Enable email notification to external address

[ckotte]

Could you implement email notification to external addresses?

Probably a copy of system-email plus at least one parameters like recipient.

[dshearer]

Could you use a notify script to do this?

…

On Sep 6, 2020, at 11:52 AM, Christian Kotte ***@***.***> wrote:  Could you implement email notification to external addresses? Probably a copy of system-email plus at least one parameters like recipient. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[ckotte]

Is there a message passed to the program as a parameter?

[ckotte]

Wouldn't is be possible to just duplicate result_sink_system_email.go and change execResult, err := common.ExecAndWait([]string{"sendmail", rec.Job.User}, msgBytes) and replace rec.Job.User with the recipient parameter?

[dshearer]

Yeah. When I get around to it ;) In the meantime, would this work: https://dshearer.github.io/jobber/doc/v1.4/#notifications

…

On Sep 7, 2020, at 3:46 AM, Christian Kotte ***@***.***> wrote: Wouldn't is be possible to just duplicate result_sink_system_email.go and change execResult, err := common.ExecAndWait([]string{"sendmail", rec.Job.User}, msgBytes) and replace rec.Job.User with the recipient parameter? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#310 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABRCWRGCK2LZJ3YSI43NMETSES2WZANCNFSM4Q4YQRHQ>.

[ckotte]

I created a job where the command errors every time. However, the job doesn't never fails. It should eventually fail if onError is set to Backoff, righ?

I configured it to run every minute, but the schedule changes after two failures:

bash-5.0# jobber log
TIME                  JOB       RESULT     NEW JOB STATUS  
Sep 13 17:09:00 2020  TestEcho  succeeded  Good            
Sep 13 17:08:45 2020  TestEcho  succeeded  Good            
Sep 13 17:08:00 2020  TestEcho  succeeded  Good            
Sep 13 17:07:45 2020  TestEcho  succeeded  Good            
Sep 13 17:07:00 2020  TestEcho  succeeded  Good            
Sep 13 17:06:45 2020  TestEcho  succeeded  Good            
Sep 13 17:06:00 2020  TestEcho  succeeded  Good            
Sep 13 17:05:45 2020  TestEcho  succeeded  Good            
Sep 13 17:05:00 2020  TestEcho  succeeded  Good            
Sep 13 17:05:00 2020  TestFail  failed     Backoff         
Sep 13 17:04:45 2020  TestEcho  succeeded  Good            
Sep 13 17:04:00 2020  TestEcho  succeeded  Good            
Sep 13 17:03:45 2020  TestEcho  succeeded  Good            
Sep 13 17:03:00 2020  TestEcho  succeeded  Good            
Sep 13 17:02:45 2020  TestEcho  succeeded  Good            
Sep 13 17:02:00 2020  TestEcho  succeeded  Good            
Sep 13 17:01:45 2020  TestEcho  succeeded  Good            
Sep 13 17:01:00 2020  TestEcho  succeeded  Good            
Sep 13 17:01:00 2020  TestFail  failed     Backoff         
Sep 13 17:00:45 2020  TestEcho  succeeded  Good            
Sep 13 17:00:00 2020  TestEcho  succeeded  Good            
Sep 13 16:59:45 2020  TestEcho  succeeded  Good            
Sep 13 16:59:00 2020  TestEcho  succeeded  Good            
Sep 13 16:59:00 2020  TestFail  failed     Backoff         
Sep 13 16:58:45 2020  TestEcho  succeeded  Good            
Sep 13 16:58:00 2020  TestEcho  succeeded  Good            
Sep 13 16:58:00 2020  TestFail  failed     Backoff         
Sep 13 16:57:45 2020  TestEcho  succeeded  Good

bash-5.0# jobber list
NAME      STATUS   SEC/MIN/HR/MDAY/MTH/WDAY  NEXT RUN TIME              NOTIFY ON SUCCESS  NOTIFY ON ERR  NOTIFY ON FAIL  ERR HANDLER
TestEcho  Good     */45 * * * * *            Sep 13 17:09:45 +0000 UTC  program            program        program         Backoff
TestFail  Backoff  */60 * * * * *            Sep 13 17:10:00 +0000 UTC                     program                        Backoff

When will it fail and will the JSON be different? I can differentiate between success and error via "succeeded", but how can I get the failure state?

{
    "job": {
        "command": "cat 'hello world'",
        "name": "TestFail",
        "status": "Backoff",
        "time": "*/60 * * * * *"
    },
    "startTime": "Sep 13 17:01:00 2020",
    "stderr": "cat: can't open 'hello world': No such file or directoryn",
    "stderr_base64": false,
    "stdout": "",
    "stdout_base64": false,
    "succeeded": false,
    "user": "root"
}

Edit: Ok. I got it. I should use Stop for testing. The job is now in failure state. I get the same JSON with notifyOnError and notifyOnFailure. How my script check if notifyOnFailure triggered it?

{
    "job": {
        "command": "cat 'hello world'",
        "name": "TestFail",
        "status": "Failed",
        "time": "*/60 * * * * *"
    },
    "startTime": "Sep 13 17:35:00 2020",
    "stderr": "cat: can't open 'hello world': No such file or directoryn",
    "stderr_base64": false,
    "stdout": "",
    "stdout_base64": false,
    "succeeded": false,
    "user": "root"
}