Re: [users@httpd] connection closed for event mpm?

2013-03-28 Thread Marcin Wanat
What error_log says ? It looks like httpd is crashing.
Maybe you are using some flv streaming mod which is not thread safe ?

2013/3/28 Esmq e...@163.com:
 hi, guys

 i have come across a problem related to event mpm in apache(v2.4.3),
 that is
 connection may closed during transfer for some times. when i change to
 prefork mpm, everything goes well...
 i also try to set EnableSendfile  EnableMMAP to Off, but without work...

 following is the full Scenario of the problem..

 wget -O a.flv http://video.test.com/a.flv; md5sum a.flv
 --2013-03-27 18:15:51-- http://video.test.com/a.flv
 Resolving video.test.com... 192.168.1.100
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 200 OK
 Length: 290899940 (277M) [video/x-flv]
 Saving to: “a.flv”

 19% [=== ] 56,424,265  2.50M/s   in 16s

 2013-03-27 18:16:07 (3.30 MB/s) - Connection closed at byte 56424265.
 Retrying.

 --2013-03-27 18:16:08--  (try: 2) http://video.test.com/a.flv
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 206 Partial Content
 Length: 290899940 (277M), 234475675 (224M) remaining [video/x-flv]
 Saving to: “a.flv”

 19% [ ] 56,424,265  --.-K/s   in 11s

 2013-03-27 18:16:20 (0.00 B/s) - Connection closed at byte 56424265.
 Retrying.

 --2013-03-27 18:16:22--  (try: 3) http://video.test.com/a.flv
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 206 Partial Content
 Length: 290899940 (277M), 234475675 (224M) remaining [video/x-flv]
 Saving to: “a.flv”

 100%[]
 290,899,940 3.37M/s   in 58s

 2013-03-27 18:17:20 (3.86 MB/s) - “a.flv” saved [290899940/290899940]

 671034be784ce6eb7bc9366572f5441f  a.flv



-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



[users@httpd] Re:Re: [users@httpd] connection closed for event mpm?

2013-03-28 Thread Esmq
i dont use flv module, and the error_log not provide any clue...~


and someone also reported the problem:
http://marc.info/?l=apache-httpd-bugsm=135712896422945
At 2013-03-28 17:19:21,Marcin Wanat marcin.wa...@gmail.com wrote:
What error_log says ? It looks like httpd is crashing.
Maybe you are using some flv streaming mod which is not thread safe ?

2013/3/28 Esmq e...@163.com:
 hi, guys

 i have come across a problem related to event mpm in apache(v2.4.3),
 that is
 connection may closed during transfer for some times. when i change to
 prefork mpm, everything goes well...
 i also try to set EnableSendfile  EnableMMAP to Off, but without work...

 following is the full Scenario of the problem..

 wget -O a.flv http://video.test.com/a.flv; md5sum a.flv
 --2013-03-27 18:15:51-- http://video.test.com/a.flv
 Resolving video.test.com... 192.168.1.100
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 200 OK
 Length: 290899940 (277M) [video/x-flv]
 Saving to: “a.flv”

 19% [=== ] 56,424,265  2.50M/s   in 16s

 2013-03-27 18:16:07 (3.30 MB/s) - Connection closed at byte 56424265.
 Retrying.

 --2013-03-27 18:16:08--  (try: 2) http://video.test.com/a.flv
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 206 Partial Content
 Length: 290899940 (277M), 234475675 (224M) remaining [video/x-flv]
 Saving to: “a.flv”

 19% [ ] 56,424,265  --.-K/s   in 11s

 2013-03-27 18:16:20 (0.00 B/s) - Connection closed at byte 56424265.
 Retrying.

 --2013-03-27 18:16:22--  (try: 3) http://video.test.com/a.flv
 Connecting to video.test.com|192.168.1.100|:80... connected.
 HTTP request sent, awaiting response... 206 Partial Content
 Length: 290899940 (277M), 234475675 (224M) remaining [video/x-flv]
 Saving to: “a.flv”

 100%[]
 290,899,940 3.37M/s   in 58s

 2013-03-27 18:17:20 (3.86 MB/s) - “a.flv” saved [290899940/290899940]

 671034be784ce6eb7bc9366572f5441f  a.flv



-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



Re: [users@httpd] Mod_proxy: Authentication-Info header lost in response

2013-03-28 Thread Nicolas Daniels

Hi,

I finally solved the problem myself by modifying the source code of 
mod_proxy_http.c:

Line 1697:
apr_table_do(addit_dammit, save_table, r-headers_out, Set-Cookie, NULL);
--
apr_table_do(addit_dammit, save_table, r-headers_out, Set-Cookie, 
Authentication-Info, NULL);


Cheers


On 27/03/2013 13:23, Nicolas Daniels wrote:

Ok, I was probably not clear enough ;-)

First I'm using mod_proxy_http and DIGEST authentication. 
Authentication-Info header is part of digest authentication:

http://rfc-ref.org/RFC-TEXTS/2069/chapter2.html

Lets say I've 2 accessed URLs:

http://mydomain.com/index.html
http://mydomain.com/tomcat/index.html

Both are using digest authentication on apache.

Proxy is configured as follow:
ProxyPass  /tomcat http://mytomcat.com/bla
ProxyPassReverse  /tomcathttp://mytomcat.com/bla
So http://mydomain.com/index.html is replied directly by apache and 
http://mydomain.com/tomcat/index.html is proxied to tomcat.


1st case: Authentication-Info replied

 GET /index.html HTTP/1.1
 User-Agent: curl/7.29.0
 Host: mydomain.com
 Accept: */*

 HTTP/1.1 401 Unauthorized
 Date: Wed, 27 Mar 2013 11:24:18 GMT
 Server: Apache/2.4.4 (Unix)
 WWW-Authenticate: Digest realm=bla, 
nonce=nxteR+bYBAA=9c9e9d4176b1ff722c18122c2a3a9af3d52b6e8a, 
algorithm=MD5, qop=auth

 Content-Length: 381
 Content-Type: text/html; charset=iso-8859-1

 GET /index.html HTTP/1.1
 Authorization: Digest username=username, realm=bla, 
nonce=nxteR+bYBAA=9c9e9d4176b1ff722c18122c2a3a9af3d52b6e8a, 
uri=/index.html, cnonce
=ICAgICAgICAgICAgICAgICAgICAgICAgICAxNDEyNjc=, nc=0001, 
qop=auth, response=bbfa7dqsdqs2c014d85sqdzaab1, algorithm=MD5

 User-Agent: curl/7.29.0
 Host: mydomain.com
 Accept: */*

 HTTP/1.1 200 OK
 Date: Wed, 27 Mar 2013 11:24:18 GMT
 Server: Apache/2.4.4 (Unix)
* Authentication-Info: rspauth=efbdcdsqdsqhiaaazqds4eee3c1, 
cnonce=ICAgICAgICAgICAgICAgICAgICAgICAgICAxNDEyNjc=, nc=0001, 
qop=auth*

 Last-Modified: Tue, 19 Feb 2013 08:24:06 GMT
 ETag: 22-4d60f909e7580
 Accept-Ranges: bytes
 Content-Length: 34
 Content-Type: text/plain


2nd case: Authentication-Info *not* replied

 GET /tomcat/index.html HTTP/1.1
 User-Agent: curl/7.29.0
 Host: mydomain.com
 Accept: */*

 HTTP/1.1 401 Unauthorized
 Date: Wed, 27 Mar 2013 12:15:25 GMT
 Server: Apache/2.4.4 (Unix)
 WWW-Authenticate: Digest realm=bla, 
nonce=5X4sqdsqdsqd456sq4dsq4d65sq78zf599bbd478c, algorithm=MD5, 
qop=auth

 Content-Length: 381
 Content-Type: text/html; charset=iso-8859-1

 GET /tomcat/index.html HTTP/1.1
 Authorization: Digest username=username, realm=bla, 
nonce=5X4sqdsqdsqd456sq4dsq4d65sq78zf599bbd478c, 
uri=/tomcat/index.html, 
cnonce=ICAgICAgICAgICAgICAgICAgICAgICAgICA0NDk5NzM=, nc=0001, 
qop=auth, response=cf10890c9dsqdsqef3bd248dsqdsqec34, algorithm=MD5

 User-Agent: curl/7.29.0
 Host: mydomain.com
 Accept: */*

 HTTP/1.1 200 OK
 Date: Wed, 27 Mar 2013 12:15:27 GMT
 Server: Apache-Coyote/1.1
 Content-Type: application/json
 Content-Length: 142
.

So my question is, is there any way to have Apache reply this 
Authentication-Info in both case ? I guess the reverse proxy should 
add is somehow...


Thanks !

On 27/03/2013 13:00, Nick Kew wrote:

On 27 Mar 2013, at 11:39, Nicolas Daniels wrote:


Everything work fine except that when the proxy is used, the 
Authentication-Info header is not included in the response. If Apache is 
replying directly without using the proxy, it is included.

There's no such header in HTTP.  Why not tell us exactly what you mean?







Re: [users@httpd] Control of reverse proxy workers

2013-03-28 Thread Alex Malek
Is there a way to have a script which can be allowed to enabled/disable
workers used in a reverse proxy.

Similar to what can be done using the balancer-manager page, but without
needing to have something that can emulate a web browser to do the job.

I'd be interested to know this too.

I wound up writing a perl script using WWW::Mechanize to handle the
web scraping.
I allowed unauthenticated access from localhost and run the script locally.

My use case was I wanted which workers were enabled/disable to persist
after an apache restart and/or failover to another server.

-Alex

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



[users@httpd] JkWorkerFile (mod_jk)

2013-03-28 Thread Chris Arnold
Apache2.12.x on SLES11 SP2.
I am trying to get apache2 mod_jk working with apache to access a webapp. I 
have loaded the mod_jk module and configured a virtual host and 
workers.properties file like so:

virtualhost:
VirtualHost *:80
ServerName share.domain.com

IfModule mod_jk.c

# The following line makes apache aware of the location of
# the /jsp-examples context
Alias /share /opt/alfresco/tomcat/webapps/share
Directory /opt/alfresco/tomcat/webapps
Options Indexes FollowSymLinks
allow from all
/Directory

# The following line mounts all JSP files and the /servlet/ uri to tomcat
#JkMount /servlets-examples/servlet/* ajp13
JkMount /share/*.jsp ajp13

# The following line prohibits users from directly accessing WEB-INF
Location /share/WEB-INF/
AllowOverride None
deny from all
/Location

/IfModule

/VirtualHost

jk.conf:
IfModule mod_jk.c

JkWorkersFile /opt/alfresco/tomcat/workers.properties
JkLogFile /var/log/alfresco/mod_jk.log
JkShmFile /var/log/alfresco/shm

# Log level to be used by mod_jk
JkLogLevel error

/IfModule

When i restart apache, i get:
JkWorkerFile only allowed once
and apache does not start until i comment out the jkworkerfile line in jk.conf.
So my question is, does apache already have a jkworkerfile somewhere that i do 
not know about? I searched for workers.properties and see only 2, 1 is mine 
that i created and the other is in /usr/share/doc/packages/apache2-mod-jk.
I do not have tomcat installed outside of the tomcat the webapp (which is 
alfresco) installed.

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



Re: [users@httpd] JkWorkerFile (mod_jk)

2013-03-28 Thread Tom Evans
On Thu, Mar 28, 2013 at 4:20 PM, Chris Arnold
carn...@electrichendrix.com wrote:
 Apache2.12.x on SLES11 SP2.
 I am trying to get apache2 mod_jk working with apache to access a webapp. I 
 have loaded the mod_jk module and configured a virtual host and 
 workers.properties file like so:

 virtualhost:
 VirtualHost *:80
 ServerName share.domain.com

 IfModule mod_jk.c

 # The following line makes apache aware of the location of
 # the /jsp-examples context
 Alias /share /opt/alfresco/tomcat/webapps/share
 Directory /opt/alfresco/tomcat/webapps
 Options Indexes FollowSymLinks
 allow from all
 /Directory

 # The following line mounts all JSP files and the /servlet/ uri to tomcat
 #JkMount /servlets-examples/servlet/* ajp13
 JkMount /share/*.jsp ajp13

 # The following line prohibits users from directly accessing WEB-INF
 Location /share/WEB-INF/
 AllowOverride None
 deny from all
 /Location

 /IfModule

 /VirtualHost

 jk.conf:
 IfModule mod_jk.c

 JkWorkersFile /opt/alfresco/tomcat/workers.properties
 JkLogFile /var/log/alfresco/mod_jk.log
 JkShmFile /var/log/alfresco/shm

 # Log level to be used by mod_jk
 JkLogLevel error

 /IfModule

 When i restart apache, i get:
 JkWorkerFile only allowed once
 and apache does not start until i comment out the jkworkerfile line in 
 jk.conf.
 So my question is, does apache already have a jkworkerfile somewhere that i 
 do not know about? I searched for workers.properties and see only 2, 1 is 
 mine that i created and the other is in 
 /usr/share/doc/packages/apache2-mod-jk.
 I do not have tomcat installed outside of the tomcat the webapp (which is 
 alfresco) installed.


Are you including this config file multiple times? Perhaps once
explicitly, and once as a glob include?

Cheers

Tom

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



Re: [users@httpd] JkWorkerFile (mod_jk)

2013-03-28 Thread Chris Arnold
On Mar 28, 2013, at 12:26 PM, Tom Evans tevans...@googlemail.com wrote:

 Are you including this config file multiple times? Perhaps once
 explicitly, and once as a 

That I know of, this is the only time it is being called. With that said, in 
httpd.conf I have an include statement pointing to /etc/apache2/conf.d. Maybe 
jk.conf is placed in a place where it is automatically being read/loaded and my 
file reads/loads it also?

Re: [users@httpd] Mod_proxy: Authentication-Info header lost in response

2013-03-28 Thread Nick Kew
On Thu, 28 Mar 2013 16:21:56 +0100
Nicolas Daniels nicolas.dani...@swing.be wrote:

 Hi,
 
 I finally solved the problem myself by modifying the source code of 
 mod_proxy_http.c:

Thanks!

I started to compose a reply yesterday, then decided I should look
at it first to determine whether it's a bug.  I might end up applying
your fix, if I don't forget about it first.


-- 
Nick Kew

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



[users@httpd] Followup to [Bug 50028] (LDAP authentication with encrypted passwords)

2013-03-28 Thread Ken Nishimura

Hi -

I've searched the archives and it looks like this was discussed back in 
2010 with a WONTFIX.  I just wanted to make sure this is still the case.


Basically, using the mod_auth_ldap module, apart from using SSL (and 
associated overhead), is it still the case that there is no way to 
encrypt just the passing of username and password from the client 
(browser) back to the server?


As others have pointed out, SSL is a fallback, but with associated 
overhead.  Has this been fixed in later versions of Apache?


Thanks,

Ken

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



Re: [users@httpd] JkWorkerFile (mod_jk)

2013-03-28 Thread Chris Arnold
On Mar 28, 2013, at 12:26 PM, Tom Evans wrote: 

 Are you including this config file multiple times? Perhaps once 
 explicitly, and once as a 

That I know of, this is the only time it is being called. With that said, in 
httpd.conf I have an include statement pointing to /etc/apache2 
/conf.d. Maybe jk.conf is placed in a place where it is automatically being 
read/loaded and my file reads/loads it also? 

Ok, i moved the jk.conf file to .opt/alfresco/tomcat/conf and now i do not get 
the only allowed once described in a previous post. 
However, now when accessing http://share.domain.com, the result is directories 
and files are listed, the jsp files are not running. Here is my complete setup: 
jk.conf- 
# simple configuration for apache (for AJP connector, modul mod_jk.so) 

IfModule mod_jk.c 

JkWorkersFile /opt/alfresco/tomcat/workers.properties 
JkLogFile /var/log/alfresco/mod_jk.log 
JkShmFile /var/log/alfresco/shm 

# Log level to be used by mod_jk 
JkLogLevel error 

# The following line mounts all JSP files and the /servlet/ uri to tomcat 
#JkMount /servlets-examples/servlet/* ajp13 
JkMount /share/*.jsp ajp13 

/IfModule 

virtualhost- 
VirtualHost *:80 
ServerName share.domain.com 

#RewriteEngine On 
#RewriteCond %{REQUEST_URI} !^/share/ 
#RewriteCond %{HTTPS} on 
#RewriteRule ^/. http://share.paradixent.com/share/ [P] 
#JkMount /share/* worker1 

IfModule mod_jk.c 


# The following line makes apache aware of the location of 
# the /jsp-examples context 
Alias /share /opt/alfresco/tomcat/webapps/share 
Directory /opt/alfresco/tomcat/webapps/share 
Options Indexes FollowSymLinks 
allow from all 
/Directory 

# The following line mounts all JSP files and the /servlet/ uri to tomcat 
#JkMount /servlets-examples/servlet/* ajp13 
JkMount /share/*.jsp ajp13 

# The following line prohibits users from directly accessing WEB-INF 
Location /share/WEB-INF/ 
#AllowOverride None 
deny from all 
/Location 

# if not specified, the global error log is used 
ErrorLog /var/log/apache2domain.com-error_log 
CustomLog /var/log/apache2/domain.com-access_log combined 

/IfModule 

/VirtualHost 

httpd.conf- 
# mod_jk 
Include /opt/alfresco/tomcat/conf/jk.conf 

Mod_jk is loaded: 
web:~ # /usr/sbin/httpd2 -M 
Loaded Modules: 
... 
jk_module (shared) 
perl_module (shared) 
php5_module (shared) 
Syntax OK 


Here is the log from apache: 
[Thu Mar 28 18:40:14 2013] [error] [client pub ip] proxy: Error reading from 
remote server returned by /error/HTTP_INTERNAL_SERVER_ERROR.html.var 
[Thu Mar 28 18:40:28 2013] [error] [client pub ip] (70007)The timeout specified 
has expired: proxy: error reading status line from remote server 
share.paradixent.com 
[Thu Mar 28 18:40:28 2013] [error] [client pub ip] proxy: Error reading from 
remote server returned by /error/HTTP_INTERNAL_SERVER_ERROR.html.var 


Any ideas why the jsp's are being listed instead of running? 


Re: [users@httpd] Followup to [Bug 50028] (LDAP authentication with encrypted passwords)

2013-03-28 Thread Eric Covener
On Thu, Mar 28, 2013 at 5:33 PM, Ken Nishimura
ken_nishim...@agilent.com wrote:
 Basically, using the mod_auth_ldap module, apart from using SSL (and
 associated overhead), is it still the case that there is no way to encrypt
 just the passing of username and password from the client (browser) back to
 the server?

 As others have pointed out, SSL is a fallback, but with associated overhead.
 Has this been fixed in later versions of Apache?

mod_authnz_ldap requires HTTP Basic Authentication, which doesn't have
any provision to encrypt the password separately from the rest of the
connection.

mod_authnz_ldap doesn't work with Digest authentication -- I don't think it can.

What does your client support that would need a fixed mod_authnz_ldap?

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



Re: [users@httpd] Followup to [Bug 50028] (LDAP authentication with encrypted passwords)

2013-03-28 Thread Ken Nishimura

Eric -

I'm not exactly sure what your last question means.  However, I think 
you answered my question.  In short, the situation has not changed.  If 
we want to ensure that the password is passed from the client (browser) 
to the server securely (to be further passed on to the LDAP server), we 
have to use SSL (https).  The path from the http server to the LDAP 
server is secure using SSL (ldaps), but from the client to the server is 
unencrypted unless the entire thing is SSL'ed.


I'm pretty new at this, but it appears that the act of popping up a 
dialog box asking for username/password cannot be encrypted separately 
from the http connection.


Thanks,

Ken
On 03/28/2013 04:11 PM, Eric Covener wrote:

On Thu, Mar 28, 2013 at 5:33 PM, Ken Nishimura
ken_nishim...@agilent.com wrote:

Basically, using the mod_auth_ldap module, apart from using SSL (and
associated overhead), is it still the case that there is no way to encrypt
just the passing of username and password from the client (browser) back to
the server?

As others have pointed out, SSL is a fallback, but with associated overhead.
Has this been fixed in later versions of Apache?

mod_authnz_ldap requires HTTP Basic Authentication, which doesn't have
any provision to encrypt the password separately from the rest of the
connection.

mod_authnz_ldap doesn't work with Digest authentication -- I don't think it can.

What does your client support that would need a fixed mod_authnz_ldap?

-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org



-
To unsubscribe, e-mail: users-unsubscr...@httpd.apache.org
For additional commands, e-mail: users-h...@httpd.apache.org