David G's questions

Is it possible to configure postfix to send via IPV4 only but receive using both IPV4 & IPV6?

My server, running on AWS Lightsail, now can have an IPV6 address. The address is relatively static (as long as I keep IPV6 active on the instance), but it's not truly static because I can't move the address from server to server (like I can do with the static IPV4 address). I also can't (currently) as Amazon to assign an PTR record for the IPV6 address.

As I have the IPV6 address available, I'd like to be able to receive mail using either IPV4 or IPV6 but only send using IPV4.

I know I can set postfix to only listen on the IPV4 address using the inet_protocols setting.

I need some help diagnosing a problem using postfix as my outbound mail server with Mailman.

Up front I want to say, I have this working FINE with my current mailman list server. I'm in the process of moving my list server to an AWS lightsail instance.

And, yes, I know that I could use postfix as both the inbound & outbound mail server ... which I probably should do, but I'm far more familiar with using sendmail and have it setup & working fine on my current list server.

I've got just about everything setup and working fine.

The problem is, I'm seeing this in my mail log...

Nov 15 15:41:43 lists postfix/smtp[18136]: AA76A61E61: to=, relay=none, delay=0.01, delays=0/0/0/0, dsn=5.4.6, status=bounced (mail for localhost loops back to myself)

It appears that postfix thinks it's the mail server for lists.atestdomain.net and can't deliver the mail because it doesn't know how to deliver the message.

Here's my current setup:

I have sendmail listening on the public SMTP ports on the server. I have postfix listening on localhost:24. Mailman is configured with the SMTPHOST & SMTPPORT set to localhost & 24.

So mail is received by sendmail, sent to & processed by Mailman, and Mailman sends the mail out by connecting to postfix on localhost:24.

The list mail is being delivered fine ... but bounces are not getting processed correctly.

I have the domain lists.atestdomain.net listed in the transport map indicating that mail for that domain should be sent to localhost:25.

You can see a relatively complete log fragment at https://code.midrange.com/e665e40f93.html

Here's (what I think are) the relevant postfix config file entries ...

main.cf:

myhostname = listmail.arealdomain.com
mydomain = $myhostname
myorigin = $mydomain
inet_interfaces = localhost
mydestination = $myhostname, localhost.$mydomain, localhost
transport_maps = hash:/etc/postfix/transport

transport:

lists.atestdomain.net smtp:localhost:25
listmail.arealdomain.com smtp:localhost:25

master.cf

24      inet  n       -       n       -       -       smtpd

Any suggestions on what to look at?

I've got some joker trying to subscribe a bunch of email addresses to my mailman system and I want to block him from accessing the apache web server.

According to the apache log, he's accessing from a very consistent hostname ... variations on x12356.example.com.br (where the numbers changed).

The problem is, the actual hostname that's appearing in the log doesn't resolve ... but, once I was able to get the IP (by turning off hostname resolution), I found that the IP did resolve on a reverse IP lookup.

I tried the following in my htaccess file, but it's not blocking.

RewriteCond %{REMOTE_HOST} ".*\.example\.com\.br"
RewriteRule ^.* - [F,L]

I also tried a simple 'Deny from', but it's not blocking either.

Deny from .example.com.br

My gut says I'm missing something simple, but I can't see it.

Here's the complete .htaccess file...

RewriteEngine on
RewriteBase /
RewriteCond %{HTTP_HOST} (.*)\.l\.mydomain\.com [NC]
RewriteRule ^$ http://lists.mydomain.com/listinfo/%1 [R=301,L]

RewriteCond %{HTTP_REFERER} mailbait [NC]
RewriteCond %{REQUEST_URI} !^/mailbait.php$
RewriteRule ^.*$ /mailbait.php [L]

SetEnvIfNoCase Request_URI "(testproxy|upload|xmlrpc|wp|blog|wordpress)" spammer=yes

RewriteCond %{ENV:spammer} ^yes$
RewriteCond %{REQUEST_URI} !^/tarpit.php$
RewriteRule ^.*$ /tarpit.php [L]

RewriteCond %{REMOTE_HOST} ".*\.example\.com\.br"
RewriteRule ^.* - [F,L]
Deny from .example.com.br

I've got a Fedora 10 system that's exporting some file systems via NFS.

One of the exported file systems is used very heavily on a client system when I run a search engine indexer. The file activity is 100% read (no writes).

Whenever I run the indexer the systems load average skyrockets (up to 25 at points) and rpcbind is often using 98% of the CPU capacity.

On the client system I'm mounting the remote directory with the following options: intr,nosuid,ro,rsize=8192,udp,noatime,bg,nodev

On the server, I'm exporting the directory with the following options: ro,no_subtree_check

I have the NFS server configured for 32 threads.

If it makes any difference, the NFS server is also a NIS server. And the client is bound to it as a NIS client.

Any suggestions on what I can do to reduce the load on the server? Having such a high load just doesn't seem normal to me.

Thanks!

I'm trying to setup the watchdog daemon to run on my Fedora 12 system.

I know exactly what I want it to do ... but I'm unsure as to exactly configure it. Mainly due to inexact documentation.

What I want to do is have my system reboot if a specific file doesn't change within a certain amount of time.

To do this I'll use the file = <filename> and change = <mtime> configuration options.

The problem is, I can't figure out what value I should use for mtime.

I want the system to restart if the specified file (a web server log file) hasn't changed in 10 minutes. What value would I use for mtime?

Thanks!