pyzabbix unexpected behavior

Question

I'm using pyzabbix for the first time. I'd like to get a history of CPU for a specific host. I feel I may be using/understanding the library incorrectly since I'm not getting the result I'm expecting. When I make the following call:

time_till = time.mktime(datetime.now().timetuple())
time_from = time_till - 60 * 60 * 4

history = zapi.history.get(hostids=["10632"],
    itemids=["78815"],
    time_from=time_from,
    time_till=time_till,
    output='extend',                                   
    limit=5,
    history=0,
    sortfield='clock', 
    sortorder='DESC'
)

I get the following result:

[
{
    "clock": "1520260023",
    "itemid": "78783",
    "ns": "353845414",
    "value": "100.0000"
},
{
    "clock": "1520260018",
    "itemid": "78778",
    "ns": "315473640",
    "value": "0.1677"
},
{
    "clock": "1520260018",
    "itemid": "78898",
    "ns": "321212433",
    "value": "40.7421"
},
{
    "clock": "1520260017",
    "itemid": "78777",
    "ns": "313120971",
    "value": "0.3103"
},
{
    "clock": "1520260016",
    "itemid": "78896",
    "ns": "308593447",
    "value": "99.6560"
}

]

Based on these results, I have two questions:

1) shouldn't all my results have "itemid": "78815" since this is what I requested in my API call?

2) what is the "ns" value?

Answer 1

Try omitting hostids from your request. The ns value is the nanosecond value in that second.