Topic: No Server Manager or Website after restore

[jamesn]

We had a hard drive going bad on a SME7.  Replaced the drive, installed a fresh copy, then restored the backup using the server-manager.  After that, everthing works but the webserver.  Can not access the server manager or the website on the box.  httpd error log shows:

[Tue May 15 13:14:17 2007] [error] SSL Proxy requested for wlpmm-firm.com:80 but not enabled [Hint: SSLProxyEngine]
[Tue May 15 13:14:17 2007] [error] proxy: failed to enable ssl support for 10.0.0.254:443 (10.0.0.254)
[Tue May 15 13:26:40 2007] [error] SSL Proxy requested for wlpmm-firm.com:443 but not enabled [Hint: SSLProxyEngine]
[Tue May 15 13:26:40 2007] [error] proxy: failed to enable ssl support for 10.0.0.254:443 (10.0.0.254)

Been searching the site for days, with no luck.

Thanks for any help.

[cjensen]

Did you do:

Code: [Select]

signal-event post-upgrade

Code: [Select]

signal-event reboot

Then if that does not work come back with log errors from relevant logs.

Craig

[jamesn]

Yes, I've done the post-upgrade and have rebooted many times.  Can you tell me which logs to post.  I gave the http error log file in the fist post.

Thanks for the help.

[cjensen]

What is the status of your services:

Code: [Select]

service httpd-admin status
and

Code: [Select]

service httpd-e-smith status

other logs that may help:

Code: [Select]

/var/log/httpd/admin_error_log

Code: [Select]

/var/log/messages

Did you change any of your domain info with the migration?

Craig

[cjensen]

Also check the last two posts here:
http://forums.contribs.org/index.php?topic=35786.0

Craig

[jamesn]

I've looked at that post.  When I set SSLEngine on  I get a squid message saying (111) Connection refused.  Also, just the website on the server without ssl (http port 80) will not work.

With SSLEngine off, I do get the IE certificate message, then after accepting, I get the Internal Server Error.

Here are the items you asked for:

/service/httpd-admin: run (pid 4663) 84092 seconds, normally down

/service/httpd-e-smith: run (pid 4791) 82926 seconds, normally down

[root@iserver ~]# cat /var/log/httpd/admin_error_log
[Tue May 15 01:12:10 2007] [notice] Digest: generating secret for digest authentication ...
[Tue May 15 01:12:10 2007] [notice] Digest: done
[Tue May 15 01:12:10 2007] [notice] Apache configured -- resuming normal operations


The message log does not have any strange errors in it.  It is very large, I can post if you think it will help.  I did not change the domain name durning the switch.  I have since tried changing it in hopes of fixing the problem, (changed name, reboot the server and tried with no luck).  Then changed it back to the current (real) domain name.

[cjensen]

Look at the post by Mortale:
http://forums.contribs.org/index.php?topic=33867.0

Craig

[jamesn]

I've tried removing the cert files and running that script, but still no luck.  Right after the restore I could still access server-manager from the admin menu, but no where else.  Now I can not even access it from the admin menu.

[cjensen]

Have you tried to manually expand the httpd.conf template?

Code: [Select]

signal-event domain-modify

Should have been run with your changes above.  Also I would be sure you are completing all the steps in the last post.

Run this:

Code: [Select]

ipcs -s | grep www | perl -e 'while (<STDIN>) { @a=split(/\s+/); print `ipcrm sem $a[1]`}'

It will remove lock files.

Craig

[cjensen]

also for custom certs, this post from Shad works well:

Custom-cert

Craig

[jamesn]

Yes I did the signal-event domain-modify.  I'm not using any custom certs, just the self signed one.

[CharlieBrady]

Have you tried to manually expand the httpd.conf template?

Craig, please don't try to troubleshoot issues on the forum. Anything that doesn't 'just work' should be taken to the Bug Tracker. Thanks.

[cjensen]

Sorry Charlie, You are right.  I have strung this one out without mentioning that reference.

jamesn, this should be entered in the bug tracker

[jamesn]

I just did a search in the bug tracker and found where someone else has had pretty much the same problem.  I've got to get this server fixed for a client, so I guess I'll just have to rebuild everything by hand.  Unless someone can give me a better idea.

Thanks for the help.

James

[cjensen]

I just did a search in the bug tracker and found where someone else has had pretty much the same problem.

Did you enter your issue there?  Or open a new bug?  That is likely your best bet and that is where this should go.

Craig