Quarkus Kafka consumer with exponential backoff retry

I have the following scenario: I have a Quarkus Kafka consumer that gets events (in JSON) from a topic and sends it on to a downstream system via their own web service).

However sometimes there is a network hiccup or the target system is just DOWN (maintenance etc).

I currently have the consumer set up to to retry indefinitely but I notice that it retries a few times and then just marks the consumer as BAD in the health endpoint and stops. Now this is fine as I can then manually just restart the consumer, but I would rather have the consumer not rely on human intervention.

My current code looks as follows:

    @Incoming("fin-in")      @Retry(delay = 120, delayUnit = ChronoUnit.SECONDS, maxRetries = -1, maxDuration = 300, durationUnit = ChronoUnit.SECONDS)      public void receive(ConsumerRecord<String, String> event) throws Exception, RetryException {try {        //Send to Client endpoint        CloseableHttpClient target = HttpClients.createDefault();        HttpPost httpPost = new HttpPost(endpoint);        UsernamePasswordCredentials creds = new UsernamePasswordCredentials(userName, passwd);        httpPost.addHeader(new BasicScheme().authenticate(creds, httpPost, null));        StringEntity entity = new StringEntity(event.value());        httpPost.setEntity(entity);        httpPost.setHeader("Accept", "application/json");        httpPost.setHeader("Content-type", "application/json");        CloseableHttpResponse response = target.execute(httpPost);        if (response.getStatusLine().getStatusCode() == okStatus) {          LG.info("Event sent successfully");        }        else {          if (response.getStatusLine().getStatusCode() == failStatus) { // Never able to process event, log error and continue to next event            HttpEntity hEntity = response.getEntity();            String entityJSON = EntityUtils.toString(hEntity);            LG.error("Fatal error sending to Client endpoint: "+ failStatus +" " + response.getStatusLine().getReasonPhrase() +" " + entityJSON);            logStdError("Error Sending record to Client endpoint" + response.getStatusLine().getReasonPhrase(), event.value(), "ClientEvent", topic, "Client", event.key());          }          else {            HttpEntity hEntity = response.getEntity();            String entityJSON = EntityUtils.toString(hEntity);            LG.error("Unable to send to Client target, retrying: " + response.getStatusLine().getStatusCode() +" " + response.getStatusLine().getReasonPhrase() +" " + entityJSON);            logStdError("Error Sending record to Client endpoint, http status = " + response.getStatusLine().getStatusCode() +" retrying later " + response.getStatusLine().getReasonPhrase(), event.value(), "ClientEvent", topic, "Client", event.key());            throw new RetryException("RETRY_EXCEPTION", "Unable to send to Client target, retrying " ++response.getStatusLine().getStatusCode() +" " + response.getStatusLine().getReasonPhrase());          }        }      }      catch (RetryException rEx) {        throw rEx;      }      catch (Exception ex) {        logStdError("Error sending event to Client endpoint: " + ex.getMessage(), event.value(), "ClientEvent", topic, "Client", event.key());        throw ex;      }

What I would like to implement is a indefinite retry but with an exponential back off. I.e. After the first failure, retry the same event in 5 minutes. If it fails again, wait 10 minutes, and if again 20 minutes etc etc etc (Effectively doubling the time between retries).