Update your OpenSSL now!

@toml, than you very much. I thought it was a DirectAdmin service because of the naming, i did this to restart all the Dell services:

Code:
srvadmin-services.sh stop
srvadmin-services.sh start

Now I only have one server that still show result when checking for services using old OpenSSL, it is a DirectAdmin pid, and I can see the pid in the "Service monitor", however I am unable to kill it and/or remove it:

First I check my server:
Code:
[root@server~]# lsof -n | grep ssl | grep DEL
directadm  7660      root  DEL       REG                8,2               2705316 /usr/lib64/libssl.so.1.0.1e

Then I kill the pid:
Code:
[root@server~]# kill 7660

Then I check my server again, but the pid is still there, and is still using old OpenSSL:
Code:
[root@server~]# lsof -n | grep ssl | grep DEL
directadm  7660      root  DEL       REG                8,2               2705316 /usr/lib64/libssl.so.1.0.1e

Also it does not help to stop/restart DirectAdmin. What can I do to get red of the above pid/problem?
 
@toml, thanks a lot! kill -9 fixed it, and finally none of my servers is using the old openssl version anymore:

Code:
[root@server ~]# lsof -n | grep ssl | grep DEL
directadm  7660      root  DEL       REG                8,2               2705316 /usr/lib64/libssl.so.1.0.1e
[root@server~]# kill -9 7660
[root@server~]# lsof -n | grep ssl | grep DEL
[root@server~]#

I guess the next thing is to spend the night buying/renew all my servers ssl certificates. Not happy about that, because I recently renewed them all by two more years. But it seems to be needed, so I will do it. :)
 
Hi,

I'm just a noob so it took me a lot of reading. I accidentally, upon finding out about this openssl bug, updated openssh. I didn't really pay close attention but it did stuff. :-)

After finding out that I had done the wrong update I got confused since I knew the update was out and yet I was not getting any openssl update.

Anyhow, I finally got enough reading down to feel confident I've done all that I needed on the openssl bug for now.

I just wanted relay some info that helped me to get a grasp of the situation. Due to the standard RH/Centos procedures, and me looking for openssl "g" revision that is widely discussed as the fix, it was kind of a head scratcher for a bit..

RE CentOS 6.5: https://www.centos.org/forums/viewtopic.php?f=9&t=45814#p194518
rpm -q openssl` and it reports version 1.0.1e and less than 1.0.1e-16.el6_5.4.0.1 then you are currently vulnerable to this problem. If it reports 1.0.1e-16.el6_5.4.0.1.centos then you have the temporary version issued before Redhat issued their official fix. If you have 1.0.1e-16.el6_5.7 or higher then you have the official fixed version."

My conclusion about OpenSSH update being responsible came from seeing:

# rpm -q openssl
openssl-1.0.1e-16.el6_5.7.x86_64

..on my box. Thank goodness! Now I can close some of the multitude of browser tabs and windows and quit shuffling between them as I open more windows.. lol
 
just NOW did yum update and it offered version e so I am not sure whats going on.

edit: I missed that rh/cent would backport the patch onto version e so I may e ok
 
Last edited:
two issues:

when I do a yum update then restart the httpd service openssl is updated to OpenSSL 1.0.1e-fips, not 1.0.1g (I thought all version < 1.0.1g are vulnerable??)

the second issue is even openssl is updated to 1.0.1, under curl the ssl version is still openssl 1.0.0??
 
Last edited:
@redjersey

try this:

rpm -qa openssl\*

if you see this:

openssl-devel-1.0.1e-16.el6_5.7.x86_64
openssl-1.0.1e-16.el6_5.7.x86_64

.. then you have the latest redhat patch -- note the key part is the _5.7

Then just [strike]restart[/strike] kill|stop -> start openssl and all ssl dependent services or reboot
 
Last edited:
when I run:

# rpm -qa openssl\*
openssl-devel-1.0.1e-16.el6_5.7.x86_64
openssl-1.0.1e-16.el6_5.7.x86_64

I do get the exact same lines, I also did a reboot, but my phpinfo is still showing curl ssl is running openssl 1.0.0 and openssl-1.0.1e

so is it mean that the latest patch are added to both 1.0.0 and 1.0.1e?
 
The SSL is good from what I read... However I am unqualified to instruct you myself on the httpd. I did read tons about this bug and recall many getting advice that they needed to recompile Apache. I guess there is something (mod_ssl?) of the installed openssl which Apache relies on when when it's installed. If that makes any sense.

I think much of the people who could answer you more authoritatively are working on their servers in light of this bug.

Let me see if I can dig up some of the links I have on this and will edit them in here in a minute.

Maybe this: "Did you just restart apache or did you kill it completely and restart it? A graceful restart doesn't kill it completely" (duh, guess that's not related since you said you rebooted..)

RE: http://www.webhostingtalk.com/showthread.php?t=1364373 post #21

..sorry I don't seem to have much more than the one link to WHT ~post 12 or so someone was commenting about recompile for apache but nobody in that conversation indicates recompile is required.
 
Last edited:
I am from Vietnam where I have not updated with YUM yet.
Follow midas's guide. my server is safe now.
thanks
 
saosangmo,

If I understand you correctly, affected redhat/centos and the openssl-1.0.1g.tar.gz AFAIK isn't the best approach.

Official centos/redhat updates patch the 1.0.1e library instead of updating the 1.0.1e library to 1.0.1g.

I guess it will break the ability for YUM to update your openssl and anything that uses openssl, at least..
 
I had to rebuild php because header file and library version not matched.
Just for your information.
 
FreeBSD users should be unaffected as they use OpenSSL 0.9.8y built-in.

Just in case, if you eventually compiled OpenSSL from ports (usually people don't do that), check your ports tree for it and see if it's up to date..
 
The SSL is good from what I read... However I am unqualified to instruct you myself on the httpd. I did read tons about this bug and recall many getting advice that they needed to recompile Apache. I guess there is something (mod_ssl?) of the installed openssl which Apache relies on when when it's installed. If that makes any sense.

I think much of the people who could answer you more authoritatively are working on their servers in light of this bug.

Let me see if I can dig up some of the links I have on this and will edit them in here in a minute.

Maybe this: "Did you just restart apache or did you kill it completely and restart it? A graceful restart doesn't kill it completely" (duh, guess that's not related since you said you rebooted..)

RE: http://www.webhostingtalk.com/showthread.php?t=1364373 post #21

..sorry I don't seem to have much more than the one link to WHT ~post 12 or so someone was commenting about recompile for apache but nobody in that conversation indicates recompile is required.

yes I did, I did restarted the httpd, I have also rebooted my server just in case.
but under curl the ssl version is still showing openssl 1.0.0

I have tried http://filippo.io/Heartbleed with mydomain.com:443 which says my server seems ok, but I just don't know how to fix that curl issue.

#curl -V
curl 7.36.0 (x86_64-unknown-linux-gnu) libcurl/7.36.0 OpenSSL/1.0.0 zlib/1.2.3 libidn/1.18
Protocols: dict file ftp ftps gopher http https imap imaps pop3 pop3s rtsp smtp smtps telnet tftp
Features: IDN IPv6 Largefile NTLM NTLM_WB SSL libz

Update: not sure why but I had to build the curl again (already did it last nite, not working)
./build curl
./build php d

now it's showing 1.0.1e
 
Last edited:
Hi.
after distribution update!

If i check the SSL Connections on Port 443 on Debian 7 and centos 6,5 (Both 64 bit) everything is OK.
If i check the port 2222 on both boxes
the centos box say everthing is ok, but the debian 7 is vulnerable
 
i have try to reinstall DA with the latest prerelease but same message on a debian 7 64bit OS when i check the port 2222
Server is vulnerable, please upgrade software ASAP
 
Back
Top