I'm getting the below error message when trying to perform "configtest" after upgrading apache to 2.4.12 version with success.
/home/apache/bin/httpd: symbol lookup error: /home/apache/bin/httpd: undefined symbol: SSL_CONF_CTX_new
Note: I received the error after recently upgrading my openssl to 1.0.2.
Just wonder how to get rid of the error message.
Running on Ubuntu Server 12.04.05 LTS 32bit.
ffmpeg: symbol lookup error: /usr/lib/libavcodec.so.52: undefined symbol: speex_header_free
I am using a centos 5 vps
I'm trying to use mysqlhotcopy on freebsd 6.1 with mysql 5.0.45, but I'm getting this error when executing it:
/libexec/ld-elf.so.1: /usr/local/lib/perl5/site_perl/5.8.8/mach/auto/DBD/mysql/mysql.so: Undefined symbol "mysql_server_init"
at first, I was having problems compiling DBD-mysql, because mysql.h and mysqld_error.h wasn't on the include dir, so I had to add -I/usr/local/mysql-5.0.45-freebsd6.0-i386/include to cflags (I also tried -I/usr/local/include/mysql , but same happened)
since the error persisted, I changed all references to mysql_server_init (deprecated) to mysql_library_init (same for mysql_server_end -> mysql_library_end)
and the error still persists!
also:
# ldd /usr/local/lib/perl5/site_perl/5.8.8/mach/auto/DBD/mysql/mysql.so
libz.so => /usr/lib/libz.so (0x28178000)
libcrypt.so.3 => /lib/libcrypt.so.3 (0x28189000)
libm.so.4 => /lib/libm.so.4 (0x281a1000)
We're having problems with the Roundcube webmail spell check, and upon checking the log we get the error 'Unable to find the socket transport SSL'. From what I can find out this is normally due to Open SSL not being enabled, but it is, you can check our php config here.
View 3 Replies View RelatedDo you have plans for building Apache 2.4 with OpenSSL 1.0.2? One good reason for upgrading Apache to OpenSSL to 1.0.2 would be the ability to disable TLS session tickets, eq. when using PFS:
Code : SSLOpenSSLConfCmd Options -SessionTicket
Here are a few references:
[URL] ....
[URL] ....
I tried to experiment with replacing OpenSSL dlls and exe in apachein directory but that did not work, because it looks like SSLOpenSSLConfCmd configuration directive is only available when Apache is compiled against OpenSSL 1.0.2.
Currently I am using Apache 2.4.9 OpenSSL 1.0.1g (VC10), want to migrate openSSL 1.0.1h to resolve the vulnerability issue.is there way to migrate openssl alone in my existing apache build ?
View 9 Replies View Relatedwe need to upgrade the openssl modules, we have apache 2.2 installed and openssl version is 1.0.0,apache and openssl are in solaris9 SPARC.
View 4 Replies View RelatedLast week I have updated the apache from 2.4.6 to 2.4.9 version in Win 2008, 64-bit server. There was no openSSL and update was successful. Later I did the update in QA with openSSL and again the updte was successfully completed. Apache services was running fine and everything looked nice.
When I did the same update in the prd where openSSL is also there, it failed to start the service.
steps to update the apache from 2.4.6 to 2.4.9
----------------------------------------------
1> stop the apache services
2> Take the backup by copying original Apache installation directory and rename it . (eg I:Program Files (x86)Apache Software FoundationApache2.2 to Apache2.2_old)
3> Unzip the latest binaries to the temp directory
4> Copy the following files apachebin , apachemodules to the Apache Inst Directory ( I:Program Files (x86)Apache Software FoundationApache2.2)
5> start the apache service
----------------------------------------------------
QA and PRD both has enabled openSSL but it was prd where we got the issue , and the apache services couldnt be started. We have had to revert the change. Find the error log in the apache directory
--------------------------
[Tue Jun 24 21:12:12.665632 2014] [ssl:emerg] [pid 3336:tid 320] AH02561: Failed to configure certificate
RGWEB58V.brotherdc.eu:443:0, check G:/Program Files (x86)/Apache Software Foundation/Apache2.2/conf/server.crt
[Tue Jun 24 21:12:12.665632 2014] [ssl:emerg] [pid 3336:tid 320] SSL Library Error: error:0906D06C:PEM
routines:PEM_read_bio:no start line (Expecting: CERTIFICATE) -- Bad file contents or format - or even just a
forgotten SSLCertificateKeyFile? [Tue Jun 24 21:12:12.665632 2014] [ssl:emerg] [pid 3336:tid 320] SSL Library Error: error:140AD009:SSL
routines:SSL_CTX_use_certificate_file:PEM lib
---------------------------------------
I read somewhere that there is bug in 2.4.9 as this version breaks the openSSL.
Also read on this forum that someone resolved the issu by changing the server certificate from DER to PEM.
c:Apache24bin>openssl.exe
WARNING: can't open config file: c:/openssl-1.0.1e-X64/ssl/openssl.cnf
OpenSSL>
I regularly have DNS issues with some of my sites hosted with a specific host... At random sites, the urls done resolve and I get a DNS error.
Just to find out that a few hours later it loads again without issues...
I always thought it had something to do with my ISP... My host says that there's nothing wrong with their DNS.
However, I asked a few people now and it seems they're having DNS problems too sometimes when accessing the problematic urls...
I am running on a VPS system, and I have some auction software and I run Vbulletin as well..
When someone trys to do a Private MEssage via Vbulletin using anything with the @ in it.. Like an e-mail address. And hit submit, it says " /private.php access is denied.
But it also does it when I post news in the auction site. Which leads me to believe their is something screwed up with one of the server settings.. But I dont know what?
I triad to make restart for my Apache I have this result:
PHP Code:
[root@server ~]# httpd restarthttpd: Syntax error on line 39 of /usr/local/apache/conf/httpd.conf:
Syntax error on line 9 of /usr/local/apache/conf/php.conf: Cannot load /usr/local/apache/modules/libphp5.so into server:
/usr/local/apache/modules/libphp5.so: undefined symbol: xmlTextReaderSchemaValidate
I simply want to rewrite a hash # or other symbol to a question mark via htaccess. So any hash in the URL anywhere should internally rewrite to a ?. All I want is to remove ? from the URL really.
View 2 Replies View RelatedRHEL4 Box
No Control Panel
Current Version of OpenSSL = OpenSSL 0.9.7a Feb 19 2003
which is the most current version available via 'up2date'.
Need to update to 0.9.8.
I have download the source for 0.9.8h and have:
#./config --prefix=/usr/local --openssldir=/usr/local/openssl
#make
#make test
#make install
All commands ran without errors.
When I run:
# openssl version
OpenSSL 0.9.7a Feb 19 2003
# whereis openssl
openssl: /usr/bin/openssl /usr/local/bin/openssl /usr/include/openssl /usr/local/openssl /usr/share/man/man1/openssl.1ssl.gz
# rpm -qa | grep openssl
openssl-devel-0.9.7a-43.17.el4_6.1
openssl-0.9.7a-43.17.el4_6.1
xmlsec1-openssl-1.2.6-3
openssl096b-0.9.6b-22.46
How can I get the version updated?
what is the best way to upgrade Openssl
to the latest version
openssl-0.9.8g
in my cpanel server
openssl version
OpenSSL 0.9.7a Feb 19 2003
I might still be a bit too close to the newbie level to be trying this
but I wanted upgrade various components of my cpanel install for security issues.
So I got openssl to update to 0.9.8g and that seemed to be working correctly.
Quote:
openssl version
OpenSSL 0.9.8g 19 Oct 2007
So now I am trying to build PHP 4.4.8 like this:
Quote:
./configure --with-litespeed --with-config-file-path=../php --with-mysql=/usr --with-zlib --with-zlib-dir=.. --with-gd --with-jpeg-dir=.. --with-png-dir=.. --enable-shmop --enable-track-vars --enable-sockets --enable-sysvsem --enable-sysvshm --enable-magic-quotes --with-openssl
And it fails because of the openssl:
Quote:
Make
...
In function `zif_openssl_seal':
/php-4.4.8/ext/openssl/openssl.c:2885: undefined reference to `EVP_CIPHER_CTX_block_size'
collect2: ld returned 1 exit status
(I tried --with-openssl=/usr with no change)
so I got OpenSSL 0.9.8k up and installed, no issued:
Quote:
# openssl
OpenSSL> version
OpenSSL 0.9.8k 25 Mar 2009
OpenSSL>
Rebuilt cURL (and then php), httpd and proftpd but all of them are still linking to the older libraries for some reason
Quote:
# curl -V
curl 7.19.4 (x86_64-unknown-linux-gnu) libcurl/7.19.4 OpenSSL/0.9.8g zlib/1.2.3
Protocols: tftp ftp telnet dict ldap http https ftps
Features: IPv6 Largefile NTLM SSL libz
Quote:
[Tue Apr 14 00:11:03 2009] [notice] Apache/2.2.11 (Unix) mod_ssl/2.2.11 OpenSSL/0.9.8g DAV/2 PHP/5.2.9 Apache configured -- resuming normal operations
Quote:
Starting proftpd: - mod_tls/2.2.1: compiled using OpenSSL version 'OpenSSL 0.9.8i 15 Sep 2008' headers, but linked to OpenSSL version 'OpenSSL 0.9.8g 19 Oct 2007' library
proftpd has its own set of issues obviously built with i headers and linked to g headers. Any ideas wtf I did? I recomplined and restarted everything. I removed the g and i libraries completely. OpenSSH seems happy and nothing is actually "wrong", the server is working fine but I'm really anal retentive this way...it's kinda how I feel "safer" at the OS level.
I'm trying to connect to the server via SFTP but when I check the box to do that in Total Commanders settings box it says I need the OpenSSL dll's. I downloaded the must current file from [url] but I can't find any dll files in it and I'm unsure how to proceed.
View 1 Replies View Related- Apache version : 2.2.21
- operating system : MAC OS Lion Server
- mod_ssl : 2.2.21
I have a certain area on a homepage which is protected by an .htaccess file. The code of the .htaccess file is the following:
Code:
AuthType Digest
AuthName "Password Required"
Options +Indexes
# code for HTTPS
RewriteEngine On
RewriteCond
%{SERVER_PORT} !^443$
RewriteRule (.*) https://%{SERVER_NAME}%{REQUEST_URI}
Require valid-user
Usually everything has been working fine with this .htaccess-file until today. There hasn't been any change on the system.
If you click on the hyperlink to the restricted area the login window appears where you need to type in your username and a password. After typing in these data the following message appears in the browser:
500 Internal Server Error
The server encountered an internal error or
misconfigruation and was unable to complete your request. (…)
The apache error log file contains the following message:
Request exceeded the limit of 10 sub request nesting levels due to probable configuration error.
Use 'LimitInternalRecursion' to increase the limit if necessary.
Use 'LogLevel debug' to get a backtrace.
The increase of the LimitInternalRecursion does not solve the problem. So I switched the LogLevel in the httpd.conf file from warn to debug and looked again on the entries of the error log.
The following appeared (I have replaced ip addresses, usernames and URL by xyz or <URL>) :
Code:
[error] [client xyz]
mod_digest_apple: Client is using a nonce not issued by this server for this context:/rb/Vorlesung/WS12/Downloads/, referer:
http:<URL>
[Code] ....
if I do a reverse lookup on my server's IP it returns a host assigned by the data center. So is this something I should even be managing too or is that always left up to the DC? I run my own DNS for the forward lookup zones.
Also just a technical question, when you do a reverse lookup, how does it know where to look to get the host? I'm not sure I fully understand how these work.
Somebody purchased a domain from a popular registrar. When you lookup the domain, it points to its own name servers (i.e. ns1.registrar.com and ns2.registrar.com), which means that it is hosted by them as well. correct?
this person is saying they bought their domain from the registrar, but they are hosted by company X. he says that he is pointing the ip address of his web site on company X to his registrar.
I want to know how serious following nameserver problem can be in reality:
[url] ?
From DNS timing someone could conclude it's a slow hosting.
Can anyone explain above results "accurately" ? I still don't know what's "good" result in DNS timing: A or F?
The problem is that
[url] loads slow (5-10 seconds on T1 connection), here's a speed test. I'm aware images and scripts contribute to slow site loading.
My question is: Beside removing some images from the Home page, is it any reason to change the hosting (regarding DNS tests).
Friend of mine asked me what "CURL named lookup issue resolved" means, he's got a CentOS 4 machine, with cPanel/WHM latest builds I tried recompiling apache and even cPanel at some point but no good..
View 1 Replies View RelatedI am having a problem. My server was able to ping google.com successfully before installing DirectAdmin.
However after installing DirectAdmin, I am no longer able to ping google.com anymore. I just get a host cannot be found error. This is the same with all other domains.
The /etc/resolv.conf is exactly the same, so I do not know the issue. I also have other servers using the same nameserver and are still able to ping and resolve successfully.
The main problem here is that exim no longer can send email as it cannot lookup the hostnames.
the error from exim is: R=lookuphost defer (-1): host lookup did not complete.
Thats the only error within the message.
My VPS Config:
centos with webmin
mail program sendmail
Its a newly configured VPS have a problem with smtp realying through outlook and other smtp programs. When i use smtp on outlook it give following error:
The message could not be sent because one of the recipients was rejected by the server. The rejected e-mail address was 'xxxxx@yahoo.com'. Subject 'Outlook', Account: 'Testing', Server: 'xxxx.com', Protocol: SMTP, Server Response: '550 5.7.1 <xxxxx@yahoo.com>... Relaying denied. IP name possibly forged [xx.xxx.xx.xxx]', Port: 25, Secure(SSL): No, Server Error: 550, Error Number: 0x800CCC79
I want to make outgoing authentication activate instead of allowing some iprange or domain for relaying through /etc/mail/access.
I have six dedicated servers on a VPR. I use one of these as the Database server and the other five connects to that server to read the Database.
When I setup my scripts for the first time I used the external IP to connect to the DB server. Everything worked perfect for these weeks:
mysql_connect('174.xxx.xxx.xx', 'user', 'pass');
Since three days ago I finally received the internal IPs of my servers and now I use the internal IP of the DB server:
mysql_connect('10.0.0.1:3306', 'user', 'pass');
When I did this last change this my database started to collapse. Queries stuck on a bottleneck every 10 minutes. Here's an example:
40468 simon 10.0.0.7:47002 pkindigo_db Query 115 Locked update `captured` set hp_left = '79', status = 'normal', att1pp = '25', att2pp = '30', att3p
40523 simon 10.0.0.2:58080 pkindigo_db Query 115 Locked update `captured` set experience = '5120', level = '16', hp_left = '58', hp = '58', status =
... (tons of locked queries before) ...
As you may see table is locked since 115 seconds for a simple update... after some hours this ends on a "#1135 - Can't create a new thread (errno 13); if you are not out of available memory, you can consult the manual for a possible OS-dependent bug" and even some of my tables finally crashing.
I think that my Database stuck because of a DNS lookup delay. When I had this problem before using the external IPs, I added skip-name-resolve and the problem was fixed. But now with the internal IPs it seems that's not enough.
I'm having tons of problems trying to get reverse lookup working on my dns server. I've scoured the net and tried lots of methods but so far none work.
Forward lookup DNS is working perfect, but reverse lookup is completely broken.
db.44.80.212.67:
Code:
$TTL 86400 ; Default TTL in secs(1 day)
@ IN SOA ns1.capturetheprize.com. info.capturetheprize.com. (
2009072902; serial number YYMMDDNN
28800 ; Refresh
7200 ; Retry
864000 ; Expire
86400 ; Min TTL
)
IN NS ns1.capturetheprize.com.
IN NS ns1.mytempmail.com.
; Reverse lookups
44 IN PTR ns2.mytempmail.com.
44 IN PTR ns1.mytempmail.com.
44 IN PTR ns2.capturetheprize.com.
44 IN PTR ns1.capturetheprize.com.
44 IN PTR mail.mytempmail.com.
44 IN PTR mail.capturetheprize.com.
44 IN PTR stats.capturetheprize.com.
44 IN PTR secure.capturetheprize.com.
named.conf
Code:
zone "80.212.67.in-addr.arpa" IN {
type master;
file "etcdb.44.80.212.67";
allow-transfer { localhost; };
};
if its possible to change the reverse lookup, or at least i think thats what it is called.
I have 2 accounts on my CentOS, WHM/cPanel VPS and they are hosted on their own seperate IPs however when you do a lookup on one domain it also lists the other domain as being hosted on the server and vice-versa.
How do i change this so they appear like they are on different servers? Each one has their own nameservers which are on seperate IPs also.
My MySQL server slowed down to a crawl a few days ago, and I finally fixed it by starting it without DNS lookups via "--skip-name-resolve".
I tested domain name resolution on the MySQL machine and it was able to return IPs from domains rather quickly, so I'm at a loss as to why DNS was the problem here? I am using "192.168.1.6" as the MySQL server's IP, in case that matters?
I have a couple of questions:Why is DNS resolution so slow when in the command line it's running normally?
Why does MySQL need to do DNS lookups at all? I don't see any log files that's converting IPs to host names or vice versa.
Where am I sending e-mails from with the below settings:
(Is it mail.freesexdoor.com, www.freesexdoor.com, 64.85.161.123 or 64.85.161.124?)
My secondary question is: Is creating own reverse IP lookup to 64.85.161.124 possible and beneficial in my case? If so, what should I replace this SPF line with: freesexdoor.com. TXT "v=spf1 a mx ip4:64.85.161.123 ?all"?
I'm trying to improve freesexdoor.com's e-mail delivery ratio because hotmail and gmail deliver my PHP mailer's sent confirmation e-mails to the junk folder. I've done everything I can to improve the sent headers etc. so there's only one more possible way to improve it: (That I can think of). Freesexdoor.com is hosted on the same server as mitorahgames.com.
DirectAdmin control panel on CentOS 4.4.
My server's primary IP is: 64.85.161.123
My server's hostname is: server.mitorahgames.com
server.mitorahgames.com's Related DNS Settings:
mailA64.85.161.123
server.mitorahgames.com.A64.85.161.123
server.mitorahgames.com.NSns1.mitorahgames.com.
server.mitorahgames.com.NSns2.mitorahgames.com.
mailMX10
123.161.85.64.in-addr.arpaPTRserver.mitorahgames.com.
server.mitorahgames.com.TXT"v=spf1 a mx ip4:64.85.161.123 ?all"
freesexdoor.com's IP is: 64.85.161.124
freesexdoor.com's Related DNS Settings:
freesexdoor.com.A64.85.161.124
mailA64.85.161.124
mailMX10
freesexdoor.com.TXT"v=spf1 a mx ip4:64.85.161.123 ?all"
I'm by the way putting www.freesexdoor.com (PHP reported server name) to the Message ID: field in the e-mail headers. Would putting IP there perhaps work better? If so, which IP?
I want to hide my NS records on WHOIS lookup, or show a fake or something so that no one can find me, Is it possible,
View 5 Replies View Related