use map instead of list for reporting cpu/mem usage in NodeStatus UVE
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
Juniper Openstack | Status tracked in Trunk | |||||
R3.1 |
Fix Committed
|
Medium
|
Santosh Gupta | |||
Trunk |
Fix Committed
|
Medium
|
Santosh Gupta |
Bug Description
These are the proposed changes:
1.
Change the stats from list to map.
The client code doing the send will also have to change.
struct NodeStatus {
1: string name (key="")
2: optional bool deleted
…
6: optional list<process_
…
10: optional list<cpuinfo.
}
Should change to:
struct NodeStatus {
1: string name (key="")
2: optional bool deleted
…
// map key is partition name
6: optional map<string,
…
// map key is <module_
10: optional map<string,
}
2.
Remove “partition_name” from DiskPartitionUs
3.
Remove "module_id and “inst_id” from ProcessCpuInfo
before and after the changes in diff.