So I just restored the original sa-update script from the qmt github,
specifically the conditional service restart section.
if [ "${PIPESTATUS[0]}" == "0" ] \
|| [ "$1" == "-r" ]; then
/sbin/service spamd condrestart >/dev/null 2>&1
else
exit ${PIPESTATUS[0]}
fi
That should do it right?
-----
Sincerely,
Fabian Santiago
Sent from my iPhone
> On May 3, 2015, at 9:22 AM, Fabian Santiago <[email protected]> wrote:
>
> Hello Eric,
>
> Sa-update cron job error:
>
> /etc/cron.daily/sa-update:
>
> svc: warning: unable to chdir to /var/qmail/supervise/spamd: file does not
> exist
> svc: warning: unable to chdir to /var/qmail/supervise/spamd: file does not
> exist
> svc: warning: unable to chdir to /var/qmail/supervise/spamd: file does not
> exist
>
>
> Contents of sa-update cron script:
>
>
> #!/bin/bash
>
> # Only restart spamd if sa-update returns 0, meaning it updated the rules
>
> /usr/bin/sa-update | logger -i -p mail.info
>
> if [ "${PIPESTATUS[0]}" != "0" ]; then
> exit ${PIPESTATUS[0]}
> fi
>
> svc -d /var/qmail/supervise/spamd | logger -i -p mail.info
> svc -t /var/qmail/supervise/spamd | logger -i -p mail.info
> svc -u /var/qmail/supervise/spamd | logger -i -p mail.info
>
>
> What I notice is spamd doesn't exist under qmail/supervise.
>
> On my cos6.6 system it runs as its own standard daemon started by its init
> script under /etc/init.d.
>
> So the likely changes needed here are, as I would imagine:
>
> Something alone the lines of changing the svc lines to either point to other
> binaries/scripts or something like "service spamd <command>", ??????
>
> What was the original sa-update cron script?
>
> -----
>
> Sincerely,
>
> Fabian Santiago
>
> Sent from my iPhone
>
>> On May 2, 2015, at 4:54 PM, Fabian Santiago <[email protected]>
>> wrote:
>>
>> Ok, that has been my observation. Thanks.
>>
>> -----
>>
>> Sincerely,
>>
>> Fabian Santiago
>>
>> Sent from my iPhone
>>
>>> On May 2, 2015, at 3:22 PM, Eric Broch <[email protected]> wrote:
>>>
>>> Fabian,
>>>
>>> According to the simscan wiki...
>>> spam_hits=number
>>> The spam_hits option changes the threshold spamassassin spam value. If a
>>> mail gets a spam score that number or above, it is rejected, otherwise it
>>> is accepted.
>>> http://www.qmailwiki.org/Simscan/Guide#regex_matching
>>>
>>> In other words (I think), if an email scores over 'spam_hits' it will be
>>> rejected and won't go to a mailbox.
>>>
>>> EricB
>>>
>>>> On 5/2/2015 9:48 AM, Fabian Santiago wrote:
>>>> Yes it does. I was just able to test it. Thanks.
>>>>
>>>> -----
>>>>
>>>> Sincerely,
>>>>
>>>> Fabian Santiago
>>>>
>>>> Sent from my iPhone
>>>>
>>>>>> On May 2, 2015, at 1:38 AM, Eric Broch <[email protected]> wrote:
>>>>>>
>>>>>> I think it should, though, I'm no expert. Maybe someone else can chime
>>>>>> in...anyone?
>>>>>>
>>>>>>> On 5/1/2015 11:16 PM, Fabian Santiago wrote:
>>>>>>> Nvm, You helped me find it. You pointed me to the actual config file
>>>>>>> where as I was looking at a file under /etc/spamassassin, which
>>>>>>> contains my prior file renamed.
>>>>>>>
>>>>>>> Once I edited the actual file all is as expected. Under
>>>>>>> /etc/mail/spamassassin.
>>>>>>>
>>>>>>> Question, can/will sa actually rewrite the header with ***spam***
>>>>>>> pre-pended if over the sa defined score limit but below simscans
>>>>>>> defined spam_hits limit?
>>>>>>>
>>>>>>> -----
>>>>>>>
>>>>>>> Sincerely,
>>>>>>>
>>>>>>> Fabian Santiago
>>>>>>>
>>>>>>> Sent from my iPhone
>>>>>>>
>>>>>>> On May 2, 2015, at 12:20 AM, Eric Broch <[email protected]> wrote:
>>>>>>>
>>>>>>> Fabian
>>>>>>>
>>>>>>> And, I might add that the error running I got when initially running
>>>>>>> 'spamassassin -D --lint' was that it could not find the configuration
>>>>>>> file
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe, e-mail: [email protected]
>>>>> For additional commands, e-mail: [email protected]
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: [email protected]
>>>> For additional commands, e-mail: [email protected]
>>>
