Daniel Liston wrote:
> Load Average is different for every machine. If you have statistical
> information regarding your machines "normal" load vs. "high" load, I
> would set the max_loadavg in majordomo.cf to your high mark. You do
> not want majordomo to be the cause of pushing your machine over the
> edge.
Thank you Daniel,
I'll set it according to the sar reports.
BR,
Matias.
> Matias Lopez Bergero wrote:
>
>>Daniel Liston wrote:
>>
>>
>>>Does this email come from the same server as your majordomo? Is any
>>>other user of the system having trouble getting mail to the outside
>>>world? As the root user, can you "mail -s test mailbox @
other .
domain"?
>>>
>>>The real question I am asking, is if your sendmail is working. If not,
>>>there is no way for majordomo to work, because it depends entirely on
>>>your MTA for email delivery. Mail might come in ok, and work between
>>>internal mailboxes, without being able to talk to the rest of the 'net.
>>
>>
>>Yes, the MTA was working fine, and that was what get me confused. But I
>>have already solved the problem.
>>The problem was entirely my fault.
>>
>>Wen the messages via majordomo started to arrive late, from four to six
>>hours of delay, I comment out the line that sets the max_loadavg
>>variable at the majordomo.conf. Looks like without this line majordomo
>>waits for the system to have 0 load average :-P.
>>
>>So, after burning my brain, I try setting this parameter again, and
>>things works out. I feel like kind of idiot, but shit happens :-D
>>
>>Now I have a question, what would be a good value to set this
>>configuration variable?
>>
>>Thanks for your help Daniel.
>>
>>BR,
>>Matias.
>>
>>
>>
>>
>>>Matias Lopez Bergero wrote:
>>>
>>>
>>>
>>>
>>>>Hello Daniel,
>>>>
>>>>Daniel Liston wrote:
>>>>
>>>>
>>>>
>>>>
>>>>>Has anything been installed on your server that could have replaced the
>>>>>/etc/smrsh/wrapper link to /usr/local/majordomo/wrapper?
>>>>
>>>>
>>>>no, nothing.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Has the machine been busy (high cpu) where majordomo might not be able
>>>>>to pass mail to the MTA?
>>>>
>>>>
>>>>Yes, I change the $max_loadavg value to release this a little bit, but
>>>>the load isn't to much anyway.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>When you refer to messages sent directly to a mailbox, do you mean by
>>>>>majordomo or by some other source?
>>>>
>>>>
>>>>Sorry, I mean sending a mail to a local user from a local account or a
>>>>remote one. The message is received with no delay at all.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Considering the failure is temporary, does the mail eventually get
>>>>>delivered?
>>>>
>>>>
>>>>I was happening that the message was getting in after four to six hours
>>>>later, but now is getting even worst, I haven't get a single message
>>>>trough majordomo since yesterday
>>>>
>>>>I have tons of similar problems reports over the list and other mailing
>>>>lists, but no one haves give me a clue of where the problem maybe.
>>>>I configured sendmail to run without smrsh and the error was the same
>>>>but instead smrsh was /bin/sh.
>>>>
>>>>Thank you,
>>>>Matias.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Dan Liston
>>>>>
>>>>>Matias Lopez Bergero wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>Hello,
>>>>>>
>>>>>>The last days I am having this problem, it seems that all the messages
>>>>>>that goes trough majordomo are getting queued, here is a mailq output:
>>>>>>
>>>>>>kBKFLeZP009438 0 30769 Dec 20 12:21 owner-prueba
>>>>>> (Deferred: prog mailer (/usr/sbin/smrsh) exited with
>>>>>>EX_TEMPFAIL)
>>>>>>
>>>>>>(---prueba:8:0:<prueba @
udesa .
edu .
ar>---)
>>>>>>
>>>>>>"|/usr/local/majordomo/wrapper resend -l prueba prueba-list-00FF
>>>>>> (Deferred: prog mailer (/usr/sbin/smrsh) exited with
>>>>>>EX_TEMPFAIL)
>>>>>>
>>>>>>A message sent directly to a mailbox get passed without a problem.
>>>>>>I have completely no idea of what is happening,
>>>>>>Any help is most welcome.
>>>>>>
>>>>>>Thanks,
>>>>>>Matias
>>>>>>
>>>>>>
>>>>>
>>>>>
>
References:
|
|
