure why GitHub complains here, the branch merges cleanly locally. |
|
cc @davisp review plz <3 |
|
+1 Look fine to me. Though you'll want to update rebar.config.script after you merge this. As to the conflicts notice it sure would be nice to know what GitHub thinks is in conflict. Assuming it applies cleanly locally I'd just do that and call it good. It could even be that GitHub thinks its a conflict precisely because they have separate roots. |
|
@janl A reminder this needs to be updated for the latest changes to couchdb-setup that I just pushed. |
|
bump. we should do this. |
|
@wohali Says merged or am I missing something else? |
|
@davisp That's a different PR. This is definitely not done, otherwise we'd have a /src/setup directory in this repo, and we don't. |
|
@wohali Let that be a lesson for me to never read through PR activity before my first coffee. |
|
@jan any desire to get this onto master (and thus 2.1.x)? |
|
Superceded by #1241. |
8 participants
PR for #566