URGENT: Apache Fails To Start After Update

httpd won’t start now that 4.03 was installed. “apachectl -t” shows no errors in config (which hasnt changed in weeks anyways)

virtualmin was upgraded (from 4.02) via yum command line, not via webmin.

“Ran post-install script” was executed afterwards and logged to webmin actions log.

please advise.

the upgrade package was wbm-virtual-server-4.03.gpl-1.noarch. i was able to yum undo to downgrade to 4.02, but apache still doesnt start.

the upgrade occurred at :12 and apache’s first error was at :13. this was apache’s first restart today, since logrotate.

the only upgrade besides virtualmin was perl-File-Next, does webmin/virtualmin even use this package?


Hmm, I wasn’t able to reproduce a problem performing an upgrade on any of my test systems.

What error do you get when trying to start Apache?

You can do that by running this command:

/etc/init.d/httpd restart

I’d be curious what errors you get on the command line, as well as to see what shows up in the Apache error log in /var/log/httpd/error_log.


Do you see anything in the error logs for the “atamaeurope” domain?

Certain Apache errors can be directed into the error logs of specific Virtual Hosts, especially SSL related problems. It looks like that domain has SSL certificates being loaded towards the end of that output.

If that doesn’t help, you may want to look at /var/log/virtualmin/, and see if any of those log files have anything being written to them as you try to start Apache.


that was it Eric, thank you. I created a new CSR this morning and I was waiting for the cert to be issued. I believe I have done this in the past and had no problems. I should NOT have any problems starting apache just because a new CSR has been issued, correct? What happened?


I get NO ERRORS in /var/log/httpd/error_log

/etc/init.d/httpd restart returns
Stopping httpd: [FAILED]
Starting httpd: [FAILED]

I’m attaching the STRACE output.

since loglevel is set to debug, this is what I get in the error_log

[Mon Oct 21 22:31:48 2013] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Mon Oct 21 22:31:48 2013] [info] mod_unique_id: using ip addr xxx.xxx.xxx.xxx
[Mon Oct 21 22:31:49 2013] [info] Init: Seeding PRNG with 256 bytes of entropy
[Mon Oct 21 22:31:49 2013] [info] Init: Generating temporary RSA private keys (512/1024 bits)
[Mon Oct 21 22:31:49 2013] [info] Init: Generating temporary DH parameters (512/1024 bits)
[Mon Oct 21 22:31:49 2013] [info] Init: Initializing (virtual) servers for SSL

Well, it’s difficult to say exactly what the issue is… however, you’re correct that simply generating a new CSR should not cause a problem.

When generating a new CSR, that key isn’t actually active until pasting in the new SSL certificate.


that’s all I did. created a CSR. i will install the new one, back it up, test by generating a new one and report back.

the new key isnt active but that shouldnt stop the old one from working, right?

thank you!

there’s something going on with webmin. i went to upload the key (the cert just got issued) and virtualmin got stuck one the first step (applying cert or something similar). then it proceeded to use a lot of CPU and memory (see http://i.imgur.com/MXGWM0s.png).

i cannot get this cert installed, and problems started when i requested a new CSR, remember? what’s going on? could I get someone to take a look at it?!

in other words, right.cgi?open=system&auto=status&open=updates&open=common&open=owner&open=reseller&open=vm2limits gets stuck in “Configuring webserver to use new certificate and key …” and webmin starts using memory and cpu thru the roof.

I was able to install the cert, but only manually. installing via virtualmin sends miniserv on some loop.