https://bugs.exim.org/show_bug.cgi?id=2693
Jeremy Harris changed:
What|Removed |Added
Status|WAIT_FIX_CONFIRMATION |RESOLVED
Resolution|---
https://bugs.exim.org/show_bug.cgi?id=2693
Git Commit changed:
What|Removed |Added
CC||g...@exim.org
--- Comment #20 from Git Commit ---
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #19 from Jeremy Harris ---
Ideally, yes. But a different bug, not this one. Possibly only an RFE; it's
not wrong to take queue-run iterations to deliver to recipients that get a 4xx.
Meantime, I added your non-pipeline testcase, commenting
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #18 from Simon Arlott ---
Created attachment 1366
--> https://bugs.exim.org/attachment.cgi?id=1366=edit
Test case without pipelining
Ideally the behaviour without pipelining should be the same. Currently Exim
will need to retry 3 times
https://bugs.exim.org/show_bug.cgi?id=2693
Jeremy Harris changed:
What|Removed |Added
Status|ASSIGNED|WAIT_FIX_CONFIRMATION
--- Comment #17 from
https://bugs.exim.org/show_bug.cgi?id=2693
Jeremy Harris changed:
What|Removed |Added
Priority|medium |high
Status|NEW
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #15 from Simon Arlott ---
Add to the SMTP transport:
hosts_avoid_pipelining = *
--
You are receiving this mail because:
You are on the CC list for the bug.
--
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #14 from Moritz Friedrich ---
Thank you very much for analysing and reproducing the problem in detail.
At the moment, a workaround would be to send each message in a separate
connection, wouldn't it?
Should it be sufficient to set the value
https://bugs.exim.org/show_bug.cgi?id=2693
Simon Arlott changed:
What|Removed |Added
Attachment #1359|0 |1
is obsolete|
https://bugs.exim.org/show_bug.cgi?id=2693
Moritz Friedrich changed:
What|Removed |Added
Version|N/A |4.94
--- Comment #12 from Moritz Friedrich
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #11 from Jeremy Harris ---
Ah, it was unclear thanks to the obfuscation of names. Simon seems to be
saying
(comment 3) that it works for him, although (4) obscures that somewhat.
It's also unclear what triggered the second connection 18
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #10 from Benoit Panizzon ---
Hi Jeremy
Look at Simon Arlott's post about the recipients and how they got rejected.
He has the tcp caputre file (not uploaded as it contains the real email
addresses of our customers).
C defer
D defer
They
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #9 from Moritz Friedrich ---
Created attachment 1360
--> https://bugs.exim.org/attachment.cgi?id=1360=edit
exim configuration
--
You are receiving this mail because:
You are on the CC list for the bug.
--
## List details at
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #8 from Jeremy Harris ---
I'm not seeing the problem. Six recipients were delivered to according to
Exim's log. Was that the number expected? Were the messages in fact received
by the destination accounts?
--
You are receiving this mail
https://bugs.exim.org/show_bug.cgi?id=2693
Moritz Friedrich changed:
What|Removed |Added
CC||m...@cyon.ch
--- Comment #7 from Moritz
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #6 from Benoit Panizzon ---
Waiting for feedback from the tech @ sending ISP to learn about the used exim
version and his logs.
Simon, your observation is correct.
The first six recipients are sent in one TCP session. The failed are
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #5 from Simon Arlott ---
There is a slight variation in the observed scenario. The first two attempts
occur in the same connection. The final attempt is in a new connection several
seconds later. My test case is using three separate
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #4 from Simon Arlott ---
Created attachment 1359
--> https://bugs.exim.org/attachment.cgi?id=1359=edit
Pipelining test case
Exim works correctly when I test it
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #3 from Simon Arlott ---
The observed behaviour of the three delivery attempts is:
A ok
B defer
C defer
D defer
E defer
F defer
B ok
C defer
D defer
E ok
F ok
E ok
F ok
The recipients are sent in alphabetical order, and they are in this
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #2 from Benoit Panizzon ---
Just got feedback from sender ISP (exim) customer. He re-tested while
pipelining off on our postfix. Result: Problem does not occur. So it is very
likely related to pipelining.
--
You are receiving this mail
https://bugs.exim.org/show_bug.cgi?id=2693
--- Comment #1 from Simon Arlott ---
I've already asked you to provide an unredacted test case, and the version of
Exim used. If you try running Exim yourself you will be able to produce this.
I can't find a test case for this specific scenario (0217
21 matches
Mail list logo
