configuration build-server¶
Defines user configuration when building a casual server.
services¶
Defines which services the server has (and advertises on startup). The actual xatmi
conformant
function that is bound to the service can have a different name.
Each service can have different transaction
semantics:
type |
description |
---|---|
automatic |
join transaction if present else start a new transaction (default type) |
join |
join transaction if present else execute outside transaction |
atomic |
start a new transaction regardless |
none |
execute outside transaction regardless |
resources¶
Defines which xa
resources to link and use runtime. If a name is provided for a given
resource, then startup configuration phase will ask for resource configuration for that
given name. This is the preferred way, since it is a lot more explicit.
examples¶
Below follows examples in yaml
and json
(casual can also handle ini
and xml
)
yaml¶
---
server:
default:
service:
transaction: "join"
category: "some.category"
resources:
- key: "rm-mockup"
name: "resource-1"
note: "the runtime configuration for this resource is correlated with the name 'resource-1' - no group is needed for resource configuration"
services:
- name: "s1"
- name: "s2"
transaction: "auto"
- name: "s3"
function: "f3"
visibility: "undiscoverable"
- name: "s4"
function: "f4"
category: "some.other.category"
...
json¶
{
"server": {
"default": {
"service": {
"transaction": "join",
"category": "some.category"
}
},
"resources": [
{
"key": "rm-mockup",
"name": "resource-1",
"note": "the runtime configuration for this resource is correlated with the name 'resource-1' - no group is needed for resource configuration"
}
],
"services": [
{
"name": "s1"
},
{
"name": "s2",
"transaction": "auto"
},
{
"name": "s3",
"function": "f3",
"visibility": "undiscoverable"
},
{
"name": "s4",
"function": "f4",
"category": "some.other.category"
}
]
}
}