Example of using a Readiness probe in portable services

This article shows an example of implementing a Readiness probe based on HTTP requests.

начало внимание

This article does not include recommendations on how to define whether you need a Readiness probe.

конец внимание

Basic information about the case

This is a module that includes two portable services. Service #1 can send information about its state via HTTP requests. Service #2 can interpret its responses or absence of responses.

We need to configure Service #2 so that it doesn’t accept incoming traffic when Service #1 is in a certain state.

We have decided to do it using a Readiness probe.

Modification of Service #2

We need to define how Service #2 processes the HTTP request that retrieves the status of Service #1.

From the SERV1URL environment variable, the address of Service #1 is retrieved. An HTTP request is sent to this service. If the response status code is 200, and the response body meets the conditions of the probe set up in Service #2, the method we added returns status 200. Otherwise, it returns 504.

func readnessFunc(w http.ResponseWriter, r *http.Request) {
if serv1URL, exists := os.LookupEnv("SERV1URL"); exists {
       response, err := http.Get(serv1URL + "/getstatus")
       if err == nil &&
               response.StatusCode == 200 &&
               isValidStatus(response.Body) {
               fmt.Fprintf(w, "Yeah, it works")
               return
       }
}
w.WriteHeader(http.StatusGatewayTimeout)
}

Configure Service #2

When the image of Service #2 is built and uploaded, we need to configure the Readiness probe in ELMA365. To do that, we need to:

1. In the module’s settings, open the Services tab, then move on to editing Service #2.
2. On the Readiness tab, select HTTP.
3. Set a delay after which the probe is performed.

начало внимание

Don’t make the delay too short. Service #2 must have enough time to be initialized in the cluster to accept requests from probes. If the value of the delay is not enough, Service #2 may become unavailable.

конец внимание

1. In Service #2, specify the port and the path to the HTTP method used for the probe.
2. Pass information about the location of Service #1 to Service #2. To do that, go to the Environment Variables tab and create a string named SERV1URL with the URL of Service #1.
3. Specify the address of the service by clicking the {+} symbol on the right side of the string and select the service from the list. A value of the following format will be generated: {$_srv_ServiceName#1}.

Additional information

When developing and implementing a Readiness probe, note that it is performed all the time while Service #2 exists. In the example provided in this article, Service #2 will constantly send HTTP requests to Service #1.

Was this helpful?

Thanks for your feedback!

Please specify why:

Recommendations did not help meArticle is hard to understandDidn`t answer my questionContent does not match the topicOther

How we can improve it?

YesNo

Please specify whyRecommendations did not help meArticle is hard to understandDidn`t answer my questionContent does not match the topicOther

Found a typo? Highlight the text, press ctrl + enter and notify us