The NodePing API and PowerShell

At NodePing we interact with our own API quite a bit from the command line.  Most of that is from bash on Linux, because that’s where we live most of our lives.  But the API works well from just about any scripting environment.  Since our documentation examples all use curl with bash syntax, it seemed like it might be a good idea to also write up some examples of using other tools.  Here, as the first installment of that effort, is a handful of examples of using PowerShell with the NodePing API.

Full disclosure: PowerShell is not an environment we spend a lot of time working with.  There are likely ways to do some of this better.  We’ve tested the example calls in this post.  Hopefully it is enough to get you started if you work with PowerShell.

Basic GET Calls

Basic calls are quite simple to make using Invoke-RestMethod.  If you have the check ID, getting a check is quite easy:

Invoke-RestMethod ‘https://api.nodeping.com/api/1/checks/201205050153W2Q4C-OJ2HSIRF?token=[token]’

This returns a JSON object, which PowerShell handles easily.

_id : 201205050153W2Q4C-0J2HSIRF
description : This is the description, if the check has one.
public : False
customer_id : 201205050153W2Q4C
queue : nyI8JSL23W
interval : 1
created : 1427513104608
pro : nodeping.com
modified : 1510332098196
parameters : @{target=https://example.com/; follow=False; threshold=5; sens=2; invert=False; verify=false}
firstdown : 0
label : Keep me
runlocations : False
enable : active
uuid : bd3eha9o-z2m5-4qg2-9k8b-jqq0kiituyxz
state : 1
status : assigned
notifications : {}
type : HTTP
acctdisable : False
suspacct : False
dep : False

Note that “parameters” is another hash.  You can refer to the fields in the check as properties of the returned object, including properties that are hashes themselves.  So, for example, this will give you the check check threshold, which is part of “parameters”.

$check = Invoke-RestMethod ‘https://api.nodeping.com/api/1/checks/201205050153W2Q4C-0J2HSIRF?token=[token]’ ;
$check.parameters.threshold;

API Authentication

PowerShell doesn’t easily support basic authentication.  Since the NodePing API supports passing the token as a parameter, the easiest approach is to include it in the query string, as I did in the example above.  In most cases I prefer to pass it in the Body, as I’ll show in examples below.  If you’re working with a script its usually easiest to set a $token variable and use that throughout your script.  If you’re just sending a URL as I did in the simple GET calls above, actually putting in the query string works fine.

If you’re using the API in scripts that make several calls to the API, you might want a more reusable way to pass the token that keeps it out of your Body.  You can do this by manually building the headers.  That requires base64 encoding the credentials and then setting that using the -Headers argument. That looks something like this:

$hash = [Convert]::ToBase64String([Text.Encoding]::ASCII.GetBytes(("{0}:{1}" -f $token,"")))

Then set the -Header argument to @{Authorization=(“Basic {0}” -f $hash)}

Or, if you are doing this several times, you will probably want to set this to a variable you can reuse.

Some sites suggest that you create a credential object using ConvertTo-SecureString and Management.Automation.PSCredential, and pass that using the -Credential argument.  However, this won’t work with the NodePing API, since the -Credential option waits for the challenge to send the authorization header.  The API is stateless, and expects the header on each request.

Put and Post Calls

We recommend that you use JSON to set fields for our API in most environments.  However, PowerShell does not automatically handle sending the body as JSON.  For POST calls, this isn’t a big deal, and you can just POST the body as it is and Invoke-RestMethod sends the data as if it were a form.  PUT and DELETE calls don’t work that way.  For those, you have to convert to JSON, or include all of your parameters in a query string.

Creating a check looks like this:

Invoke-RestMethod 'https://api.nodeping.com/api/1/checks' -Method Post -Body @{ 
    label='test label' 
    type="HTTP"
    target="http://example.com"
    token=$token 
}

This call will return an object with the new check.  You can add any of the other fields listed in our documentation.

Updating a check is almost the same.  This can all be done on one line, but we’ll split it out here to make it easier to see.

$data = @{ label='a new label'; type="HTTP"; interval=5; token=$token } | ConvertTo-Json
Invoke-RestMethod 'https://api.nodeping.com/api/1/checks/201205050153W2Q4C-0J2HSIRF' -Method Put -Body $data

Note that since we’re updating a check, we need to include the check ID.  As with the Post call, the Put call returns the updated check object.

Working with Lists of Checks

Working with a list of checks is slightly trickier.  The call returns a JSON object with a list of checks using the ID as the key.  For PowerShell, this is a hash table of hashes.  So to list checks, you would do something like this:

Invoke-RestMethod 'https://api.nodeping.com/api/1/checks' -Body @{ token=$token } | %{ $_.PSOBJECT.Properties.value }

This is actually fairly handy, because you can fairly easily list a specific field from the response.  For example, this lists all of the targets from all of the checks on this account:

Invoke-RestMethod 'https://api.nodeping.com/api/1/checks' -Body @{ token=$token } | %{ $_.PSOBJECT.Properties.value.parameters.target }

You could do all sorts of things at this point.  For example, here’s a list of all checks that include “nodeping” somewhere in the check’s target:

Invoke-RestMethod 'https://api.nodeping.com/api/1/checks' -Method Get -Body @{ token=$token } | %{ foreach($value in $_.PSOBJECT.Properties.value){ if($value.parameters.target -like "*nodeping*"){ $value } } }

This is practical as a way to find a check with a specific target up to a few thousand checks.

Getting Results

Applying the same principles should let you do just about anything with the NodePing API.  Applying the same pattern to a results call, for example:

Invoke-RestMethod 'https://api.nodeping.com/api/1/results/201205050153W2Q4C-0J2HSIRF' -Method Get -Body @{ token=$token; limit=2; clean=1 }

Note that you’ll want to always include the “clean” parameter for results.  The unclean response takes more parsing.

The options available for results calls are documented here:
https://nodeping.com/docs-api-results.html

Profit!

That’s the basics of using PowerShell to interact with the NodePing API.  With the API you can add checks, remove checks, get your results and uptime, manage contacts and notifications, and add and manage subaccounts.  Our customers use the API both for programmed integrations, and from the command line using quick scripts like the ones I demonstrated here to make quick changes to several (or lots) of checks at once.

If you’re doing interesting things with PowerShell and the NodePing API, we’d like to hear about it!  Please email support and let us know.  That’s also a great place to ask us questions.  We’re happy to help people interact with our service.

Support for Multiple Public Status Pages

Our public status report is a critical part of keeping your customers informed of your site and service status—which after all is one of the points of monitoring. Our status pages are customizable to your company, and support a custom domain so you can display your status at status.yourdomain.com, or whatever is most applicable to your business.

We have had a public status page as one of our key features for some time.  Now, we’re adding support for multiple status pages on one account. For instance, NodePing has a status report page for our websites at status.nodeping.com, and one specifically for our probe servers at probestatus.nodeping.com.  Business and Provider accounts can optionally set up an SSL cert for their status pages (contact support for info how).

To create a new status report, just log into your account , and go to the “Account Settings” tab, then the “Reporting” subtab. Click “Add new status report” and add as many checks as you want to your new status report.  

We hope you will find this feature just as useful as we do.  We also have several more enhancements for public status pages coming soon.  Let us know what you think at support@nodeping.com, by posting comments here, or by using our Contact Page.

IPv6 Monitoring

NodePing now supports IPv6… mostly.

Our website and server monitoring service can now monitor IPv6 services. With real, honest-to-goodness ICMP pings as well as the funky bracket notation for URLs:

http://[2606:c700:4020:11::53:4a3b]/

Our IPv6 support is pretty extensive.

Even though IPv6 adoption has only just crossed the 10% mark, we know many of our customers are out in front of that pack and we’re happy to keep an eye on your v6 stuff.

If you’re using hostnames that resolve to both IPv4 and IPv6, you’ll need to set up separate checks and specify the IPv6 services using the IP as our services will continue to prefer IPv4 for resolution of hostnames. The exception here is our ping check, where you can specify the protocol version to use with a hostname.  You’ll need two checks to cover both protocols.

Unfortunately there are a few caveats and addendums… A couple of our check type don’t yet support IPv6 routing – RBL and SIP checks.  Not many RBLs (Real-time BlackLists) support IPv6 addresses and our SIP check relies on a library that doesn’t support IPv6.  We’ll keep an eye on these and add IPv6 support to them when we can.

We also aren’t able to provide IPv6 monitoring from our Latin America region, sorry.  IPv6 connectivity in Latin America is nearly non-existent currently. We’re sure that will change in the future and we’ll be able to bring that online eventually as providers there start offering it. Our other regions (North America, Europe, and East Asia/Oceania) are well supported.

IPv6 support is available on all NodePing plans. If you’ve got IPv6 services that need monitoring and don’t have a NodePing account yet, please sign up for our free 15-day free trial.  We’ve got your back.

 

rDNS: Monitoring the Flip Side of DNS

DNS monitoring is an important part of keeping your services available. DNS is what allows your browser to turn the name ‘nodeping.com’ into the IP address 192.95.37.22. Without proper functioning DNS, your website, email, and other services would be unreachable.

One often neglected part of DNS monitoring is the rDNS, or reverse DNS, entries. As the name suggests, rDNS is the reverse of DNS. It maps an IP address to a hostname using a special PTR DNS record type. In essence, it associates an IP address to a specific hostname or domain.

PTR records are used by all kinds of utilities and services like the humble ‘ping’ and ‘traceroute’ as well as more complex FCrDNS-enabled services. A forward-confirmed reverse DNS (FCrDNS) verification uses the PTR record to associate a domain owner with an IP address. It’s not a rock-solid form of validating ownership but is usually considered enough to be used for whitelisting servers for SMTP services because spammers usually can’t fake an rDNS record when they forge domains.

If you’re sending email from a server, you should have a proper PTR record in place that includes the domain for the ‘from’ address. This will help ensure your email from that server will not get sent to the spam bucket. A PTR, or ‘pointer’, record is usually configured by whomever owns the IP address so you often have to put in a ticket with your colocation or service provider to set or change rDNS entries.

NodePing is one of very few server monitoring services that can monitor rDNS entries. To set up a rDNS check, select ‘DNS’ from the check type and ‘PTR’ from the record type dropdown. It’s important to note that your PTR record will be in what is commonly called ‘arpa’ format. An example of the ‘arpa’ format for the IP 192.95.37.22 is ‘22.37.95.192.in-addr.arpa’ – please note how the octets are in reverse order, not the numbers. Set the ‘arpa’ address for your IP address in the ‘Query’ text field. You’ll also want to set the ‘Expected Response:’ field to the hostname for that IP, example: ‘api.nodeping.com’.

For more information about rDNS/PTR monitoring or our DNS check capabilities in general, check out our DNS check documentation.

HTTP Advanced Check

Our HTTP checks for website monitoring at NodePing already include our standard HTTP Check, the HTTP Content Check that lets you verify that specific content is present or is not present in the page, and the HTTP Parse Check that allows you to track and alert on arbitrary data points in the response. Today we’re excited to announce that we’re adding the HTTP Advanced Check to our HTTP line up.

The new HTTP Advanced Check adds the following capabilities:

  • simulate a form POST to your web site and verify the expected response
  • check for arbitrary HTTP response status codes for custom API servers
  • send HTTP headers
  • verify specific HTTP headers are being received
  • send PUT, DELETE, HEAD, TRACE, or CONNECT methods

This will allow you to do more in-depth monitoring of your HTTP services. Use cases may include:

  • POST incorrect credentials to log in pages and verify the HTTP status code of 403 is returned.
  • Send mobile browser User-Agent headers and use the content checking to verify the mobile version of your site is being shown
  • Verify a PDF link is returning a PDF file by checking the return header for the correct ‘Content-Type’:’application/pdf’
  • Verify your redirect script is returning a 302 status code and not an error.

Additional information about this new check type can be found in our documentation.

The HTTP Advanced check is now available on all NodePing accounts. All accounts also include unlimited notifications, including international SMS. If you don’t have a NodePing account yet, please sign up for our free 15-day free trial.

Sendgrid’s outage impacted our notification systems

Our notification systems use Sendgrid for email delivery. Overall, Sendgrid has been very reliable and a solid service provider. Today Sendgrid had a fairly significant outage, and that has impacted the timely delivery of some of our email notifications. Obviously for a service like ours this is a big deal.

We are in the process of reviewing our notifications infrastructure to see how our failover systems can be improved to make sure our notifications always get out in a timely fashion. Our apologies to our users who were impacted by this problem.

If you have any questions about this issue, please feel free to email us at support@nodeping.com.

Twitter Notifications

NodePing is happy to announce our newest notification method – twitter direct messages.  The ability to receive a twitter direct message is a great addition to our current notification system that already includes unlimited email, international SMS, and voice calls.

Twitter notifications are in testing at this point.  They are available on all accounts so please do kick the tires and let us know how things work for you at support@nodeping.com.

You’ll need to follow @NodePing in order to get direct message alerts.  Then add your twitter handle in your contact record and in your check’s notification section and we’ll send you a private and discrete ‘direct message’ (not an embarrassing public tweet) when that check goes down and again when it comes back up.

Let us know in the comments how this new notification type is working for you and what you’d like to see added next – instant message (IM), HTTP POST to url, carrier pigeon, etc?