Plesk 12.x / Linux :: Too Many Received-SPF Headers
Dec 15, 2014
I maintain a server with a few domains. The server runs Plesk and Mailman. There are about 50 mailing lists with number of subscribers ranging from a couple people to a few hundred people. All work fine.
There is an additional list with around 2500 subscribers. Messages sent to this list are being received with nearly 500 identical "Received-SPF: pass" lines. Besides seeming odd, it's particularly problematic as some domains are bouncing the emails because the header is too large.
This behavior started after I upgraded from Plesk 11.x to 12.x.
The SPF lines look like this:
Received-SPF: pass (SERVER-HOSTNAME-HERE: localhost is always allowed.) client-ip=127.0.0.1; envelope-from=BOUNCE-LIST-ADDRESS-HERE; helo=SERVER-HOSTNAME-HERE;
Where SERVER-HOSTNAME-HERE is our hostname and BOUNCE-LIST-ADDRESS-HERE is the list's bounce address.
For what it's worth the server hostname and the domain of the list do not match.
This issue seems to be tied to the number of subscribers. We made a new list with a few subscribers and messages looked fine. We added the 2500 subscribers to the new list and then we saw the plentiful Received-SPF: pass entries.
Not sure if all notifications are working or not but I do know that when I create a new customer with a new subscription no notification is received. I have tested with several email addresses.
My wordpress installation in plesk some times give me:
[Sun Apr 12 14:26:08 2015] [warn] (104)Connection reset by peer: mod_fcgid: error reading data from FastCGI server [Sun Apr 12 14:26:08 2015] [error] Premature end of script headers: index.php
How do I stop getting all those Received-SPF: headers prepended to the mailman list? It tags on a header that's well over 32K for a message that is less than 1K.
When I set up spamassassin to store spam in the spam folder of my mailbox, it doesn't add any X-Sapm headers nor tags the subject. All I found so far was an old thread where a plesk employee stated that it will be fixed in 'a' future version.
I have an issue with running of the php-scripts. It runs via Cron (as it consumes a lot of memory), after aprox. 10 minutes after the start-time of the script the following error appears in the error-log file (path - /var/www/vhosts/system/subdomain.domain.tld/logs/error_log)
[Tue Feb 17 19:46:57.003326 2015] [fcgid:warn] [pid 28629] (104)Connection reset by peer: [client XX.XXX.XXX.XXX:48450] mod_fcgid: error reading data from FastCGI server [Tue Feb 17 19:46:57.003368 2015] [core:error] [pid 28629] [client XX.XXX.XXX.XXX:48450] End of script output before headers: Name_of_the_script.php
For a very strange reason another PHP-script, that has the same code, but just runs less arrays, and therefore consumes less memory and can be run via browser - works perfectly from both browser and the cron. The php.ini for this domain set up correctly (see the attachment)...
We are facing this problem on a Plesk (serverpsa v8.6.0_build86080910.19 os_CentOS 5) The scenario is that on this server we have around 5 domains, one of them is the main domain that is also in the hostname.
On the main domain some custom social networking PHP script is running that sends out mails using SMTP authentication, now when welcome mail or invites or notification mails or any such mail are sent through the script, it includes other domains in the headers randomly:
Received: (qmail 15896 invoked from network); 24 Nov 2008 18:45:05 -0500 Received: from somedomain.com (HELO server.maindomain.com) (xyz.xyz.xyz.xyz).....
I have a problem with my site where users log in (vbulletin) and yet the script doesn't log them in. After some research i found out this happens due to http headers.
I own a dedicated server which has plesk as the control panel, i do not know where to go and find the option to enable Http Headers.
I am running plesk 11.05 Plesk on windows server. I have noticed this things twice...
While migrating accounts one server to another Smartermail not showing actual mail received date. For all accounts and emails Smartermail showing Migration date..
My more than 10000 emails showing migration date as received date..
My server seems to be set up okay and almost everything works except for email.
Sending using Outlook or Telnet both appear to send okay and checking for email also appears to work okay. It is just that no emails are ever received (neither sending or receiving produce any errors on the email client).
In my logs I do see an error: Jan 2 11:15:05 cobalt sendmail[29380]: l02AF5mA029380: localhost [127.0.0.1] did not issue MAIL/EXPN/VRFY/ETRN during connection to MTA Jan 2 11:15:05 cobalt dovecot: pop3-login: Aborted login: rip=127.0.0.1, lip=12
I have a detected server and it host one domain with many sub domains account. I never used the cpanel DNS service, instead I used the hosting company’s DNS. .Everything was working fine (including the email) until I decided to have my own DNS. I used the domain register company DNS service to create a DNS records for the domain and all my sub-domains. The problem is the email did not work yet!! Here what I done with the DNS :
eg.com my domain and 1.1.1.1 the server IP address (e.g.) Type--- Name ---- Domain -- Data ---Pref A---- ---- eg.com ---- 1.1.1.1 A---- *. ---- eg.com ---- 1.1.1.1 A---- forum. ---- eg.com ---- 1.1.1.1 A---- mail. ---- eg.com ---- 1.1.1.1 MX---- ---- eg.com ---- 1.1.1.1----0 MX---- *. ---- eg.com ---- 1.1.1.1----0
I update the nameserver in the cpanel basic setup. It should work fine, I don’t know what goes wrong? I never touch the server configuration, the server was relay on external DNS service.
Here are the error that I got when I tried to send an email to the server :
info@eg.com SMTP error from remote mail server after RCPT TO:: host mail.eg.com [1.1.1.1]: 550-Verification failed for ....
A client is sometimes not receiving email (ie. they are informed by phone that one was sent a day or two earlier but they didn't receive it). When this happens the sender doesn't get any error message back. I had the server logs checked out for an email I sent the client that he didn't receive and the server did receive it. I logged on to the webmail and it wasn't there, but client says it's not in his inbox. I asked them to check all spam or other folders in his email software but to no avail. The problem can happen with emails that have attachments and those without. It's not happening for any particular sender either. My client cannot see any pattern to it.
I wanted to try HyperVM/LxAdmin setup from sometime now. I finally have one but I am getting trouble with its Qmail MTA. I am posting here hoping to get some help.
Let me tell you the issue.
I have my main vps IP: 10.10.10.22 (I have 3 additional ips).
I have setup 2 domains on their dedicated IPs. (10.10.10.23 & 10.10.10.24). However, whenever I send an email, I get outgoing IP to be 10.10.10.22. I searched on Google and found out that I need to patch qmail with outgoingip.patch to change the outgoing IP.
But how to patch my already installed qmail??? I tried but failed. I am running Centos 5.3 with HyperVM/Lxadmin
Also, even if I host a domain on main IP, my SPF record is always
Quote:
Received-SPF: neutral (google.com: 10.10.10.22 is neither permitted nor denied by domain domain.com) client-ip=10.10.10.22;
Sending and receiving from yahoo,google mail is just fine.
But when i try to receive email from another server or an localdomain on same server, i receive an email with the below content:
This message was created automatically by mail delivery software.
A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed:
support@xxx1.com SMTP error from remote mail server after RCPT TO:<support@xxx1.com>: host xxx1.com [xxx.xxx.31.49]: 550-Verification failed for <admin@xxx2.com> 550-No Such User Here 550 Sender verify failed
I am not sure it is the issue of sender or receiver?
I am using Cute FTP 7.0 Professional. The FTP server that I am trying to connect is firewall protected and hence I am using port method. But still I am facing problems in logging in. The following is the transcript of the command window:
*** CuteFTP 7.0 - build Jun 7 2005 ***
STATUS:> Getting listing ""... STATUS:> Connecting to FTP server 216.185.43.181:21 (ip = 216.185.43.181)... STATUS:> Socket connected. Waiting for welcome message... 220 Microsoft FTP Service STATUS:> Connected. Authenticating... COMMAND:> USER bizconsultancy 331 Password required for bizconsultancy. COMMAND:> PASS ***** 230 User bizconsultancy logged in. STATUS:> Login successful. COMMAND:> PWD 257 "/bizconsultancy" is current directory. STATUS:> Home directory: /bizconsultancy COMMAND:> FEAT 211-FEAT SIZE MDTM 211 END STATUS:> This site supports features. STATUS:> This site supports SIZE. COMMAND:> REST 100 350 Restarting at 100. STATUS:> This site can resume broken downloads. COMMAND:> REST 0 350 Restarting at 0. COMMAND:> PORT 192,168,0,14,7,111 200 PORT command successful. COMMAND:> LIST 150 Opening ASCII mode data connection for /bin/ls. ERROR:> Can't read from data socket. Socket error = #10054. 425 Can't open data connection. ERROR:> Trashed response received.
Some of the e-mails I send are received several times by my recipients. I use Outlook Express. I send the e-mail only once but some people are complaining that they receive the same message as many as 10 times sometimes. Do you know what can be causing it?
I always presumed the Expires: date is part of the page cache instructions, however I was told its to do with the cookie. I thought the cookie expire was set as expires=date in the set cookie line and not on a new line with the :
I have a dedicated server and I am using the SMTP of one of my domain names to send mail. The IP of this domain name ends in .25 and the server hostname IP ends in .24.
The emails that are received list the server hostname and IP in the message headers instead of the IP of the domain name that is sending the email. Is there something that can be done about this? I read that a message could be mistaken as spam if it doesn't originate from the exact IP of the domain in the 'from' field.
I'm in the process of moving some websites between hosts... from Dreamhost to a VPS.
Basically I want to make sure that no emails get lost in the transfer, as it's possible some people haven't collected their emails. I'm only talking about 5 or 6 sites here with about 15 mailboxes in total.
I have pop3 set up on the VPS with exim4 and courier using Maildir's. I have imap/pop3 access to emails on dreamhost and all my clients only use pop3.
Dreamhost doesn't provide direct access to the Maildir's of users, so I was planning on using offlineimap to save a copy of each mailbox and move it into the Maildir of the user on the VPS. I.e. move the contents of INBOX/new into user/Maildir/new etc. I'll do this once the dns settings have propogated so I can be sure to have every single email.
Does this sound reasonable? Is anything likely to break? Will any email clients break with any of these transferred emails? I tried it with a test account and it seemed OK, but I would like an experts opinion on whether or not this is the best way of achieving what I'm trying to do
I'd love to contact the domain registrars of the recipient spammers who are barraging me with the latest "google riches", or "ebay success" sales pitch emails, but I lack the knowledge of how to reveal the email headers and interpret them.
Is there a particular email client program that would allow me to view all parts of the email ? Maybe I'm using the wrong client (outlook), that would not allow me to view the more technical parts of the email, but but I don't know.
If I go ahead and click on their "unsub me" links at their bottom, that's their tricky way of finding out what emails are active and possibly make it worse for me. I know that it's illegal to send me email that I didn't opt into, in the first place, even if they do provide the unsubscribe link at the bottom, with the usual "Icann" spam disclaimer that they include.
I have a problem where all email sent from uaccb.edu (studentmail.uaccb.edu [208.190.226.130]} is not arriving at my VPS. The people sending email from the above domain are not getting any errors, it is not being marked as spam, it is just being blackholed somehow.
I have checked the above IP and domain, and can't find it on any blacklist, and I am using pretty standard cPanel setup on my VPS. I have spam set to be marked and receive a fair amount and have it autorouted to a folder rather than auto-deleted.
From uaccb.edu, the only thing I have access to is Outlook Web Access to send test emails, and they are sent fine with no errors returned. If I send test emails to gmail and other similar email accounts, they go through fine, the only problem is to ANY domain hosted on my VPS. I have even recently transferred all the domains to a second VPS (with the same hosting company) at a different datacenter (for different reason) and the problem persists.
Is there any WHM setting or any ideas on how to troubleshoot this?
I am having a peculiar problem, when ever I attempt to access my server globally I get the dreaded 'No Data Received' error.
The machine is running Ubuntu 13.10 and is a dedicated server (I do not have GUI or physical access to the machine). All ports are open, iptables disabled, yet nothing works (running on port 80). I loaded the files onto a different computer at home and it works nicely, but not on the dedicated server.
Also, when the Apache server is off it says it can not connect in my browser, and when the server is on it says 'No Data Received' ...
