Search Results

Search found 4244 results on 170 pages for 'ssl certificate'.

Page 69/170 | < Previous Page | 65 66 67 68 69 70 71 72 73 74 75 76  | Next Page >

  • Provisioning problem with certificate and APNs

    - by Florent
    Hi all ! I hava a big problem, here is the deal : I've made an app which has an id, on an another machine i've submit the app to the app store, now i have another mac, and i have made the AppIds using my old mac (so i used an old cer to make the push certificates), next made my apns cert for my server and i have : unable to connect to ssl://gateway.... I do not know what to do because i cannot create another appId using the same id and can't remove an app Id. Thanks to all

    Read the article

  • X.509 certificate based authentication with OpenSSL (without using sockets)

    - by hartem
    Hi, Is there an alternative in OpenSSL to SSL_set_connect_state()/SSL_set_accept_state() for X.509 certificate based authentication? The problem is that in my application the client and server do not communicate using sockets, and the establishment of direct connection between them is not possible. So what I want from OpenSSL is to 'expose' the intermediate SSL context establishment messages which I would then convey to the party at the other end. Thanks for your help!

    Read the article

  • Post bacc CS certificate of Java Certification?

    - by gwozdz
    I have a BS in biology, MS in Environmental Science, but I've been writing software in C++ and Java throughout grad school and a bit in my current job. I'd love to transition to a programming career. I've thought about getting what's called a "Post baccalaureate Certificate" or Java certification in lieu of another degree in CS. Which is more useful in terms of getting a job?

    Read the article

  • SSL in tomcat with apr and Centos 6

    - by Jonathan
    I'm facing a problem setting up my tomcat with apr native lib, I have the following: Tomcat: 7.0.42 Java: 1.7.0_40-b43 OS: Centos 6.4 (2.6.32-358.18.1.el6.i686) APR: 1.3.9 Native lib: 1.1.27 OpenSSL: openssl-1.0.0-27.el6_4.2.i686 My server.xml looks like: ... <Listener className="org.apache.catalina.core.AprLifecycleListener" SSLEngine="on" /> ... <Connector port="8443" protocol="HTTP/1.1" SSLEnabled="true" maxThreads="150" scheme="https" secure="true" clientAuth="false" sslProtocol="TLS" SSLCertificateFile="/tmp/monitoringPortalCert.pem" SSLCertificateKeyFile="/tmp/monitoringPortalKey.pem" SSLPassword="hide" /> ... I compiled the native lib as follow: ./configure --with-apr=/usr/bin/apr-1-config --with-ssl=yes --prefix=$CATALINA_HOME make && make install The APR is loaded ok: Oct 06, 2013 7:55:14 PM org.apache.catalina.core.AprLifecycleListener init INFO: Loaded APR based Apache Tomcat Native library 1.1.27 using APR version 1.3.9. But I'm still having this error: SEVERE: Failed to initialize the SSLEngine. org.apache.tomcat.jni.Error: 70023: This function has not been implemented on this platform ./configure outcome [root@localhost native]# ./configure --with-apr=/usr/bin/apr-1-config --with-ssl=yes -- prefix=$CATALINA_HOME && make && make install checking build system type... i686-pc-linux-gnu checking host system type... i686-pc-linux-gnu checking target system type... i686-pc-linux-gnu checking for a BSD-compatible install... /usr/bin/install -c checking for working mkdir -p... yes Tomcat Native Version: 1.1.27 checking for chosen layout... tcnative checking for APR... yes setting CC to "gcc" setting CPP to "gcc -E" checking for JDK location (please wait)... /usr/java/jdk1.7.0_40 from environment checking Java platform... checking Java platform... checking for sablevm... NONE adding "-I/usr/java/jdk1.7.0_40/include" to TCNATIVE_PRIV_INCLUDES checking os_type directory... linux adding "-I/usr/java/jdk1.7.0_40/include/linux" to TCNATIVE_PRIV_INCLUDES checking for gcc... gcc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ISO C89... none needed checking for OpenSSL library... using openssl from /usr/lib and /usr/include checking OpenSSL library version... ok checking for OpenSSL DSA support... yes setting TCNATIVE_LDFLAGS to "-lssl -lcrypto" adding "-DHAVE_OPENSSL" to CFLAGS setting TCNATIVE_LIBS to "" setting TCNATIVE_LIBS to " /usr/lib/libapr-1.la -lpthread" configure: creating ./config.status config.status: creating tcnative.pc config.status: creating Makefile config.status: executing default commands make[1]: Entering directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Nothing to be done for `local-all'. make[1]: Leaving directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Entering directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Nothing to be done for `local-all'. make[1]: Leaving directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' /usr/lib/apr-1/build/mkdir.sh /usr/apache-tomcat-7.0.42/include/apr-1 /usr/apache- tomcat-7.0.42/lib/pkgconfig \ /usr/apache-tomcat-7.0.42/lib /usr/apache-tomcat-7.0.42/bin /usr/bin/install -c -m 644 tcnative.pc /usr/apache-tomcat-7.0.42/lib/pkgconfig/tcnative- 1.pc list=''; for i in $list; do \ ( cd $i ; make DESTDIR= install ); \ done /bin/sh /usr/lib/apr-1/build/libtool --mode=install /usr/bin/install -c -m 755 libtcnative-1.la /usr/apache-tomcat-7.0.42/lib libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.so.0.1.27 /usr/apache- tomcat-7.0.42/lib/libtcnative-1.so.0.1.27 libtool: install: (cd /usr/apache-tomcat-7.0.42/lib && { ln -s -f libtcnative- 1.so.0.1.27 libtcnative-1.so.0 || { rm -f libtcnative-1.so.0 && ln -s libtcnative- 1.so.0.1.27 libtcnative-1.so.0; }; }) libtool: install: (cd /usr/apache-tomcat-7.0.42/lib && { ln -s -f libtcnative- 1.so.0.1.27 libtcnative-1.so || { rm -f libtcnative-1.so && ln -s libtcnative-1.so.0.1.27 libtcnative-1.so; }; }) libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.lai /usr/apache-tomcat- 7.0.42/lib/libtcnative-1.la libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.a /usr/apache-tomcat- 7.0.42/lib/libtcnative-1.a libtool: install: chmod 644 /usr/apache-tomcat-7.0.42/lib/libtcnative-1.a libtool: install: ranlib /usr/apache-tomcat-7.0.42/lib/libtcnative-1.a libtool: install: warning: remember to run `libtool --finish /usr/local/apr/lib' make && make install outcome: make[1]: Entering directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Nothing to be done for `local-all'. make[1]: Leaving directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Entering directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' make[1]: Nothing to be done for `local-all'. make[1]: Leaving directory `/usr/apache-tomcat-7.0.42/bin/tomcat-native-1.1.27- src/jni/native' /usr/lib/apr-1/build/mkdir.sh /usr/apache-tomcat-7.0.42/include/apr-1 /usr/apache- tomcat-7.0.42/lib/pkgconfig \ /usr/apache-tomcat-7.0.42/lib /usr/apache-tomcat-7.0.42/bin /usr/bin/install -c -m 644 tcnative.pc /usr/apache-tomcat-7.0.42/lib/pkgconfig/tcnative- 1.pc list=''; for i in $list; do \ ( cd $i ; make DESTDIR= install ); \ done /bin/sh /usr/lib/apr-1/build/libtool --mode=install /usr/bin/install -c -m 755 libtcnative-1.la /usr/apache-tomcat-7.0.42/lib libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.so.0.1.27 /usr/apache- tomcat-7.0.42/lib/libtcnative-1.so.0.1.27 libtool: install: (cd /usr/apache-tomcat-7.0.42/lib && { ln -s -f libtcnative- 1.so.0.1.27 libtcnative-1.so.0 || { rm -f libtcnative-1.so.0 && ln -s libtcnative- 1.so.0.1.27 libtcnative-1.so.0; }; }) libtool: install: (cd /usr/apache-tomcat-7.0.42/lib && { ln -s -f libtcnative- 1.so.0.1.27 libtcnative-1.so || { rm -f libtcnative-1.so && ln -s libtcnative-1.so.0.1.27 libtcnative-1.so; }; }) libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.lai /usr/apache-tomcat- 7.0.42/lib/libtcnative-1.la libtool: install: /usr/bin/install -c -m 755 .libs/libtcnative-1.a /usr/apache-tomcat- 7.0.42/lib/libtcnative-1.a libtool: install: chmod 644 /usr/apache-tomcat-7.0.42/lib/libtcnative-1.a libtool: install: ranlib /usr/apache-tomcat-7.0.42/lib/libtcnative-1.a libtool: install: warning: remember to run `libtool --finish /usr/local/apr/lib' It seems everything is fine, but the error is not self-explanatory Could you guys help to understand where my error is? What am I missing? Thanks in advance for your support.

    Read the article

  • "Can't Connect to Server" from 2nd virtual host on VPS

    - by chaoskreator
    I'm using Debian 7 Wheezy and Apache 2.2.22, and I'm setting up Virtual Hosts for a number of websites on my VPS. I've successfully configured the VirtualHost directives for one of the sites, but the second one continually gives "Problem Loading Page" in Firefox. I've run configtest and it has verified all my syntax is correct, and I've checked all the permissions. Everything on the 2nd domain is pretty much copy/pasted from the first, so I'm not sure what the issue is, as there are no entries into /var/log/apache2/error.log other than where I have reloaded the configurations: /# cat /var/log/apache2/error.log [Thu May 29 01:19:00 2014] [notice] Graceful restart requested, doing restart [Thu May 29 01:19:00 2014] [info] Init: Seeding PRNG with 656 bytes of entropy [Thu May 29 01:19:00 2014] [info] Init: Generating temporary RSA private keys (512/1024 bits) [Thu May 29 01:19:00 2014] [info] Init: Generating temporary DH parameters (512/1024 bits) [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(253): shmcb_init allocated 512000 bytes of shared memory [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(272): for 511920 bytes (512000 including header), recommending 32 subcaches, 133 indexes each [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(306): shmcb_init_memory choices follow [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(308): subcache_num = 32 [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(310): subcache_size = 15992 [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(312): subcache_data_offset = 3208 [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(314): subcache_data_size = 12784 [Thu May 29 01:19:00 2014] [debug] ssl_scache_shmcb.c(316): index_num = 133 [Thu May 29 01:19:00 2014] [info] Shared memory session cache initialised [Thu May 29 01:19:00 2014] [info] Init: Initializing (virtual) servers for SSL [Thu May 29 01:19:00 2014] [info] mod_ssl/2.2.22 compiled against Server: Apache/2.2.22, Library: OpenSSL/1.0.1e [Thu May 29 01:19:00 2014] [notice] Apache/2.2.22 (Debian) PHP/5.4.4-14+deb7u9 mod_ssl/2.2.22 OpenSSL/1.0.1e mod_perl/2.0.7 Perl/v5.14.2 configured -- resuming normal operations [Thu May 29 01:19:00 2014] [info] Server built: Mar 4 2013 22:05:16 [Thu May 29 01:19:00 2014] [debug] prefork.c(1023): AcceptMutex: sysvsem (default: sysvsem) I've ensured to enable each vhost with a2ensite {sitename.conf} with no errors there, either. Below are the contents of the configuration files... /etc/apache2/apache2.conf # Global configuration # LockFile ${APACHE_LOCK_DIR}/accept.lock PidFile ${APACHE_PID_FILE} Timeout 300 KeepAlive On MaxKeepAliveRequests 100 KeepAliveTimeout 5 ## ## Server-Pool Size Regulation (MPM specific) ## # prefork MPM # StartServers: number of server processes to start # MinSpareServers: minimum number of server processes which are kept spare # MaxSpareServers: maximum number of server processes which are kept spare # MaxClients: maximum number of server processes allowed to start # MaxRequestsPerChild: maximum number of requests a server process serves <IfModule mpm_prefork_module> StartServers 5 MinSpareServers 5 MaxSpareServers 10 MaxClients 150 MaxRequestsPerChild 0 </IfModule> # worker MPM # StartServers: initial number of server processes to start # MinSpareThreads: minimum number of worker threads which are kept spare # MaxSpareThreads: maximum number of worker threads which are kept spare # ThreadLimit: ThreadsPerChild can be changed to this maximum value during a # graceful restart. ThreadLimit can only be changed by stopping # and starting Apache. # ThreadsPerChild: constant number of worker threads in each server process # MaxClients: maximum number of simultaneous client connections # MaxRequestsPerChild: maximum number of requests a server process serves <IfModule mpm_worker_module> StartServers 2 MinSpareThreads 25 MaxSpareThreads 75 ThreadLimit 64 ThreadsPerChild 25 MaxClients 150 MaxRequestsPerChild 0 </IfModule> # event MPM # StartServers: initial number of server processes to start # MinSpareThreads: minimum number of worker threads which are kept spare # MaxSpareThreads: maximum number of worker threads which are kept spare # ThreadsPerChild: constant number of worker threads in each server process # MaxClients: maximum number of simultaneous client connections # MaxRequestsPerChild: maximum number of requests a server process serves <IfModule mpm_event_module> StartServers 2 MinSpareThreads 25 MaxSpareThreads 75 ThreadLimit 64 ThreadsPerChild 25 MaxClients 150 MaxRequestsPerChild 0 </IfModule> # These need to be set in /etc/apache2/envvars User ${APACHE_RUN_USER} Group ${APACHE_RUN_GROUP} # # AccessFileName: The name of the file to look for in each directory # for additional configuration directives. See also the AllowOverride # directive. # AccessFileName .htaccess # # The following lines prevent .htaccess and .htpasswd files from being # viewed by Web clients. # <Files ~ "^\.ht"> Order allow,deny Deny from all Satisfy all </Files> DefaultType None HostnameLookups Off ErrorLog ${APACHE_LOG_DIR}/error.log LogLevel debug # Include module configuration: Include mods-enabled/*.load Include mods-enabled/*.conf # Include list of ports to listen on and which to use for name based vhosts Include ports.conf # # The following directives define some format nicknames for use with # a CustomLog directive (see below). # If you are behind a reverse proxy, you might want to change %h into %{X-Forwarded-For}i # # LogFormat "%v:%p %h %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" vhost_combined LogFormat "%h %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" combined LogFormat "%h %l %u %t \"%r\" %>s %O" common LogFormat "%{Referer}i -> %U" referer LogFormat "%{User-agent}i" agent <Directory "/var/www"> Order allow,deny Allow from all Require all granted </Directory> # Include generic snippets of statements Include conf.d/ # Include the virtual host configurations: Include sites-enabled/*.conf NameVirtualHost *:80 /etc/apache2/sites-available/site1.net.conf <VirtualHost *:80> ServerName site1.net ServerAlias site1.net *.site1.net DocumentRoot "/var/www/site1" ErrorLog "/var/www/site1/logs/error.log" CustomLog "/var/www/site1/logs/access.log" vhost_combined <Directory "/var/www/site1"> Options None AllowOverride All Order allow,deny Allow from all Satisfy Any </Directory> </VirtualHost> /etc/apache2/sites-available/site2.com.conf <VirtualHost *:80> ServerName site2.com ServerAlias site2.com *.site2.com DocumentRoot "/var/www/site2" ErrorLog "/var/www/site2/logs/error.log" CustomLog "/var/www/site2/logs/access.log" vhost_combined <Directory "/var/www/site2"> Options None AllowOverride All Order allow,deny Allow from all Satisfy Any </Directory> </VirtualHost> I've also tried setting NameVirtualHost like: Listen 80 NameVirtualHost 23.88.121.82:80 NameVirtualHost 127.0.0.1:80 and the VirtualHost Directives: <VirtualHost 23.88.121.82:80> ... </VirtualHost> for both sites, but that causes the first site to fail, as well. I'm wondering if I need to set up individual IPs for each site, possibly? I have 2 more IPv4 and 3 IPv6 addresses available, if that would make a difference. Also, in the grand scheme of things, I will need to enable SSL for the first site. I've been reading that I'll need to basically just mimic the directives for listening on port 80, only on port 443, and make sure mod_ssl is enabled? EDIT: I just ran apache2 -t to test the config files that way, and got the error: apache2: bad user name ${APACHE_RUN_USER}. However, apachectl configtest returns Syntax OK. There are no other mentions of errors with the mutex anywhere else, however. I was pretty sure if there was an error with the user apache was supposed to run under, the server wouldn't start at all... EDIT 2: Restarting apache fixed the bad user name error.

    Read the article

  • Cannot decode complete cipher list in .NET SslStream handshake.

    - by karmasponge
    While attempting to move from a 'C' based SSL implementation to C# using the .NET SslStream and we have run into what look like cipher compatibility issues with the .NET SslStream and a AS400 machine we are trying to connect to (which worked previously). When we call SslStream.AuthenticateAsClient it is sending the following: 16 03 00 00 37 01 00 00 33 03 00 4d 2c 00 ee 99 4e 0c 5d 83 14 77 78 5c 0f d3 8f 8b d5 e6 b8 cd 61 0f 29 08 ab 75 03 f7 fa 7d 70 00 00 0c 00 05 00 0a 00 13 00 04 00 02 00 ff 01 00 Which decodes as (based on http://www.mozilla.org/projects/security/pki/nss/ssl/draft302.txt) [16] Record Type [03 00] SSL Version [00 37] Body length [01] SSL3_MT_CLIENT_HELLO [00 00 33] Length (51 bytes) [03 00] Version number = 768 [4d 2c 00 ee] 4 Bytes unix time [… ] 28 Bytes random number [00] Session number [00 0c] 12 bytes (2 * 6 Cyphers)? [00 05, 00 0a, 00 13, 00 04, 00 02, 00 ff] - [RC4, PBE-MD5-DES, RSA, MD5, PKCS, ???] [01 00] Null compression method The as400 server responds back with: 15 03 00 00 02 02 28 [15] SSL3_RT_ALERT [03 00] SSL Version [00 02] Body Length (2 Bytes) [02 28] 2 = SSL3_RT_FATAL, 40 = SSL3_AD_HANDSHAKE_FAILURE I'm specifically looking to decode the '00 FF' at the end of the cyphers. Have I decoded it correctly? What does, if anything, '00 FF' decode too? I am using the following code to test/reproduce: using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Net.Sockets; using System.Net.Security; using System.Security.Authentication; using System.IO; using System.Diagnostics; using System.Security.Cryptography.X509Certificates; namespace TestSslStreamApp { class DebugStream : Stream { private Stream AggregatedStream { get; set; } public DebugStream(Stream stream) { AggregatedStream = stream; } public override bool CanRead { get { return AggregatedStream.CanRead; } } public override bool CanSeek { get { return AggregatedStream.CanSeek; } } public override bool CanWrite { get { return AggregatedStream.CanWrite; } } public override void Flush() { AggregatedStream.Flush(); } public override long Length { get { return AggregatedStream.Length; } } public override long Position { get { return AggregatedStream.Position; } set { AggregatedStream.Position = value; } } public override int Read(byte[] buffer, int offset, int count) { int bytesRead = AggregatedStream.Read(buffer, offset, count); return bytesRead; } public override long Seek(long offset, SeekOrigin origin) { return AggregatedStream.Seek(offset, origin); } public override void SetLength(long value) { AggregatedStream.SetLength(value); } public override void Write(byte[] buffer, int offset, int count) { AggregatedStream.Write(buffer, offset, count); } } class Program { static void Main(string[] args) { const string HostName = "as400"; TcpClient tcpClient = new TcpClient(HostName, 992); SslStream sslStream = new SslStream(new DebugStream(tcpClient.GetStream()), false, null, null, EncryptionPolicy.AllowNoEncryption); sslStream.AuthenticateAsClient(HostName, null, SslProtocols.Ssl3, false); } } }

    Read the article

  • Citrix Access Gateway with Citrix Receiver

    - by vm370
    I'm currently using a Citrix Access Gateway with firmware 5.0.4 to provide access over an SSL-VPN to an isolated environment, which is connected to over the Citrix Access Gateway Client, which is delivered by the device by default. However we encountered different problems, e.g. that it's somehow not possible to get it out of the autostart (not registered as a service or in the autostart?!) and it killed the Cisco VPN Client, which is used in the company and unfortunately cannot be replaced. The Cisco client can also just be used again after a procedure with cleaning the registry from all CAG Client remains, which requires a lot of effort. Because of that, I'd like to check if there is an alternative to this client, since is this is a real pain... Unfortunately I couldn't find a way to use the Receiver with the CAG yet, but if you have any resources on how to build this workaround, I'd be very happy. Thanks a lot in advance UPDATE: If there are other alternatives I'd be even more happy, since using the Receiver would also mean that there is an issue with the ICA-Client, which is also used in our environments. From my experience, the Receiver and the ICA-Client are also no good friends...

    Read the article

  • HTTP Upload Problems

    - by jfoster
    We are running a marketplace on ColdFusion8 and IIS with a widely geographically distributed user base and have been receiving complaints of issues with some HTTP uploads. Most of the complaints are coming from geographically distant locations from our main datacenter on the US east coast. I've attempted to upload the same 70MB file from a US West coast test server to both our main site and a backup running the same code on a different network route and I saw the same issues fairly consistently in both places, so I've ruled out the code, route, and internal network errors. I've also tested uploads using both the native cf upload tag and a third party tool called SaFileUp. I saw the same issues with both upload tools, so I also don't think this is necessarily a ColdFusion problem. I don't have any problems uploading the test file from the East coast to other east coast servers, so I'm beginning to think that the distance between our users and our equipment is a factor. I've also found that smaller files are more likely to succeed than large ones (< 10MB) I tried the test upload with both IE and FF and did notice a difference in the way that the browsers seemed to handle packet errors. IE seemed to have a tough time continuing an upload after dropped / bad packets, whereas FF seemed to have the ability to gracefully resume an upload after experiencing packet problems. Has anyone experienced similar issues? Is there anything we can do on our side to make uploads more forgiving to packet loss or resumable after an error? A different upload tool etc… Do we need upload servers in more than one location to shorten the network routes between clients and servers? Does anyone think that switching uploads to SSL will help (no layer7 packet sniffing may lead to a smoother upload). Thanks.

    Read the article

  • How can a Postfix/Dovecot(ssl)/Apache/Roundcube(non-ssl) setup leak email addresses?

    - by Jens Björnhager
    I have a linux box email server with Postfix as the MTA, Dovecot as the IMAP server and Apache with Roundcube as webmail. In my /etc/postfix/aliases I have just above a hundred different aliases which makes as many email addresses on my domain. I use one address per website so I easily can shut down spam infested addresses. During the half a year or so that I have had this setup, I have received 3 spam from 2 sources. As I know exactly where I entered this address, it should be easy to pinpoint email leaking websites and services. However, these sources are, according to me, not likely email sellers. And for one of them to sell my email twice? I contacted one of the sources and they are adamant that their system is tight. They suggested the possibility that it is my server that is doing the leaking. So, my question is: How likely is it that my box is leaking email addresses, and how? I don't store fully qualified email addresses anywhere in my system except in my maildir. I use SSL connection to IMAP I do not use https on webmail

    Read the article

  • Exchange 2003 ActiveSync problem

    - by colemanm
    We're having problems getting iPhones to sync properly with SBS 2003 Exchange. When you add a new Exchange ActiveSync account on an iPhone and enter all the pertinent information, it shows a "Verifying Exchange account info" message for a minute or so, then says everything's verified and asks what you want to sync, Mail, Contacts, Calendars... so it looks like it's working. However, when you go to the Mail app and select the Exchange email account, it just shows an "Inbox" folder with nothing in it. When you try refreshing, it attempts for a second, then says "Last Updated" with a timestamp, as if it worked, but there's no mail and no error message/feedback at all. I think I've narrowed it down to some sort of certificate issue, but I'm having trouble finding out where to go from here... I ran MS's Exchange connectivity testing tool with these results: Our cert was purchased from Network Solutions, and I'd already added it to the IIS Default Website for OWA purposes. But this report makes it look like the cert is somehow problematic. I don't know what to do now... Here's a shot of the cert details, just in case:

    Read the article

  • Google chrome not accepting any security certificates

    - by Jerry
    I've recently developed a problem with Google Chrome that's really annoying. I'm using Firefox at the moment with no problems whatsoever and it's the same with IE, so it's safe to say this problem is specific to Chrome. The problem is that it's not accepting security certificates from certain sites. I suppose the best place to start would be google itself. I can't search. The google search page will load but when I type some search term into the search box and hit 'search' I get the message: "You attempted to reach www.google.com, but the server presented an invalid certificate. You cannot proceed because the website operator has requested heightened security for this domain." No matter what the search term is, this is the result. Also when I try to log in to facebook - same message. Youtube works and many other sites that I know present security certs so I'm baffled. I've searched and there are other people who have had similar issues but I can't find a solution anywhere. The most common answer I'm picking up for this is to "check your system time" but I can safely say that it's not my system time. If anyone knows what is going on, I'd very much appreciate being informed. It's not super urgent as I can use Firefox to access those places Chrome won't, but it IS super annoying because I can usually sort out issues like this in no time.

    Read the article

  • What are the attack vectors for passwords sent over http?

    - by KevinM
    I am trying to convince a customer to pay for SSL for a web site that requires login. I want to make sure I correctly understand the major scenarios in which someone can see the passwords that are being sent. My understanding is that at any of the hops along the way can use a packet analyzer to view what is being sent. This seems to require that any hacker (or their malware/botnet) be on the same subnet as any of the hops the packet takes to arrive at its destination. Is that right? Assuming some flavor of this subnet requirement holds true, do I need to worry about all the hops or just the first one? The first one I can obviously worry about if they're on a public Wifi network since anyone could be listening in. Should I be worried about what's going on in subnets that packets will travel across outside this? I don't know a ton about network traffic, but I would assume it's flowing through data centers of major carriers and there's not a lot of juicy attack vectors there, but please correct me if I am wrong. Are there other vectors to be worried about outside of someone listening with a packet analyzer? I am a networking and security noob, so please feel free to set me straight if I am using the wrong terminology in any of this.

    Read the article

  • Apache: redirect to https before AUTH for server-status

    - by Putnik
    I want to force https and basic auth for server-status output (mod_status). If I enable auth and user asks for http://site/server-status apache first asks for pass, then redirects to httpS, then asks for pass again. This question is similar to Apache - Redirect to https before AUTH and force https with apache before .htpasswd but I cannot get it work because we are speaking not about generic folder but Location structure. My config (shortly) is as follows: <Location /server-status> SSLRequireSSL <IfModule mod_rewrite.c> RewriteEngine on RewriteBase /server-status RewriteCond %{HTTPS} off RewriteCond %{SERVER_PORT} 80 RewriteRule ^ - [E=nossl] RewriteRule (.*) https://site/server-status} [R=301,L] </IfModule> SetHandler server-status Order deny,allow Deny from all Allow from localhost ip6-localhost Allow from 1.2.3.0/24 Allow from env=nossl AuthUserFile /etc/httpd/status-htpasswd AuthName "Password protected" AuthType Basic Require valid-user Satisfy any </Location> I assume Allow from env=nossl should allow everyone with RewriteCond %{HTTPS} off and server port 80, then force it to redirect but it does not work. Please note, I do not want force to SSL the whole site but /server-status only. If it matters the server has several sites. What am I doing wrong? Thank you.

    Read the article

  • Authority Information Access local path being ignored

    - by Kevin
    I have a CA set up in Server 2008 R2, and generally it is working, but I can't control the local path/filename it writes its own certificate to for the Authority Information Access publishing. Here's a screen shot of the dialog I'm trying to set this on: From these settings I would expect to get the file: C:\Windows\system32\CertSrv\CertEnroll\DAMNIT.crt But instead I get: C:\Windows\system32\CertSrv\CertEnroll\SERVER.domain.com_My Issuing Authority(1).crt Of course, the actual change shown wouldn't be very useful, but it's illustrative; no matter what path/filename I use, it always lands up in the same place and with the same name. I actually wanted to change the name from <ServerDNSName>_<CaName><CertificateName>.crt to <CaName><CertificateName>.crt, since the latter corresponds to the HTTP URL whereas the former does not. Admittedly, I haven't set up many CAs so perhaps I'm just deluded as to what this dialog is supposed to be setting, but if so this is notoriously bad UI design. (Incidentally, I have a couple other complaints with the same dialog.) What's going on here and is there some way to get the filename pattern I want?

    Read the article

  • Freebsd jail for an small company - checklist - what shouldn't forget

    - by cajwine
    Looking for an checklist for an "small company freebsd/jail server". Having pretty common starting point: FreeBSD jail (remote/headless) for the company: public web, email, ftp server, and private (maybe in the future partially public) wiki (foswiki) 4 physical persons, (6 email addresses) + one admin - others will never use ssh) have already done usual hardening on the host side (like pf, sshguard etc). my major components are: dovecot, exim, apache22, proftpd, perl5.14. Looking for an checklist, what I shouldn't forget. My plan: openssl self-signed certificates for exim, dovecot and proftpd (wildcard keys) openssl self-signed certificate for apache (later will go for "trusted-signed" key) My questions are: is is an "good practice" having one pair of wildcard SSL-certificates for many programs? (exim, dovecot, proftpd) - or should I generate one key for each service? should I add all 4 persons as standard (unix) users, or I should go with virtual users? Asking because: have only small count of users, and it is more simple to configure everything (exim, dovecot) for local users ($HOME/Maildir), plus ability to set $HOME/.forward/vacation and etc. is here some (special) things what I should consider? (e.g. maybe, in the future we want setup our own webmail - will make this any difference?) any other recommendation? Thank you, hoping that this question fit into the http://serverfault.com/faq under the: Server and Business Workstation operating systems, hardware, software Operations, maintenance, and monitoring Looking for an checklist, but please explain why you're recommending it. See Good Subjective, Bad Subjective. related: What's your suggested mail server configuration for a FreeBSD server?

    Read the article

  • JBoss https on port other than 8080 not working

    - by MilindaD
    We have a server with two JBoss instances where one runs on 8080, the other on 8081. We need to have HTTPS enabled for the 8081 server, firstly we tried enabling https on the 8080 port instance by generating the keystore and editing the server.xml and it successfully worked. However when we tried the same thing for 8081 it did not, note that we removed https for the 8080 server first before enabling it for 8081. This is what was used for both server.xml for 8080 and 8081. The only difference was that the port was changed from 8080 to 8081 when trying to enable https for 8081 port instance. What am I doing wrong and what needs to be changed? NOTE : When I meant enabled for 8080 I meant when you visit https:// URL:8484 you will actually be visiting the 8080 port instance. However when ssl is enabled for 8081 and I visit https:// URL:8484 I get that the web page is unavailable. COMMENTLESS VERSION <Server> <Listener className="org.apache.catalina.core.AprLifecycleListener" SSLEngine="on" /> <Listener className="org.apache.catalina.core.JasperListener" /> <Service name="jboss.web"> <!-- https --> <Connector port="8080" address="${jboss.bind.address}" maxThreads="350" maxHttpHeaderSize="8192" emptySessionPath="true" protocol="HTTP/1.1" enableLookups="false" redirectPort="8443" acceptCount="100" connectionTimeout="20000" disableUploadTimeout="true" compression="on" ompressableMimeType="text/html,text/css,text/javascript,application/json,text/xml,text/plain,application/x-javascript,application/javascript"/> <Connector port="8443" protocol="HTTP/1.1" SSLEnabled="true" maxThreads="150" scheme="https" secure="true" clientAuth="false" sslProtocol="TLS" address="${jboss.bind.address}" keystoreFile="${jboss.server.home.dir}/conf/supun1.keystore" keystorePass="aaaaaa" truststoreFile="${jboss.server.home.dir}/conf/supun1.keystore" truststorePass="aaaaaa" /> <!-- https1 --> <Connector port="8009" address="${jboss.bind.address}" protocol="AJP/1.3" emptySessionPath="true" enableLookups="false" redirectPort="8443" /> <Engine name="jboss.web" defaultHost="localhost" jvmRoute="khms1"> <Realm className="org.jboss.web.tomcat.security.JBossSecurityMgrRealm" certificatePrincipal="org.jboss.security.auth.certs.SubjectDNMapping" allRolesMode="authOnly" /> <Host name="localhost" autoDeploy="false" deployOnStartup="false" deployXML="false" configClass="org.jboss.web.tomcat.security.config.JBossContextConfig" > <Valve className="org.jboss.web.tomcat.service.sso.ClusteredSingleSignOn" /> <Valve className="org.jboss.web.tomcat.service.jca.CachedConnectionValve" cachedConnectionManagerObjectName="jboss.jca:service=CachedConnectionManager" transactionManagerObjectName="jboss:service=TransactionManager" /> </Host> </Engine> </Service> </Server> WITH COMMENTS VERSION <Server> <!--APR library loader. Documentation at /docs/apr.html --> <Listener className="org.apache.catalina.core.AprLifecycleListener" SSLEngine="on" /> <!--Initialize Jasper prior to webapps are loaded. Documentation at /docs/jasper-howto.html --> <Listener className="org.apache.catalina.core.JasperListener" /> <!-- Use a custom version of StandardService that allows the connectors to be started independent of the normal lifecycle start to allow web apps to be deployed before starting the connectors. --> <Service name="jboss.web"> <!-- A "Connector" represents an endpoint by which requests are received and responses are returned. Documentation at : Java HTTP Connector: /docs/config/http.html (blocking & non-blocking) Java AJP Connector: /docs/config/ajp.html APR (HTTP/AJP) Connector: /docs/apr.html Define a non-SSL HTTP/1.1 Connector on port 8080 --> <Connector port="8080" address="${jboss.bind.address}" maxThreads="350" maxHttpHeaderSize="8192" emptySessionPath="true" protocol="HTTP/1.1" enableLookups="false" redirectPort="8443" acceptCount="100" connectionTimeout="20000" disableUploadTimeout="true" compression="on" ompressableMimeType="text/html,text/css,text/javascript,application/json,text/xml,text/plain,application/x-javascript,application/javascript"/> <!-- Define a SSL HTTP/1.1 Connector on port 8443 This connector uses the JSSE configuration, when using APR, the connector should be using the OpenSSL style configuration described in the APR documentation --> <!-- <Connector port="8443" protocol="HTTP/1.1" SSLEnabled="true" maxThreads="150" scheme="https" secure="true" keystoreFile="${jboss.server.home.dir}/conf/zara.keystore" keystorePass="zara2010" clientAuth="false" sslProtocol="TLS" compression="on" /> --> <Connector port="8443" protocol="HTTP/1.1" SSLEnabled="true" maxThreads="150" scheme="https" secure="true" clientAuth="false" sslProtocol="TLS" address="${jboss.bind.address}" keystoreFile="${jboss.server.home.dir}/conf/supun1.keystore" keystorePass="aaaaaa" truststoreFile="${jboss.server.home.dir}/conf/supun1.keystore" truststorePass="aaaaaa" /> <!-- Define an AJP 1.3 Connector on port 8009 --> <Connector port="8009" address="${jboss.bind.address}" protocol="AJP/1.3" emptySessionPath="true" enableLookups="false" redirectPort="8443" /> <Engine name="jboss.web" defaultHost="localhost" jvmRoute="khms1"> <!-- The JAAS based authentication and authorization realm implementation that is compatible with the jboss 3.2.x realm implementation. - certificatePrincipal : the class name of the org.jboss.security.auth.certs.CertificatePrincipal impl used for mapping X509[] cert chains to a Princpal. - allRolesMode : how to handle an auth-constraint with a role-name=*, one of strict, authOnly, strictAuthOnly + strict = Use the strict servlet spec interpretation which requires that the user have one of the web-app/security-role/role-name + authOnly = Allow any authenticated user + strictAuthOnly = Allow any authenticated user only if there are no web-app/security-roles --> <Realm className="org.jboss.web.tomcat.security.JBossSecurityMgrRealm" certificatePrincipal="org.jboss.security.auth.certs.SubjectDNMapping" allRolesMode="authOnly" /> <!-- A subclass of JBossSecurityMgrRealm that uses the authentication behavior of JBossSecurityMgrRealm, but overrides the authorization checks to use JACC permissions with the current java.security.Policy to determine authorized access. - allRolesMode : how to handle an auth-constraint with a role-name=*, one of strict, authOnly, strictAuthOnly + strict = Use the strict servlet spec interpretation which requires that the user have one of the web-app/security-role/role-name + authOnly = Allow any authenticated user + strictAuthOnly = Allow any authenticated user only if there are no web-app/security-roles <Realm className="org.jboss.web.tomcat.security.JaccAuthorizationRealm" certificatePrincipal="org.jboss.security.auth.certs.SubjectDNMapping" allRolesMode="authOnly" /> --> <Host name="localhost" autoDeploy="false" deployOnStartup="false" deployXML="false" configClass="org.jboss.web.tomcat.security.config.JBossContextConfig" > <!-- Uncomment to enable request dumper. This Valve "logs interesting contents from the specified Request (before processing) and the corresponding Response (after processing). It is especially useful in debugging problems related to headers and cookies." --> <!-- <Valve className="org.apache.catalina.valves.RequestDumperValve" /> --> <!-- Access logger --> <!-- <Valve className="org.apache.catalina.valves.AccessLogValve" prefix="localhost_access_log." suffix=".log" pattern="common" directory="${jboss.server.log.dir}" resolveHosts="false" /> --> <!-- Uncomment to enable single sign-on across web apps deployed to this host. Does not provide SSO across a cluster. If this valve is used, do not use the JBoss ClusteredSingleSignOn valve shown below. A new configuration attribute is available beginning with release 4.0.4: cookieDomain configures the domain to which the SSO cookie will be scoped (i.e. the set of hosts to which the cookie will be presented). By default the cookie is scoped to "/", meaning the host that presented it. Set cookieDomain to a wider domain (e.g. "xyz.com") to allow an SSO to span more than one hostname. --> <!-- <Valve className="org.apache.catalina.authenticator.SingleSignOn" /> --> <!-- Uncomment to enable single sign-on across web apps deployed to this host AND to all other hosts in the cluster. If this valve is used, do not use the standard Tomcat SingleSignOn valve shown above. Valve uses a JBossCache instance to support SSO credential caching and replication across the cluster. The JBossCache instance must be configured separately. By default, the valve shares a JBossCache with the service that supports HttpSession replication. See the "jboss-web-cluster-service.xml" file in the server/all/deploy directory for cache configuration details. Besides the attributes supported by the standard Tomcat SingleSignOn valve (see the Tomcat docs), this version also supports the following attributes: cookieDomain see above treeCacheName JMX ObjectName of the JBossCache MBean used to support credential caching and replication across the cluster. If not set, the default value is "jboss.cache:service=TomcatClusteringCache", the standard ObjectName of the JBossCache MBean used to support session replication. --> <Valve className="org.jboss.web.tomcat.service.sso.ClusteredSingleSignOn" /> <!-- Check for unclosed connections and transaction terminated checks in servlets/jsps. Important: The dependency on the CachedConnectionManager in META-INF/jboss-service.xml must be uncommented, too --> <Valve className="org.jboss.web.tomcat.service.jca.CachedConnectionValve" cachedConnectionManagerObjectName="jboss.jca:service=CachedConnectionManager" transactionManagerObjectName="jboss:service=TransactionManager" /> </Host> </Engine> </Service> </Server>

    Read the article

  • Exchange - inbound email only works from some servers

    - by Kryptonite
    I am having a problem where inbound mail from outside only works when sent from certain hosts. For example, when I send myself an email from my personal gmail account all is well, as the logs show: 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 EHLO 250 - - 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 STARTTLS 220 - - 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 STARTTLS 220 - - 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 EHLO 250 - - 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 MAIL 250 - - 2012-09-05 18:14:16 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 RCPT 250 - - 2012-09-05 18:14:48 209.85.223.175 mail-ie0-f175.google.com SMTPSVC1 MAILSVR 192.168.1.79 0 QUIT 240 - - However, if I sent from my personal Yahoo account, I get this response: Sorry, we were unable to deliver your message to the following address. <[email protected]>: Remote host said: 530 5.7.0 Must issue a STARTTLS command first [MAIL_FROM] (NB: Nothing appeared in the smtp log for this message.) Any suggestions where to start looking? EDIT ---- I don't know if it matters, but the certificate I am using for TLS is self signed.

    Read the article

  • Apache proxy: Why is one vhost returning Forbidden while the other one works?

    - by Stefan Majewsky
    I have a Java application that needs to talk to another intranet website using HTTPS in both directions. After fighting with Java's SSL implementations for some time, I gave up on that, and have now set up an Apache that's supposed to act as a bidirectional reverse proxy: external app ---(HTTPS request)---> Apache ---(local HTTP request)---> Java app This direction works just fine, however the other direction does not: Java app ---(local HTTP request)---> Apache ---(HTTPS request)---> external app This is the configuration for the vhost implementing the second proxy: Listen 127.0.0.1:8081 <VirtualHost appgateway:8081> ServerName appgateway.local SSLProxyEngine on ProxyPass / https://externalapp.corp:443/ ProxyPassReverse / https://externalapp.corp:443/ ProxyRequests Off AllowEncodedSlashes On # we do not need to apply any more restrictions here, because we listened on # local connections only in the first place (see the Listen directive above) <Proxy https://externalapp.corp:443/*> Order deny,allow Allow from all </Proxy> </VirtualHost> A curl http://127.0.0.1:8081/ should serve the equivalent of https://externalapp.corp, but instead results in 403 Forbidden, with the following message in the Apache error log: [Wed Jun 04 08:57:19 2014] [error] [client 127.0.0.1] Directory index forbidden by Options directive: /srv/www/htdocs/ This message completely puzzles me: Yes, I have not set up any permissions on the DocumentRoot of this vhost, but everything works fine for the other proxy direction where I haven't. For reference, here's the other vhost: Listen this_vm_hostname:443 <VirtualHost javaapp:443> ServerName javaapp.corp SSLEngine on SSLProxyEngine on # not shown: SSLCipherSuite, SSLCertificateFile, SSLCertificateKeyFile SSLOptions +StdEnvVars ProxyPass / http://localhost:8080/ ProxyPassReverse / http://localhost:8080/ ProxyRequests Off AllowEncodedSlashes On # Local reverse proxy authorization override <Proxy http://localhost:8080/*> Order deny,allow Allow from all </Proxy> </VirtualHost>

    Read the article

  • Does SNI represent a privacy concern for my website visitors?

    - by pagliuca
    Firstly, I'm sorry for my bad English. I'm still learning it. Here it goes: When I host a single website per IP address, I can use "pure" SSL (without SNI), and the key exchange occurs before the user even tells me the hostname and path that he wants to retrieve. After the key exchange, all data can be securely exchanged. That said, if anybody happens to be sniffing the network, no confidential information is leaked* (see footnote). On the other hand, if I host multiple websites per IP address, I will probably use SNI, and therefore my website visitor needs to tell me the target hostname before I can provide him with the right certificate. In this case, someone sniffing his network can track all the website domains he is accessing. Are there any errors in my assumptions? If not, doesn't this represent a privacy concern, assuming the user is also using encrypted DNS? Footnote: I also realize that a sniffer could do a reverse lookup on the IP address and find out which websites were visited, but the hostname travelling in plaintext through the network cables seems to make keyword based domain blocking easier for censorship authorities.

    Read the article

  • Can we do a DNSSEC 101? [closed]

    - by PAStheLoD
    Please share your opinions, FAQs, HOWTOs, best practices (or links to the one you think is the best) and your fears and thoughts about the whole migration (or should I just call it a new piece of tech?). Is DNSSEC just for DNS providers (name server operators)? What ought John Doe to do, who hosts johndoe.com at some random provider (GoDaddy, DreamHost and such)? Also, what if the provider's name server doesn't do automatic signing magic, can John do it manually? In a fire-and-forget way, without touching KSKs and ZSKs rollovers and updating and headaches?) Does it bring any change regarding CERT records? Do browsers support it? How come it became so complex? Why didn't they just merged it with SSL? DKIM is pretty straightforward, IANA/IETF could've opted for something like that. (Yes I know that creating a trust anchor would be still problematic, but browsers are already full of CA certs. So, they could've just let anyone get a cert for a domain for shiny green padlocks, or just generate one for a poor blue lock, put it into a TXT record, encrypt the other records and let the parent zone sign the whole for you with its cert.) Thanks! And for disclosure (it seemed like the customary thing to do around here), I've asked the same on the netsec subreddit.

    Read the article

  • Apache load balancer with https real servers and client certificates

    - by Jack Scheible
    Our network requirements state that ALL network traffic must be encrypted. The network configuration looks like this: ------------ /-- https --> | server 1 | / ------------ |------------| |---------------|/ ------------ | Client | --- https --> | Load Balancer | ---- https --> | server 2 | |------------| |---------------|\ ------------ \ ------------ \-- https --> | server 3 | ------------ And it has to pass client certificates. I've got a config that can do load balancing with in-the-clear real servers: <VirtualHost *:8666> DocumentRoot "/usr/local/apache/ssl_html" ServerName vmbigip1 ServerAdmin [email protected] DirectoryIndex index.html <Proxy *> Order deny,allow Allow from all </Proxy> SSLEngine on SSLProxyEngine On SSLCertificateFile /usr/local/apache/conf/server.crt SSLCertificateKeyFile /usr/local/apache/conf/server.key <Proxy balancer://mycluster> BalancerMember http://1.2.3.1:80 BalancerMember http://1.2.3.2:80 # technically we aren't blocking anyone, but could here Order Deny,Allow Deny from none Allow from all # Load Balancer Settings # A simple Round Robin load balancer. ProxySet lbmethod=byrequests </Proxy> # balancer-manager # This tool is built into the mod_proxy_balancer module allows you # to do simple mods to the balanced group via a gui web interface. <Location /balancer-manager> SetHandler balancer-manager Order deny,allow Allow from all </Location> ProxyRequests Off ProxyPreserveHost On # Point of Balance # Allows you to explicitly name the location in the site to be # balanced, here we will balance "/" or everything in the site. ProxyPass /balancer-manager ! ProxyPass / balancer://mycluster/ stickysession=JSESSIONID </VirtualHost> What I need is for the servers in my load balancer to be BalancerMember https://1.2.3.1:443 BalancerMember https://1.2.3.2:443 But that does not work. I get SSL negotiation errors. Even when I do get that to work, I will need to pass client certificates. Any help would be appreciated.

    Read the article

  • Why does Google Analytics use two domains?

    - by AKeller
    I'm building a distributed widget that is comparable to Google Analytics. Users will add a <script> tag to their site that references my widget's JavaScript file. The Google Analytics tracking code looks like this: var _gaq = _gaq || []; _gaq.push(['_setAccount', 'UA-XXXXXXXX-X']); _gaq.push(['_trackPageview']); (function () { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Can anyone explain the reasoning behind separate HTTP and HTTPS hostnames? My instinct is to just secure the www address and then use the protocol-less syntax, like //www.google-analytics.com/ga.js. But I'm sure the Google Analytics architects put a lot of thought into this approach. I'd love to understand their logic before I follow/ignore their model.

    Read the article

  • Virtualhost one https site, the rest http

    - by RJP1
    I have a linode server with Apache2 running a handful of sites with virtualhosting. All sites work fine on port 80, but one site has a ssl certificate and also runs okay. My problem is as follows: The non-https sites, if visiting https://domain.com - show the contents of the only secure site... Is there a way of disabling the *:443 match for these non-secure sites? Thanks! EDIT (more information): Here's a typical config in sites-available for a normal insecure http site: <VirtualHost *:80> ServerName www.insecure.com ServerAlias insecure.com ... </VirtualHost> The secure https site is as follows: <VirtualHost *:80> ServerName www.secure.com Redirect permanent / https://secure.com/ </VirtualHost> <VirtualHost *:80> ServerName secure.com RedirectMatch permanent ^/(.*) https://secure.com/$1 </VirtualHost> <VirtualHost *:443> SSLEngine on SSLProtocol all SSLCertificateChainFile ... SSLCertificateFile ... SSLCertificateKeyFile ... SSLCACertificateFile ... ServerName secure.com ServerAlias secure.com ... </VirtualHost> So, visiting: http:/insecure.com - works http:/www.insecure.com - works http:/secure.com - redirects to https:/secure.com - works http:/www.secure.com - redirects to https:/secure.com - works https:/insecure.com - shows https:/secure.com - WRONG!

    Read the article

< Previous Page | 65 66 67 68 69 70 71 72 73 74 75 76  | Next Page >