Subject: Warning: could not send message for past 4 hours
**********************************************
** THIS IS A WARNING MESSAGE ONLY **
** YOU DO NOT NEED TO RESEND YOUR MESSAGE **
**********************************************
The original message was received at Fri, 14 Dec 2007 17:55:07 -0500
from
[172.20.1.147]
----- Transcript of session follows -----
<pv....ton@charter.net>...
Deferred: Resource temporarily unavailable
Warning: message still undelivered after 4 hours Will keep trying
until
message is 5 days old
Anyone have ideas as to what can cause this type of problem? Seems
once it's in this state it will eventually bounce, even though it
appears to continue to try to send.
Thanks!
1) Use "mailq" command to find out queue id of "stalled" message(s)
2) Push the message in verbose mode (usually requires root privileges)
sendmail -v -qI_the-queue_id_
It should show you very detailed indicator of what causes problems
--
[pl>en: Andrew] Andrzej Adam Filip : an...@priv.onet.pl : an...@xl.wp.pl
Open-Sendmail: http://open-sendmail.sourceforge.net/
I tried sendmail -v -ql_lBEMFmrY029698 but get this error??? Any
ideas? Probably something simple???
sendmail -v -ql_lBEMFmrY029698
Invalid time unit `l'
Invalid time unit `_'
Invalid time unit `l'
Invalid time unit `B'
Invalid time unit `E'
Invalid time unit `M'
Invalid time unit `F'
Invalid time unit `r'
Invalid time unit `Y'
Ok...I see now...I'm a bonehead sometimes....anyway. basically not
much to go on...this comes up over and over with the ones stalled in
the queue...
Forcing the attempted delivery of 1 selected messages ..
Running /var/spool/mqueue/lBFGnHMI021689 (sequence 1 of 1)
<US...@hotmail.com>... Connecting to mx2.hotmail.com. via esmtp...
<US...@hotmail.com>... Connecting to mx3.hotmail.com. via esmtp...
<US...@hotmail.com>... Connecting to mx4.hotmail.com. via esmtp...
<US...@hotmail.com>... Connecting to mx1.hotmail.com. via esmtp...
<US...@hotmail.com>... Deferred: Resource temporarily unavailable
I don't understand...most of the time these go through, but when they
get stuck, they seem to stay there...
This is odd...If I keep trying to push these messages out over and
over, they may eventually go out??? No guarentee, but seems to
eventually flush them after many tries?
If I do a nslookup I get an answer...possibly a timeout problem?
Ok...this is odd...there must be something timing out...if I just
constantly push these messages, they will finally be delivered. I'm
pretty sure DNS is ok because I've done several different lookups and
checks and all seems fine...it seems, that once a message gets in the
queue, it will hang for a while and may or may not be delivered. But
now, as I've discovered "thanks to your help" is that I can sit there
and force these out...maybe I should setup a script to keep doing
this, but I shouldn't have to go to that extreme...
If you have any ideas I'd love to hear...thanks:)
It seems that your sendmail is unable to contact hotmail.com MXes.
*IF* you *CAN* telnet port 25 on the above listed hosts just after such
"push test" *AND* get SMTP greeting
*THEN* it is a sendmail problem
*ELSE* must must blame somebody/something else.
Most likely its not sendmail problem but do the "finally conclusive" tests:
telnet mx1.hotmail.com 25
telnet mx2.hotmail.com 25
...
--
Does your host/sendmail send many messages to hotmail.com?
One possibility is some "3 wheels present" system limiting incoming
messages at the receiving end.
[ From a joke: 3 wheels present = 1 wheel missing ]
--
This seems to be random with others as well...but if I keep trying to
flush the queue they will eventually be delivered....
I can telnet directly to port 25 without a problem...
LOL! 3 wheels!!!
Well...seems that running sendmail -q from cron corrects the problem,
but I've never had to do anything like this in the past. I'm not sure
if this version of sendmail is buggy or maybe just a problem with the
server it's self.
>> > This seems to be random with others as well...but if I keep trying to
>> > flush the queue they will eventually be delivered....
>>
>> > I can telnet directly to port 25 without a problem...
>>
>> LOL! 3 wheels!!!
>
> Well...seems that running sendmail -q from cron corrects the problem,
It *SHOULD* be no different than periodic queue runs by sendmail daemon.
[ you state it is not -> it is *VERY* interesting *why* ]
> but I've never had to do anything like this in the past. I'm not sure
> if this version of sendmail is buggy or maybe just a problem with the
> server it's self.
1) How often your sendmail daemon processes the queue?
[ -q parameter passed to sendmail when sendmail daemon is started]
2) Do you use hoststatus info? [cached info about previous attempts]
* "grep -i hoststatus /etc/mail/sendmail.cf " should if your sendmail
uses hoststatus info
* "sendmail -bh" (by root) should print hoststatus info
--
This is what is in sendmail.cf:
#O HostStatusDirectory=.hoststat
# single thread deliveries (requires HostStatusDirectory)?
#O Timeout.hoststatus=30m
Hmmmmm....Interesting!!
> This is what is in sendmail.cf:
>
> #O HostStatusDirectory=.hoststat
> # single thread deliveries (requires HostStatusDirectory)?
> #O Timeout.hoststatus=30m
>
> Hmmmmm....Interesting!!
You may consider turning the feature on with relatively short timeout
e.g. 10 minutes.
The idea is to stop your sendmail delivery attempts for a short period
after failed delivery attempt to give "rate control" some time to
forget&forgive.
If you have such problems only with a few destinations you may create
separate queue group for them (one for all or one per every).
It will help sendmail send more messages over single SMTP connection to
the domains.
URL(s):
http://www.sendmail.org/m4/tweaking_config.html#confHOST_STATUS_DIRECTORY
http://www.sendmail.org/m4/tweaking_config.html#confTO_HOSTSTATUS
--
You are my hero! I would have had a tough time getting through this
without your suggestions. I've installed sendmail at least 50 times
and never really had a problem...probably because being such small
scale, but this is the first time I've seen this kind of behavior.
Seems like it happens more often or consistently with roadrunner and
comcast.
Thanks again for your help!!!
Thanks Much for the info Res...really appreciated! I was planning to
upgrade sendmail, but this was one of those emergency installs...our
single server died, so I just ran with what I had installed at the
time to get it on line asap. I've just started working with this
company...but I do have plans to get a backup mail server online.
I'll have to keep an eye on charter1 and their mailstore:)