can not send email through pop3

cool38

Member
Aug 1, 2006
12
0
151
Hi,

one of my client told me that he can not send email through pop3.

however, i tried his account on my computer, and i can send it by pop3.

so, i told him how i set pop3 on outlook, exactly the same settings as mine,

but he still can't send email.

here is the log i found under /var/log/exim_mainlog

2006-08-07 02:53:27 H=61-230-141-179.dynamic.hinet.net (PC251709360118) [61.230.141.179] F=<[email protected]> rejected RCPT <[email protected]>: 61-230-141-179.dynamic.hinet.net (PC251709360118) [61.230.141.179] is currently not permitted to relay through this server. Perhaps you have not logged into the pop/imap server in the last 30 minutes or do not have SMTP Authentication turned on in your email client.
2006-08-07 02:54:32 H=61-230-141-179.dynamic.hinet.net (PC251709360118) [61.230.141.179] F=<[email protected]> rejected RCPT <[email protected]>: 61-230-141-179.dynamic.hinet.net (PC251709360118) [61.230.141.179] is currently not permitted to relay through this server. Perhaps you have not logged into the pop/imap server in the last 30 minutes or do not have SMTP Authentication turned on in your email client.
2006-08-07 02:55:43 fixed_login authenticator failed for (PC251709360118) [61.230.141.179]: 535 Incorrect authentication data ([email protected])



can anyone tell me how i can fix this problem?

because it seens like the exim server is fine.

note : i've told him to receive the mail first, then try to send email. (pop before smtp)
 

cool38

Member
Aug 1, 2006
12
0
151
it seems like the pop before smtp script, only adds MY ip to /etc/relayhosts file and /etc/relayhostsusers

but it doesn't add other client users.
 

cool38

Member
Aug 1, 2006
12
0
151
webignition said:
The user should be able to send using standard SMTP authentication.

Have you tried this?
i've tried that. but it didn't work.

i've tried to login with the same computer, but different ISP (of cuz it's a different IP) to check and send emails.

BINGO.

my client was right. his outlook setting was fine.

i can only see my pervious IP was added to the files
/etc/relayhosts
and /etc/relayhostsusers

but my new IP was not added to the file.

that's so strange
 

cool38

Member
Aug 1, 2006
12
0
151
hmmn..
it seems like the maillog file was not being written.
so, i ran

/scripts/restartsrv_syslogd

and it actually fixed my problem.

Hmmn..

my exim went strange these days.
 

nxds

Well-Known Member
Jan 6, 2006
53
0
156
Sounds like the CentOS/RH log rotation bug when TMPDIR is mounted noexec. It will probably happen again (maillog not getting written to) if you don't make a permanent fix.
 

carock

Well-Known Member
Sep 25, 2002
272
9
168
St. Charles, MO
Is it possible to change the postrotate script to /scripts/restartsrv_syslogd instead of

/bin/kill -HUP `cat /var/run/syslogd.pid 2> /dev/null` 2> /dev/null || true

Or would that not work?

Chuck
 

carock

Well-Known Member
Sep 25, 2002
272
9
168
St. Charles, MO
I was having a similar issue with one customer on the box. several accounts could successfully POP their mail account, but the sending mail was not authenticating.

exim_rejectlog:

fixed_login authenticator failed for (salem) [69.153.102.49]: 535 Incorrect authentication data ([email protected])

POP3 was downloading mail just fine though.

I fixed this person's account by going into cPanel and clicking the Change Pass button to reset the password for this mailbox. I used the same password they had before.

Now they can send and receive without error.

My question is though, why or how is the SMTP authentication using a different password database than POP3???

Also, is there a way to just resynchronize them instead of resetting the password on the account?

Thanks,
Chuck
 

cool38

Member
Aug 1, 2006
12
0
151
nxds said:
Sounds like the CentOS/RH log rotation bug when TMPDIR is mounted noexec. It will probably happen again (maillog not getting written to) if you don't make a permanent fix.

actually..

the bug was not fixed.
i have submitted a support ticket, and cpanel support team came into my server, and worked on this issue for 4 days already.

and yes, i'm running WHM on my Fedora Core 4.

but no. it doesn't look like it's the log rotation bug. (hopefully not)
because the latest log is written to /etc/maillog file