Skip to content

Latest commit

 

History

History
266 lines (216 loc) · 10.6 KB

guide.en-sg.md

File metadata and controls

266 lines (216 loc) · 10.6 KB
title excerpt updated
Using the vScope API
Find out how to use the vScope API to collect monitoring data in your applications
2020-11-18

Objective

OVHcloud provides you with a tool called vScope, for supervising and monitoring your virtual machines and infrastructures.

It is a webpage that displays useful information on your resource usage.

You can also access this information via the APIv6 and Metrics API.

This guide explains how to use these APIs.

Requirements

Instructions

vScope provides two types of information:

  • Live information which corresponds to the information on different components at a given time.
  • Graphs presenting logged performance data for different components, such as a virtual machine’s CPU and RAM.

Collecting live data

Live data is the data available via the vScope interface’s main page.

vScope-API{.thumbnail}

You can retrieve live data for the following components:

  • filers
  • hosts
  • virtual machines

You can use the API via the following three APIv6 calls:

Filers

[!api]

@api {v1} /dedicatedCloud GET /dedicatedCloud/{serviceName}/datacenter/{datacenterId}/filer/{filerId}

Hosts

[!api]

@api {v1} /dedicatedCloud GET /dedicatedCloud/{serviceName}/datacenter/{datacenterId}/host/{hostId}

Virtual machines

[!api]

@api {v1} /dedicatedCloud GET /dedicatedCloud/{serviceName}/datacenter/{datacenterId}/vm/{vmId}

Collecting logged data (graphs)

To collect and use logged data (graphs), we use the Metrics Data Platforms product.

Via OpenTSDB or WARP10 protocol, you will be able to retrieve your data as points. You can use these points via your application, or directly display them in your choice of rendering.

The following instructions cover the use of the OpenTSDB protocol for a raw data display (no graph rendering).

To use Metrics Data Platforms, you will need to get a read token. With the new version of vScope, each infrastructure user has a read token.

Use the following APIv6 call to retrieve the read token for the user you want:

[!api]

@api {v1} /dedicatedCloud POST /dedicatedCloud/{serviceName}/user/{userId}/metricsToken

Your token will be in the result’s token field.

{
    "warpEndpoint": "https://warp10.gra1-ovh.metrics.ovh.net",
    "token": "XXXXXXXXXXXX_XXXXXXXXXXXZZZZZZZZZZZ_YYYYYYYYYYYYYY-XXXXXXXXX",
    "opentsdbEndpoint": "https://opentsdb.gra1-ovh.metrics.ovh.net"
}

For each component type, there is a list of metrics, and it requires a number of very precise settings (also called labels).

Filers

Metrics Description Labels
vscope.filer.datastore.diskspace.used Filer usage in KB datacenter : pcc-37-187-228-180_datacenter869,
datastore : pcc-000443

Hosts

Metrics Description Labels
vscope.host.cpu.usage.perc Host processor usage in percent - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
vscope.host.mem.usage.perc Host RAM usage in percent - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
vscope.host.net.tx Host outgoing network usage - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
- nicname : vmnic0/vmnic1/vmnic2/vmnic3
vscope.host.net.rx Host incoming network usage - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
- nicname : vmnic0/vmnic1/vmnic2/vmnic3
vscope.host.net.packetstx Number of network packets sent from the host - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
- nicname : vmnic0/vmnic1/vmnic2/vmnic3
vscope.host.net.packetsrx Number of network packets received from the host - datacenter : pcc-37-187-228-180_datacenter869,
- host : 172.17.86.51
- nicname : vmnic0/vmnic1/vmnic2/vmnic3

Virtual machines

Metrics Description Labels
vscope.vm.cpu.usage.perc VM processor usage in percent - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.mem.usage.perc VM RAM usage in percent - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.cpu.ready VM’s CPU Ready time in milliseconds - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.net.rx VM incoming network usage - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.net.tx VM outgoing network usage - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.net.packetsrx Number of network packets received from the VM - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.net.packetstx Number of network packets sent from the VM - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.io.read IO number per second in read mode from the VM - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.io.write IO number per second in write mode from the VM - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.bandwidth.read VM disk bandwidth in read mode - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.bandwidth.write VM disk bandwidth in write mode - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.latency.read VM disk latency in read mode - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254
vscope.vm.disk.latency.write VM disk latency in write mode - datacenter : pcc-37-187-228-180_datacenter869,
- vm : vm-01254

Example of data gathered using the OpenTSDB protocol

Now that you have retrieved your token, your endpoint, and your list of metrics, you can retrieve a host’s RAM usage data for a period of 1 day.

Below is an example query:

curl -XPOST https://read:XXXXXXXXXXXX_XXXXXXXXXXXZZZZZZZZZZZ_YYYYYYYYYYYYYY-XXXXXXXXX@opentsdb.gra1-ovh.metrics.ovh.net/api/query
-d '{ 
    "start":1564407950, 
    "queries":[ 
        { 
            "metric":"vscope.host.mem.usage.perc", 
            "aggregator":"sum",
            "downsample":"20m-max-zero",
            "tags": {
                "datacenter":"pcc-37-187-228-180_datacenter869",
                "host":"172.17.86.51" 
            } 
        } 
    ] 
}'

Below are explanations of the fields used:

  • read: user used to send the query (will always be read).
  • XXXXXXXXXXXX_XXXXXXXXXXXZZZZZZZZZZZ_YYYYYYYYYYYYYY-XXXXXXXXX: the token retrieved earlier via the APIv6.
  • opentsdb.gra1-ovh.metrics.ovh.net: OpenTSDB endpoint, also retrieved via the APIv6. This endpoint may vary depending on your location.
  • start: a timestamp corresponding to the query’s start date.
  • queries: a table containing the metrics to retrieve. Several metrics may be retrieved in a single query.
  • metric: the name of the metric to retrieve.
  • aggregator: the name of the aggregation feature (please refer to OpenTSDB documentation for further details).
  • downsample: the name of the downsampling feature (can be used to reduce the amount of data to retrieve. Optional setting).
  • tags: a list of labels in the form of key/value.

Other settings may also be provided. Please refer to the documentation for the OpenTSDB API for further details.

You will then receive a json with the query summary, as well as the timestamps associated with their value in the dps field. For example:

[
    {
        "metric":"vscope.host.mem.usage.perc",
        "tags":{
            "datacenter":"pcc-37-187-228-180_datacenter869",
            "env":"prod",
            "host":"172.17.86.51",
            "servicename":"pcc-37-187-228-180",
            "servicetype":"vscope"
        },
        "query":{
            "index":0
        },
        "aggregateTags":[],
        "dps":{
            "1564409391":4.38,
            "1564410591":4.35,
            "1564411791":4.37,
            "1564412991":4.38,
            "1564414191":4.35,
            "1564415391":4.38,
            "1564416591":4.35,
            "1564417791":4.36,
            "1564418991":4.36,
            "1564420191":4.37,
            "1564421391":4.37,
            "1564422591":4.37,
            "1564423791":4.37,
            "1564424991":4.38,
            "1564426191":4.36,
            "1564427391":4.35,
            "1564428591":4.37,
            "1564429791":4.36,
            "1564430991":4.38,
            "1564432191":4.35,
            "1564433391":4.37,
            "1564434591":4.36,
            "1564435791":4.37,
            "1564436991":4.37,
            "1564438191":4.37,
            "1564439391":4.38,
            "1564440591":4.36,
            "1564441791":4.36,
            "1564442991":4.37,
            "1564444191":4.37,
            "1564445391":4.35,
            "1564446591":4.36,
            "1564447791":4.36,
            "1564448991":4.36,
            "1564450191":4.35,
            "1564451391":4.37,
            "1564452591":4.37,
            "1564453791":4.35,
            "1564454991":4.36,
            "1564456191":4.37,
            "1564457391":4.37,
            "1564458591":4.36,
            "1564459791":4.37,
            "1564460991":4.34,
            "1564462191":4.36,
            "1564463391":4.34,
            "1564464591":4.37,
            "1564465791":4.34,
            "1564466991":4.37,
            "1564468191":4.34,
            "1564469391":4.36,
            "1564470591":4.36,
            "1564471791":4.36,
            "1564472991":4.37,
            "1564474191":4.37,
            "1564475391":4.36,
            "1564476591":4.35,
            "1564477791":4.36,
            "1564478991":4.35,
            "1564480191":4.35,
            "1564481391":4.37,
            "1564482591":4.36,
            "1564483791":4.34,
            "1564484991":4.37,
            "1564486191":4.38,
            "1564487391":4.35,
            "1564488591":4.34,
            "1564489791":4.36,
            "1564490991":4.35,
            "1564492191":4.36,
            "1564493391":4.36,
            "1564494591":4.36
        }
    }
]

For further details on OpenTSDB queries, please refer to the following documentation: OpenTSDB API query

Go further

Join our community of users on https://community.ovh.com/en/.