Apache conf/etc

sslsite

New member
Joined
Sep 22, 2005
Messages
4
Hello all,

We are thinking about using directadmin for dedicated servers.

We have a few questions.

1) Does direct admin use the apache include directive to add sites so they have their own http.conf? We host between 1,000 and 2,000 domains per server and we found if all the sites are in one httpd.conf it takes too long to rebuild.

2) Does direct admin run cgi's under suexec and log the connections?

3) What mail server does direct admin use and does it record sendmail connections to find bad formmail scripts?

Thanks for anyone's input. We really appreciate it :-)

- Dave
 
I'll answer what I can...

1) Yes

2) You need to install this yourself, once installed it will work.

Question 3 I'll leave to more qualified people to answer.


Regards,
Onno Vrijburg
 
3) DirectAdmin uses exim.

I don't know how you'd use logging to find bad form-to-email scripts, but exim's logging is quite configurable.

Jeff
 
Hi,

Thanks for all the info.

1) If we enable suexec is there an easy way for our customers to add the user directive to their sites httpd.conf?

2) Does exim have a sendmail wrapper to replace /usr/sbin/sendmail ?

If so we could probably use a perl script to log connections to sendmail from php and cgi scripts. That's what we do with qmail.

Thanks again :-)

Dave
 
No, and yes.

Only the admin can change the default httpd.conf file, and the changability is quite rudamentary <frown>.

Jeff
 
Hi,

That's too bad but not so bad.

We have been using Plesk since pre sw-soft days and it use to be really smooth on upgrades but now it seams every upgrade has a problem plus we pay more for software than hardware ?:-)

That said even if scripts run as nobody (as they do in php anyway) top should still show the offending "hog" and we could use find to find the offending script.

How do upgrades go with Directadmin when you have 1,900 sites and about 90GB of used space...

I know that's a loaded question but we host allot of low bandwidth sites and everything runs fine until we do upgrades.

I also like how active these forums are and all the responses I have received.
Thanks for all the input.

If we decide to go with directadmin I will register under our company name "hostexpress" and become more active in helping users in any way I can.

Thanks again,

Dave
 
Dave, we were Plesk Gold Partners long for the buyout and I know what you mean.

I don't see why having lots of sites would make a difference in updates, but any way you look at it it's going to mean starting up httpd and named is going to take a while, and in any control-panel based server just when they get restarted is NOT under your control.

We stop after a few hundred sites. Even with cheap sites our servers more than pay for themselves.

Jeff
 
Hi Jeff,

Thank you so much for your reply.

We have allot of resellers...

What we have fond is upgrading from Plesk 7 to 7 reloaded requires 2 dns updates of all dns records and the upgrade tries to resolve each zone.

This takes allot of time (over 2 hours) with 1,000 sites. The total upgrade time was 4 hours with 1,900 sites, on dual cpu Dell server. If it fails and we have to restore 90GB of data... It happened twice last month.... There is no simple revert from an upgrade that fails.

Please keep in mind all is well unless we upgrade.

That said Plesk upgrades are not fun. We never had these issues with Plesk 5 and 6, pre sw-soft days.

Oh, by the way, I am just venting, I really appreciate your input :-)

Thanks again Jeff,

- Dave
 
Hi Dave,

I have several servers with 600-800 sites per server and the last upgrade of DA took 3 minutes per server. DA does not upgrade Bind/Apache/Exim/.... when you upgrade the control panel you do only that you upgrade the control panel.

Other upgrades like Apache/Exim take longer but that is to be expected.

Regards,
Onno Vrijburg
 
Back
Top