The health check with pub/health_check.php, that is configured in the probe section of the Varnish (version 6) configuration, seems to take a couple of seconds and causes some significant CPU usage. Is this normal or is there anything wrong? I don't know that and I'd like to know what could be expected.

Afaik, the default interval time is set to 5 seconds. Is this a reasonable value or should it be set to another?


These are the settings I use:

    backend apache_1 {
    .host = "magento2_php-apache_1";
    .port = "80";
    .first_byte_timeout = 180s;
    .probe = {
        .request =  "GET /health_check.php HTTP/1.1"
                    "Host: magento2.dev”
                    "Connection: close"
                    "Accept: text/html" ;
        .timeout  = 60s;
        .interval = 120s;
        .window    = 10;
        .threshold = 8;

  • Not very useful information. – robsch May 27 '20 at 13:42
  • well I can’t comment on whether a 5 second interval for the healthcheck probe reasonable is because I don’t know your setup or buisness requirements. If you are clustering varnish and cannot allow more than 5 second failover time then it’s reasonable. If you want to reduce processor overhead caused by the probe then extending the interval to 120 seconds is also reasonable. Those are the settings, it’s your call. – paj May 27 '20 at 19:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.