On 9/23/2014 6:56 AM, Francis SOUYRI wrote: > Hello Peter, > > Thank you for your reply, I thought this is multiple messages in a > single connection, but greater than 4GB... :( ? > > Also how can a decode de delays ?
from the docs: "delays=a/b/c/d" where a=time before queue manager, including message transmission; b=time in queue manager; c=connection setup including DNS, HELO and TLS; d=message transmission time. > On 09/23/2014 01:49 PM, Peter wrote: >> On 09/23/2014 11:41 PM, Francis SOUYRI wrote: >>> Sep 23 11:29:42 vador postfix/smtp[4194]: 832AF60511: to=<xxxxx>, >>> relay=xxxxxxx.com[xxx.xxx.xxx.xxx]:25, delay=63197, >>> delays=63191/0.43/5.1/0.19, dsn=2.0.0, status=sent (250 ok >>> 1411464583 qp >>> 1725) a=63191 seconds before queue manager. This message waited ~17.5 hours before the queue manager attempted delivery. Some combination of very slow initial transmission to postfix, and/or the queue is very busy with other mail. b=0.43 seconds in the queue manager. This may seem like not much time, but still high enough that I suspect there's a lot of mail queued. On a lightly loaded system this will be near zero. c=5.1 seconds for the connection setup. That's really slow, but could be from delays at the receiving end. A more typical setup time would be 0.5 or less (but this could be a destination with a 5-second greet wait, and a "real" setup time of 0.1 -- not enough info here to tell). If connections to *all* destinations are slow, this could indicate DNS problems and/or an overloaded network connection. d=0.19 seconds for message transmission. The message transmission was quick once it started. Maybe it was a very small message; no info shown here on the size. Postfix logs the message size in a separate "qmgr" log line. Maybe you have lots of mail backed up in your queue. See the "Problem Solving" section of the docs for more ideas. http://www.postfix.org/documentation.html -- Noel Jones >>> Sep 23 11:29:42 vador postfix/smtp[4193]: 8824560692: to=<xxxxx>, >>> relay=xxxxxxx.com[xxx.xxx.xxx.xxx]:25, delay=62340, >>> delays=62334/0.45/5.1/0.36, dsn=2.0.0, status=sent (250 ok >>> 1411464583 qp >>> 1727) >>> Sep 23 11:29:53 vador postfix/smtp[4250]: 59083603A7: to=<xxxxx>, >>> relay=xxxxxxx.com[xxx.xxx.xxx.xxx]:25, delay=63308, >>> delays=63303/0.06/5.1/0.2, dsn=2.0.0, status=sent (250 ok >>> 1411464594 qp >>> 1825) >>> >>> What is it 1411464583, 1411464583, 1411464594 ? >> >> It's part of the response from the remote server and is >> meaningless to >> postfix. It looks to me like a timestamp (unixtime seconds since >> epoch). >> >>>> After checking de CheckPoint firewall Log, I see some very big >>>> (more >>>> than 1Go) smtp transfers from our Postfix gateway. >>>> >>>> But I have this parameter "message_size_limit = 13631488" and >>>> when I try >>>> to send a bigger mail I am blocked. >>>> >>>> How can is this possible ? The max transfer I see was 4Go ! >> >> Multiple messages in a single connection. >> >> >> Peter >> >
