Uptime percentage for today, 7 days, 1 month, 3 months

[kovoks]

Hi There,

Thanks for making a very good tool. I love it.

We are currently using uptimerobot, but we plan to migrate to Bluewave. For our current status page we have stats for today, week and month. ( https://kovoks-status.nl ). Can I get those numbers from the API?

I see some hints in monitors/stats/:montitorid

    "uptimePercentage": 0.9783653846153846,
    "status": true,
    "periodAvgResponseTime": 0,
    "periodUptime": 0,
    "aggregateData": [],
    "periodIncidents": 0,
    "periodTotalChecks": 0,
    "checks": [],

But I'm not sure how for example aggregateDate gets filled over time.

And for reporting to our customers, we currently use a 90 day average, I found a setting to save the data for 90 days, but I'm not sure how to retrieve the number for that.

I hope someone who uses Bluewave for a bit longer can help.

[ajhollid]

Hi @kovoks,

Happy to hear you find the software useful!

Could you let me know what kind of metrics you are looking for?

Are you looking for average uptime percentage for day, week, and month?

Let us know what kind of data you'd be interested in and if we don't currently provide it we can look at adding that functionality if it fits with our software!

[kovoks]

Hi,

Thanks for replying! Yes, I would love to have an average uptime percentage for today, 7 days, 30 days and if possible 90 days.

It would be even better if /v1/monitors could return that as uptimePercentage object. But if for speed reasons 30 and 90 days would be to heavy for that call /v1/monitors/stats/:monitorId is fine too for that.

If we have today and 7 days, we can replace the columns in https://kovoks-status.nl for the stats coming from BlueWave.

side note:
uptimerobot allows a custom period (https://uptimerobot.com/api/#getMonitorsWrap), so you can call getMonitors with a parameter custom_uptime_ratios=7-30-45 to get the averages for 7, 30 and 45 days in the result. Maybe an idea for you, but it might be a bit overkill.

[ajhollid]

No problem, glad to have some feedback.

Let me discuss those ideas with the team, I'm sure we can work something out for you.

A few more questions just so I'm clear on what you'd like.

1. Are you using the server headless? Interacting with the api through curl or postman or some thing along those lines? Or do you have some UI needs as well? I assume since you're displaying it on your own website you just need the raw data.

2. Could you show me a sample of data that you'd like to receive? JSON format would be great 👌

If it's just simply uptime percentage for n number of days I can probably have that for you pretty quickly.

[kovoks]

Yes, we love to use the API for this, so we can cache the results for a minute or two and also many of our server names have a direct relation to our customers, so we would like to anonymously report the servers in the list. So the API is fine. We can curl the information indeed and format it as we currently do.

How about something for v1/monitors like this (removed the other fields):

{
  "_id": "123",
  "uptimePercentage": 0.7,
  "aggregatedUptimePercentages": {
    "1": 0.7,
    "7": 0.77,
    "30": 0.9,
    "90": 0.999
  },
  "status": true
}

you probably want uptimePercentage to be backwards compatible.

[ajhollid]

Yes, we love to use the API for this, so we can cache the results for a minute or two and also many of our server names have a direct relation to our customers, so we would like to anonymously report the servers in the list. So the API is fine. We can curl the information indeed and format it as we currently do.

How about something for v1/monitors like this (removed the other fields):

{
  "_id": "123",
  "uptimePercentage": 0.7,
  "aggregatedUptimePercentages": {
    "1": 0.7,
    "7": 0.77,
    "30": 0.9,
    "90": 0.999
  },
  "status": true
}

you probably want uptimePercentage to be backwards compatible.

Yep most definitely want to be backwards compatible.

Do you want the rest of the data as well? If you're only interested in uptime percentage I may just make you a new endpoint to crunch as little data as possible on the back end.

[ajhollid]

@kovoks feature is implemented, you've got a new endpoint at api/v1/monitors/uptime that should return what you're looking for:

curl --request GET \
  --url http://localhost:5000/api/v1/monitors/uptime \
  --header 'Authorization: <bearer_token>

{
	"success": true,
	"msg": "Got all monitors successfully",
	"data": [
		{
			"1": 100,
			"7": 100,
			"30": 100,
			"90": 100,
			"_id": "672adc421f92af2a06be3315",
			"userId": "672acf1a641a68559b6f5a1c",
			"teamId": "672acf1a641a68559b6f5a1a",
			"name": "Google",
			"description": "Google",
			"type": "http",
			"url": "https://google.com",
			"isActive": true,
			"interval": 60000,
			"createdAt": "2024-11-06T03:02:26.041Z",
			"updatedAt": "2024-11-06T03:02:26.267Z",
			"__v": 1,
			"notifications": [],
			"status": true,
			"uptimePercentage": 1
		}
	]
}

The branch is in PR review queue right now, feel free to checkout the branch if you'd like to try it out while we wait for it to be merged in.

Swagger docs are also available at <host>:<port>/api-docs, ex localhost:5000/api-docs if you want to see details for this or other API endpoints as well.