From: "Steven Levine" <ecs-isp@2rosenthals.com>
Message-ID: <69d5f145.8.mr2ice.fgrirsq@earthlink.net>
Date: Tue, 07 Apr 2026 23:10:13 -0700
To: "eCS ISP Mailing List" <ecs-isp@2rosenthals.com>
In-Reply-To: <list-2030831@2rosenthals.com>
Subject: Re: [eCS-ISP] LE certificate renewal fails when using CRON2

In <list-2030831@2rosenthals.com>, on 04/07/26
   at 10:13 PM, "Massimo S." <ecs-isp@2rosenthals.com> said:

Hi Massimo,

>if i start the script manually from the command line everything works
>correctly, but instead when the script is started by the scheduler
>(CRON2) it fails with this:

When did the start happening or did it never work from cron?

>uacme: fetching directory at
>https://acme-v02.api.letsencrypt.org/directory uacme: curl_get: GET
>https://acme-v02.api.letsencrypt.org/directory failed: Problem with the
>SSL CA cert  (path? access rights?)
>uacme: curl_get: waiting 5 seconds before retrying

When Let's Encrypt is overloaded, it can have problems access with
accessing URLs on it's server.  The relevant code is

curlwrap.c:116
        res = curl_easy_perform(curl);
        if (res != CURLE_OK) {
            warnx("curl_get: GET %s failed: %s", url,
                    curl_easy_strerror(res));
            curldata_free(c);
            c = NULL;

I would normally suspect a problem at the server end, but since you can
run from the command line, it must be something else.

ATM, I have no idea what cron2 might be doing to mess up the runtime
environment for uacme, but it must be doing something.

You could try turning on logging.  You could try running full screen or
not minimized.


Steven

-- 
----------------------------------------------------------------------
"Steven Levine" <steve53@earthlink.net>  Warp/DIY/BlueLion etc.
www.scoug.com www.arcanoae.com www.warpcave.com
----------------------------------------------------------------------