[CentOS] CentOS 8.1 cron does not send mail

Sat Mar 21 10:56:38 UTC 2020
Tobias Kirchhofer <collect at shift.agency>

On 27 Feb 2020, at 14:42, Jonathan Billings wrote:

> On Feb 27, 2020, at 08:01, Tobias Kirchhofer <collect at shift.agency> 
> wrote:
>>
>> Hi,
>>
>> we experience difficulties with crond behaviour sending mail since 
>> CentOS 8.1. The cron job is the same like we used in CentOS 7.
>>
>> crontab -l
>> /usr/bin/python3 -c 'import random; import time; 
>> time.sleep(random.random() * 3600)' && /usr/local/bin/backup.sh
>
> Is this literally what your crontab looks like?   Because that’s not 
> valid crontab syntax.

This is what it is literally:

0 5 * * * /usr/bin/python3 -c 'import random; import time; 
time.sleep(random.random() * 3600)' && /usr/local/bin/backup.sh

>> backup.sh writes the backup to the remote backup server. There are 
>> around 30 machines with the same cron job. The python part spreads 
>> the jobs over an hour so that the backup server is not struck at once 
>> from about 30 machines.
>>
>> The change in behaviour is that crond sends no mail anymore. It is 
>> because of the logical and (&&). Without this crond sends mails.
>>
>> What has changed in CentOS 8 and does anyone has an idea how we could 
>> fix it?
>
> There aren’t any significant changes in ‘cronie’ in 8.1, looking 
> at the spec file.
>
> Assuming the crontab you wrote above included the time spec too, I’d 
> check to make sure the first command isn’t exiting with a non-zero 
> exit code.

The command chain is running propery, STDOUT output is visible in system 
log (excerpt from logwatch email):


```
--------------------- Cron Begin ------------------------

  **Unmatched Entries**
  CMDOUT (### Starting backup. Host: host.example.com Backupserver: 
10.9.1.5 Path: /borgbackup/vm/host-example-com Date: Fri Mar 20 05:13:46 
CET 2020 ###)
  CMDOUT (Creating archive at 
"borg at IP:/borgbackup/vm/host-example-com::{now:%Y-%m-%d_%H:%M}")
…
```

Its just not sending an email.

What is additionally irritating is that it is only on a few machines. 
All machines are identically provisioned with ansible.

The situation is not mission critical but it should work. Strange.

Tobias
	

> --
> Jonathan Billings <billings at negate.org>
>
>
> _______________________________________________
> CentOS mailing list
> CentOS at centos.org
> https://lists.centos.org/mailman/listinfo/centos


-- 
collect at shift.agency