Get Device Loss And Latency History

[eroche]

Thanks @KH  for the fix on my other post!

I have question about the results. Does the following result constitute a loss? I am seeing packet loss on the uplink in the web dashboard in this time frame but the only thing I get back that isn't 0% loss is the None.

startTs: 2024-08-06T23:47:00Z
endTs: 2024-08-06T23:48:00Z
lossPercent: None
latencyMs: None
jitter: None
goodput: None

[sungod]

'None' means there is no data, rather than a zero value.

Can you post the exact way you are calling this, all the parameters/values you are using.

Especially the ip parameter, normally use 8.8.8.8 (unless you changed the value in Dashboard config.)

Btw for MX only, I prefer this call as it returns all devices/interfaces in the org...

https://developer.cisco.com/meraki/api-v1/get-organization-devices-uplinks-loss-and-latency/

[eroche]

Hi

Here are the parameters and example of what I am getting back.

serial = 'Q2FY-XT5L-5Z4F'

ip = '8.8.8.8'



response = dashboard.devices.getDeviceLossAndLatencyHistory(

    serial, ip,

    t0 = '2024-08-06T23:45:000Z',

    t1 = '2024-08-07T00:15:00Z',

    #timespan = '2678400',

    resolution = '60',

    uplink = 'wan1',

startTs: 2024-08-06T23:46:00Z
endTs: 2024-08-06T23:47:00Z
lossPercent: 0
latencyMs: 5.82
jitter: 0.14
goodput: 100000
------
startTs: 2024-08-06T23:47:00Z
endTs: 2024-08-06T23:48:00Z
lossPercent: None
latencyMs: None
jitter: None
goodput: None
------
startTs: 2024-08-06T23:48:00Z
endTs: 2024-08-06T23:49:00Z
lossPercent: 0
latencyMs: 5.78
jitter: 0.07
goodput: 100000

Thanks!

[sungod · Accepted solution]

That means no samples in that middle period, as you are using the minimum resolution period perhaps there's an edge effect at work.

I suggest use a longer resolution, say 300, and see if you still get any None returns.

[eroche]

Thanks I used 600 and there were no "None" values.

Its just weird though if I look at the Uplink tab for the MX I see loss minimal like 1% but nothing in the results from the API.

[sungod]

What you see in Dashboard is based upon logic/parameters that are not public and also not necessarily derived from the published API calls.

So they may show differing values, that doesn't mean either is 'wrong', simply that the values were arrived at differently.

For instance, this API call only reflects performance for the specified IP, I'd assume Dashboard is reporting loss across all traffic.

This isn't unique to Meraki, I see the same with other platforms with the UI and API not matching.

It's just something you have to live with, trends and qualitative indicators are still valuable, just don't rely on/expect absolute accuracy, nor make assumptions about UI being based on the published API 😀