• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Login Problem

J

Joern@

Guest
Hello all together,

I have a really big problem to reach the admin interface for one of my Servers working with PLESK 7.50 .

Two days before everything is working correkt but today there is no LOGIN.PHP on my screen.

There is no error, it is only a white scrren without Loginfield.

I thought after restarting the server erverything will be OK, but it fails.

The apache Server seems to be running correctly. The only way to get access to the server is via ssh. That works.

So, is that problem ever known before? What can I do for running PLESK well?

Thanks for help?

Greetings
Jörn
 
Usually a blank screen on a php page indicates that php is erroring out rather than processing. Trying turning php errors, hopefully the errors will help indicate where the problem lies. You can turn on php errors by setting default_errors=On in the /etc/php.ini file. You'll need to restart apache once you've turned errors on.
 
Hello,

the point default_errors does not exist in my php.ini.

So, what I have done is to turn on this points:

display_errors On
display_startup_errors On

but the screen stays white actually. There is no error after opening the login.php3

Maybe, it is a posibility to replace some php.files in

/usr/local/psa/admin/htdocs/ ??

The fact is that other php applications are running well. It's not a nice feeling to work without the plesk interface...

Thanks
Jörn
 
There are a number of possibilities:

1) Did you recently run an update to Plesk?

After you turned on error logging, did you then check the logs via the SSH connection?

There have been several posts in recent past regarding this same or similar problem.

If you recently updated Plesk and it never even gets you to the screen to put in the admin username and password, then you may have to force a re-install of your rpms, see post:

http://forum.plesk.com/showthread.php?s=&threadid=24000&highlight=login+blank

2) If you are able to enter the admin and password and then get the blank screen, see this post:

http://forum.plesk.com/showthread.php?threadid=25322&highlight=blank+AND+login

These are just 2 possible similar cases, hopefully will be helpful, but not sure of course.
 
Ok, thanks for answering ...

I have done nothing like updating Plesk or my Suse System.
I also can't login to Plesk because of the invisible Login fielt.

The good pint is that I think I have found the problem.

I checked this file:

/usr/local/psa/admin/logs/httpsd_access_log and
/usr/local/psa/admin/logs/httpsd_error_log

For today there is the following message:

mod_ssl: SSL handshake failed: HTTP spoken on HTTPS port; trying to send HTML error page (OpenSSL library error follows)
[Tue Jul 19 12:22:02 2005] [error] OpenSSL: error:1407609C:SSL routines:func(118):reason(156)
[Tue Jul 19 15:35:34 2005] [notice] caught SIGTERM, shutting down
Processing config directory: /usr/local/psa/admin/conf/httpsd.*.include
[Tue Jul 19 15:37:08 2005] [notice] Apache/1.3.33 (Unix) mod_ssl/2.8.22 OpenSSL/0.9.7d PHP/4.3.9 configured -- resuming normal operations

------------------------------------------------------------------------------------

That sounds like a broken ssl module. Thats why I get nothing else a while screen. Plesk works with SSL

So what do I have to do now?

Thanks
 
First, just to verify, you are actually browsing to the https url with :8443 on the end, right? Such as:

https://www.yourdomain.com:8443

Second, this type of error was reported quite some time ago related to an Apache/mod_ssl worm. The vulnerability exploited by the Apache/mod_ssl worm
was fixed beginning with OpenSSL version 0.9.6e, so unless there has been another worm which can exploit 0.9.7d, this may not be the problem, but something to consider...

Third, if the above URL is correct, *and* if no changes were made to files such as httpd.conf, httpd.include, *and* no changes to vhost.conf or vhost_ssl.conf or .htaccess, such as rewrite rules... then you could try re-installing openssl to see if that resolves the problem.

You could try forcing a re-install of the openssl package. First you need to find where the openssl rpm file is, from the SSH command prompt, do:

locate openssl-0.9.7d

You should get a location for the .rpm file.
(location will be server specific, so I cannot give you a definite path)

This will also tell you the exact name of the file (also server dependent)

Change directory into the location, and issue the command:

rpm -Uvh --force openssl-0.9.7d-whatever else.
(use the exact filename as reported by the locate command above)

If all goes well, that should cause the openssl package to be reinstalled. Then, personally, I would restart the server, but you may not have to.

That's all I can think of right now. Hopefully it is just that you are not using the HTTPS: on the url.
 
Damm,

what a really bad situation.

I have done everything you told me and the installation was OK:
Here is the alert of installation :

[Wed Jul 20 14:31:17 2005] [notice] caught SIGTERM, shutting down
Processing config directory: /usr/local/psa/admin/conf/httpsd.*.include
[Wed Jul 20 14:32:52 2005] [notice] Apache/1.3.33 (Unix) mod_ssl/2.8.22 OpenSSL/0.9.7d PHP/4.3.9 configured -- resuming norma
l operations
[Wed Jul 20 14:32:52 2005] [notice] Accept mutex: sysvsem (Default: sysvsem)


sounds nice, but my screen is white either.

Its right : I get access to my interface with

https://192.0.0.0:8443 --
When I try to connect to this URL: https://192.0.0.0 I get a good
response.

So, I do not know what to do next.
Which posibilities do I have?

Thanks for response..

Jörn
 
Do you have any other browser, such as Firefox or Mozilla which you can use to test?

I have recently had a number of consulting (non-hosting) clients who's WinXP autoupdated and now they cannot browse to some sites such as financial and stock trading sites, they also get blank white screens.

Just thought I'd mention it...

There have been some posts from people who had problems with one browser, but they tried a different one and was able to get things to work. Not a solution, but maybe a viable workaround.
 
Hello together,

jamesyeeoc, thanks a lot for responping all my questions.

It ' s definineily not a problem with a browser, because I reach another Server Plesk System with the same session.

So I get a illuminating alert from my Opera Browser:

Repaeted attemps failed to load this page completly. Tere maybe a problem on the server.

That is it...

Than I have remembered to this information from the psa_log:

[Wed Jul 20 14:32:52 2005] [notice] Accept mutex: sysvsem (Default: sysvsem).

I get the same notice for today.

I have found that:

File list
-rwxr-xr-x root root 9707 Jul 1 19:09 /usr/lib/php/extensions/sysvsem.so

That rpm: php4-sysvsem-4.3.4-43.8.i586.rpm
Maybe I should try forcing a re-install of this package.
What do you mean??

Best regards
Jörn
 
You could always force a reinstall of all the PHP related packages, I don't think it would hurt anything, and it will eliminate those as problems.

Always do backups before making changes
 
Back
Top