RE: MSIE *Again*

2000-07-28 Thread Burns, Robert

William,

That *DID* workdo you happen to have any explaination as to why?

It doesn't make sense that having to turn on revocation checking would allow
it to work?

Is this true for all Verisign certs?  If so, why do I not get that error
when going to other sites with a Verisign cert using IE?

- Bob

 -Original Message-
 From: Wallace, William [mailto:[EMAIL PROTECTED]]
 Sent: Thursday, July 27, 2000 10:17 AM
 To: '[EMAIL PROTECTED]'
 Subject: RE: MSIE *Again*
 
 
 Does changing the "Check for server certificate revocation (requires
 restart)" advanced security setting in IE change the behavior?
 
  -Original Message-
  From: Burns, Robert [mailto:[EMAIL PROTECTED]]
  Sent: Wednesday, July 26, 2000 10:38 AM
  To: '[EMAIL PROTECTED]'
  Subject: MSIE *Again*
  
  
  Folks,
  
  I believe I'm experiencing the same MSIE problems that
  have been discussed on this list over the past few weeks,
  but with a little more information.  Perhaps it will help.
  
  I'm running Apache 1.3.12 + modssl 2.6.4 + openssl 0.9.5a on 
  an UltraSparc 10 + Solaris7.
  
  First, I created a dummy certificate (i.e. signed by Snake-Oil CA)
  and everything works just fine.  Both IE and Netscape connect
  without incident.
  
  Next, I got a generated new keys and got a Verisign certificate.
  I installed this certificate (along with the intermediate 
 certificate)
  and that's when things started breaking for IE only.  Netscape will
  connect just fine, but IE gives that 'very informative' 
 error screen.
  
  Here is the tail end of the log with debug turned on:
  
  [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: write 67/67 bytes
  to BIO#0014D048 [mem: 001749F0] (BIO dump follows)
  +-
  +
  | : 14 03 00 00 01 01 16 03-00 00 38 7c 9b f8 cc 94  
  ..8| |
  | 0010: 73 0a b9 2b e8 ec 32 91-c2 88 86 52 2b d6 f3 12  
  s..+..2R+... |
  | 0020: 8c 67 0d 7a f9 c2 0c 1e-4c c8 6d 7a 95 3e 21 d9  
  .g.zL.mz.!. |
  | 0030: 02 16 c0 7d 94 4d 47 7d-70 49 9a 4c d6 db 82 c9  
  ...}.MG}pI.L |
  | 0040: 72 09 17 r..  
  |
  +-
  +
  [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Loop: SSLv3 flush data
  [26/Jul/2000 09:55:20 27052] [trace] Inter-Process Session Cache:
  request=SET
  status=OK
  id=460730715DA5C519241676A466979A8EC3B3813DC8A8803C81BCA4658A094BD8
  timeout=299s (session caching)
  [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Handshake: done
  [26/Jul/2000 09:55:20 27052] [info]  Connection: Client IP: 
  192.168.8.109,
  Protocol: SSLv3, Cipher: RC4-MD5 (128/128 bits)
  [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: read 0/18437 
 bytes from
  BIO#0014D048
  [mem: 001675C8] (BIO dump follows)
  +-
  +
  +-
  +
  [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: write 23/23 bytes to
  BIO#0014D048
  [mem: 0016FDD8] (BIO dump follows)
  +-
  +
  | : 15 03 00 00 12 d4 c5 65-6a a4 01 3f bd 11 49 75  
  ...ej..?..Iu |
  | 0010: 12 43 94 83 8f 2c a5 
  .C...,.  |
  +-
  +
  [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Write: SSL negotiation
  finished
  successfully
  [26/Jul/2000 09:55:20 27052] [info]  Connection to child 1 
 closed with
  standard
  shutdown (server 192.168.8.84:443, client 192.168.8.109)
  
  It appears that in the line above (read 0/18437 bytes 
 from...) that IE
  shutdown the TCP/IP connection, forcing the SSL connection to 
  be closed by
  the server.  The question is, why does IE shutdown the 
 connection, but
  Netscape continued on without problem?
  
  I'm going to try to sniff the TCP line to see what is 
  actually happening,
  but until then, any additional insight would be helpfull.
  
  Thanks,
  
  - Bob
  
  --
  Bob BurnsZaxus
  [EMAIL PROTECTED]   1-888-744-4976, X6510
  (local) 1-954-846-6510
  -- 
  
 __
  Apache Interface to OpenSSL (mod_ssl)   
www.modssl.org
 User Support Mailing List  [EMAIL PROTECTED]
 Automated List Manager[EMAIL PROTECTED]
 
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]

Running Mod_ssl when starting apache automatically.

2000-07-28 Thread Gene Lamoreaux

I found a couple of sections in the archives stating that the way to do this
was to remove the passphrase section of the key.

Is there a way to automatically start the ./apachectl startssl and have it
pickup the passphrase from a file or script or something. I have so far had
no luck trying either of these approaches as it seems to require keyboard
input to take the passphrase.

Any and all input is greatly appreciated.

Gene Lamoreaux
Systems Integration Engineer
Syndeo Corporation
Office:(408) 861-1000 Ext.5002
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Problem with Failed to generate temporary 512 bit RSA private key

2000-07-28 Thread Simon Dubey
Hello

I have just installed mod-ssl on a solaris /sparc machine and get the
above error.

I have read the FAQ and tried to following what it is suggesting with
$HOME/.rnd but do not quite follow it - well what I did, did not work.

I have also tried truerand as well but that did not work either.

Please advise

Simon.

__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]


Re: apache+modssl+... got SEGVs

2000-07-28 Thread David Rees

On Wed, Jul 26, 2000 at 11:17:14AM +0200, Hugues Pisapia wrote:

 And sometimes (well, more often with MSIE than with Mozilla :), apache gets
 Segmentation faults. It seems that it comes from openssl or modssl as i tried
 many configurations. Apache gets SEGVs only when the virtual host with modssl 
 is running. I tried many thing to figure out where it comes from, but i have no
 more idea... Could someone help me, at least to give me a clue ?
 
 Last things : i tried the solution in the FAQ, i.e. to change the
 SSLSessionCache directive arguments, but i'm running apache from a rpm, so i
 don't have mm support, and my project manager will kick me if i say that i
 have to recompile apache :/, so i'd like to avoid that.

No one has had good luck when using RPMs to install mod_ssl.  I'm afraid that
recompiling Apache from scratch is the way to go.

Why will your project manager kick you for recompiling Apache?  You'll only
be down for a second or two when you install the newly compiled Apache.

-Dave
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: starting ssl

2000-07-28 Thread David Rees

On Fri, Jul 28, 2000 at 01:27:32PM +0800, Raymond wrote:
 thanks for the reply david but that did not work also. any other suggestions.

Can you make your config file available somewhere so we can take a look at
it?  I've most often found that config file problems are the most frequent
cause of your problem.

Any reason you're not using the apachectl script to start/stop Apache?

-Dave
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



FW: Re: verisign certificate installation

2000-07-28 Thread Myint Myint Moe

Hi, 

I've clicked "Reply via email to Victor"  but unfortunately mail bounced
back .

Thanks and Regards,
Myint Myint Moe

  -Original Message-
 From: Myint Myint Moe  
 Sent: Friday, July 28, 2000 6:42 PM
 To:   '[EMAIL PROTECTED]'
 Subject:  Re: verisign certificate installation
 
 Hi,
 
 Thanks for your reply. In Apache-modssl, httpd.conf also have the
 following directive lines and default is making comment that lines. 
 
 #   Client Authentication (Type):
 #   Client certificate verification type and depth.  Types are
 #   none, optional, require and optional_no_ca.  Depth is a
 #   number which specifies how deeply to verify the certificate
 #   issuer chain before deciding the certificate is not valid.
 #SSLVerifyClient require
 #SSLVerifyDepth  10
 
 Shall I remark-out and set
 SSLVerifyClient none
 SSLVerifyDepth  10
 
 [none: no client Certificate is required at all optional: the client
 may present a valid Certificate require: the client has to present a
 valid Certificate optional_no_ca: the client may present a valid
 Certificate but it need not to be (successfully) verifiable]
 Please advise me and also other directive variable need to set (e.g
 SSLCACertificatePath and etc).Currently all lines with comments except
 SSLCertificateFile  SSLCertificatekeyFile directive lines. 
 (I'm quite new in SSL setting and server cannot be start and stop
 frequently. I'm waiting your favorable reply.) 
 
 Thanks in advance.
 
 
 Regards,
 Myint Myint Moe
 
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



FW: Re: verisign certificate installation

2000-07-28 Thread Myint Myint Moe

Hi, 

I've clicked "Reply via email to Victor"  but unfortunately mail bounced
back .

Thanks and Regards,
Myint Myint Moe

  -Original Message-
 From: Myint Myint Moe  
 Sent: Friday, July 28, 2000 6:42 PM
 To:   '[EMAIL PROTECTED]'
 Subject:  Re: verisign certificate installation
 
 Hi,
 
 Thanks for your reply. In Apache-modssl, httpd.conf also have the
 following directive lines and default is making comment that lines. 
 
 #   Client Authentication (Type):
 #   Client certificate verification type and depth.  Types are
 #   none, optional, require and optional_no_ca.  Depth is a
 #   number which specifies how deeply to verify the certificate
 #   issuer chain before deciding the certificate is not valid.
 #SSLVerifyClient require
 #SSLVerifyDepth  10
 
 Shall I remark-out and set
 SSLVerifyClient none
 SSLVerifyDepth  10
 
 [none: no client Certificate is required at all optional: the client
 may present a valid Certificate require: the client has to present a
 valid Certificate optional_no_ca: the client may present a valid
 Certificate but it need not to be (successfully) verifiable]
 Please advise me and also other directive variable need to set (e.g
 SSLCACertificatePath and etc).Currently all lines with comments except
 SSLCertificateFile  SSLCertificatekeyFile directive lines. 
 (I'm quite new in SSL setting and server cannot be start and stop
 frequently. I'm waiting your favorable reply.) 
 
 Thanks in advance.
 
 
 Regards,
 Myint Myint Moe
 
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



[BugDB] compiling mod_ssl without EAPI (PR#435)

2000-07-28 Thread modssl-bugdb

Full_Name: Dimulka
Version: 0.9.5a
OS: Solaris
Submission from: (NULL) (193.232.88.16)


Hi all!
I need to compile apache with mod_ssl with plain API but after configurin'
mod_ssl EAPI is enabled even if I use '--disable-rule=EAPI' option. And after
runnin' apache I have got an error message like this: "[Fri Jul 28 16:21:53
2000] [warn] Loaded DSO some_module.so uses plain Apache 1.3 API, this module
might crash under EAPI! (please recompile it with -DEAPI)"
What have I to do to avoid this warning?

__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



RE: MSIE *Again*

2000-07-28 Thread Wallace, William

Sorry Robert,

I don't have any explaination. I discovered the same problem mid-June and
have only just got around to investigating it.

I've done the same SSL log analysis as you and a packet trace as well. At
the packet level what happens is as soon as the handshake completes IE
closes the connections (it sends a FIN).

It seems to only happen with the X509 v3 certificates from Verisign so
perhaps it's something to do with the x509 version or the fact that their v3
certificates have an additional certificate in the chain. I've seen similar
certificates work though with IE (but a different web server).


On a somewhat wierd note, we both have famous Scottish names!


 -Original Message-
 From: Burns, Robert [mailto:[EMAIL PROTECTED]]
 Sent: Thursday, July 27, 2000 10:33 AM
 To: '[EMAIL PROTECTED]'
 Subject: RE: MSIE *Again*
 
 
 William,
 
 That *DID* workdo you happen to have any explaination as to why?
 
 It doesn't make sense that having to turn on revocation 
 checking would allow
 it to work?
 
 Is this true for all Verisign certs?  If so, why do I not get 
 that error
 when going to other sites with a Verisign cert using IE?
 
 - Bob
 
  -Original Message-
  From: Wallace, William [mailto:[EMAIL PROTECTED]]
  Sent: Thursday, July 27, 2000 10:17 AM
  To: '[EMAIL PROTECTED]'
  Subject: RE: MSIE *Again*
  
  
  Does changing the "Check for server certificate revocation (requires
  restart)" advanced security setting in IE change the behavior?
  
   -Original Message-
   From: Burns, Robert [mailto:[EMAIL PROTECTED]]
   Sent: Wednesday, July 26, 2000 10:38 AM
   To: '[EMAIL PROTECTED]'
   Subject: MSIE *Again*
   
   
   Folks,
   
   I believe I'm experiencing the same MSIE problems that
   have been discussed on this list over the past few weeks,
   but with a little more information.  Perhaps it will help.
   
   I'm running Apache 1.3.12 + modssl 2.6.4 + openssl 0.9.5a on 
   an UltraSparc 10 + Solaris7.
   
   First, I created a dummy certificate (i.e. signed by Snake-Oil CA)
   and everything works just fine.  Both IE and Netscape connect
   without incident.
   
   Next, I got a generated new keys and got a Verisign certificate.
   I installed this certificate (along with the intermediate 
  certificate)
   and that's when things started breaking for IE only.  
 Netscape will
   connect just fine, but IE gives that 'very informative' 
  error screen.
   
   Here is the tail end of the log with debug turned on:
   
   [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: write 67/67 bytes
   to BIO#0014D048 [mem: 001749F0] (BIO dump follows)
   +-
   +
   | : 14 03 00 00 01 01 16 03-00 00 38 7c 9b f8 cc 94  
   ..8| |
   | 0010: 73 0a b9 2b e8 ec 32 91-c2 88 86 52 2b d6 f3 12  
   s..+..2R+... |
   | 0020: 8c 67 0d 7a f9 c2 0c 1e-4c c8 6d 7a 95 3e 21 d9  
   .g.zL.mz.!. |
   | 0030: 02 16 c0 7d 94 4d 47 7d-70 49 9a 4c d6 db 82 c9  
   ...}.MG}pI.L |
   | 0040: 72 09 17 r..  
   |
   +-
   +
   [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Loop: SSLv3 
 flush data
   [26/Jul/2000 09:55:20 27052] [trace] Inter-Process Session Cache:
   request=SET
   status=OK
   
 id=460730715DA5C519241676A466979A8EC3B3813DC8A8803C81BCA4658A094BD8
   timeout=299s (session caching)
   [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Handshake: done
   [26/Jul/2000 09:55:20 27052] [info]  Connection: Client IP: 
   192.168.8.109,
   Protocol: SSLv3, Cipher: RC4-MD5 (128/128 bits)
   [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: read 0/18437 
  bytes from
   BIO#0014D048
   [mem: 001675C8] (BIO dump follows)
   +-
   +
   +-
   +
   [26/Jul/2000 09:55:20 27052] [debug] OpenSSL: write 23/23 bytes to
   BIO#0014D048
   [mem: 0016FDD8] (BIO dump follows)
   +-
   +
   | : 15 03 00 00 12 d4 c5 65-6a a4 01 3f bd 11 49 75  
   ...ej..?..Iu |
   | 0010: 12 43 94 83 8f 2c a5 
   .C...,.  |
   +-
   +
   [26/Jul/2000 09:55:20 27052] [trace] OpenSSL: Write: SSL 
 negotiation
   finished
   successfully
   [26/Jul/2000 09:55:20 27052] [info]  Connection to child 1 
  closed with
   standard
   shutdown (server 192.168.8.84:443, client 192.168.8.109)
   
   It appears that in the line above (read 0/18437 bytes 
  from...) that IE
   shutdown the TCP/IP connection, forcing the SSL connection to 
   be closed by
   the server.  The question is, why does IE shutdown the 
  connection, but
   Netscape continued on without problem?
   
   I'm going to try to sniff the 

Re: : apache+modssl+... got SEGVs

2000-07-28 Thread Magnus Stenman


On Fri, Jul 28, 2000 at 02:07:51AM -0700, David Rees wrote:
 On Wed, Jul 26, 2000 at 11:17:14AM +0200, Hugues Pisapia wrote:
 
  And sometimes (well, more often with MSIE than with Mozilla :), apache gets
  Segmentation faults. It seems that it comes from openssl or modssl as i tried
  many configurations. Apache gets SEGVs only when the virtual host with modssl 
  is running. I tried many thing to figure out where it comes from, but i have no
  more idea... Could someone help me, at least to give me a clue ?
  
  Last things : i tried the solution in the FAQ, i.e. to change the
  SSLSessionCache directive arguments, but i'm running apache from a rpm, so i
  don't have mm support, and my project manager will kick me if i say that i
  have to recompile apache :/, so i'd like to avoid that.


The RPMs at modssl.org have mm compiled in.


 
 No one has had good luck when using RPMs to install mod_ssl.  I'm afraid that
 recompiling Apache from scratch is the way to go.


What's the problem with the RPMs?

I roll them, and if you got a RPM specific porblem,
I'll be happy to look at it.


 
 Why will your project manager kick you for recompiling Apache?  You'll only
 be down for a second or two when you install the newly compiled Apache.

Maintenence? It's easier with RPMs...


/magnus


 
 -Dave
 __
 Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
 User Support Mailing List  [EMAIL PROTECTED]
 Automated List Manager[EMAIL PROTECTED]

-- 
 Magnus Stenman   mailto:[EMAIL PROTECTED]   http://www.hkust.se

 Get it up, keep it up.  Linux -- Viagra for your PC
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: : apache+modssl+... got SEGVs

2000-07-28 Thread David Rees

  No one has had good luck when using RPMs to install mod_ssl.  I'm afraid that
  recompiling Apache from scratch is the way to go.
 
 What's the problem with the RPMs?
 
 I roll them, and if you got a RPM specific porblem,
 I'll be happy to look at it.

Maybe if everyone used the RPMs you rolled there wouldn't be any problems,
but it seems that people want to take one RPM from every site, throw them
all together, and hope they work.  Then they don't because one package
wasn't compiled with the right options (like EAPI) There just seems to be
too many variables if you want to use RPMs to build Apache/mod_ssl,
especially since a lot of people would also like to use RPMs to install
other Apache modules I'm sure (php, mod_perl).  It also seems that a lot
of the people that come to the list reporting Apache/mod_ssl crashes are
using RPMs.

  Why will your project manager kick you for recompiling Apache?  You'll only
  be down for a second or two when you install the newly compiled Apache.
 
 Maintenence? It's easier with RPMs...

Can't argue with you there.

-Dave
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Frames Javascript - insecure?

2000-07-28 Thread Lee Feigenbaum

Hi Everyone,

I've searched the archives and FAQ for the answer to this, to no avail.

I have a page on my site that has one frameset with two frames, one of
which is a frameset with two more frames (yah, i know, frames are evil
yadda yadda). 

The outermost frameset is called with a secure URL, and all of the frame
src="..." tags use only relative URIs. Nevertheless, when viewed on IE
the page gives the warning "this page contains both secure and insecure
elements" - it's not an IE specific bug, as Netscape's security summary
says the same thing - listing the individual page elements in Netscape
shows that it thinks that the html pages, all of which it correctly lists
as https://whatever have:

Security: Status unknown

One of the frames does use Javascript (ya, i know, javascript is as evil
as frames, yadda yadda) extensively to write to another frame; I don't
know if it's the frames or the Javascript (or something completely
different?) that's causing this problem... which is why I'm coming here
for advice :)

TIA,
Lee


__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: Frames Javascript - insecure?

2000-07-28 Thread Cliff Woolley



 [EMAIL PROTECTED] 07/28/00 12:12PM The 
outermost frameset is called with a secure URL, and all of the 
framesrc="..." tags use only relative URIs. Nevertheless, when 
viewed on IEthe page gives the warning "this page contains both secure 
and insecureelements" - it's not an IE specific bug, as Netscape's 
security summarysays the same thing - listing the individual page 
elements in Netscapeshows that it thinks that the html pages, all of 
which it correctly listsas https://whatever have:
You get this if ANY of the elements of the page (eg, images) are delivered 
via an insecure method (http://, file://, etc). Make sure that*all* 
images are delivered via HTTPS along with the HTML documents, and the error 
should go away. The javascript ought not matter.

Hope this helps,

--Cliff

Cliff WoolleyCentral Systems Software AdministratorWashington and 
Lee Universityhttp://www.wlu.edu/~jwoolley/

Work: (540) 463-8089Pager: (540) 462-2303


Re: Frames Javascript - insecure?

2000-07-28 Thread Lee Feigenbaum

I did check everything, and everything is being served via HTTPS
(everything is specified relative to the main frameset pages, which are
all https). In fact, Netscape lists the individual images as being secure,
but all of the .html components of the page are listed as 'Status
unknown'.

Thanks for the suggestion, though,
Lee

On Fri, 28 Jul 2000, Cliff Woolley wrote:

  [EMAIL PROTECTED] 07/28/00 12:12PM 
 The outermost frameset is called with a secure URL, and all of the
 frame
 src="..." tags use only relative URIs. Nevertheless, when viewed on IE
 the page gives the warning "this page contains both secure and insecure
 elements" - it's not an IE specific bug, as Netscape's security summary
 says the same thing - listing the individual page elements in Netscape
 shows that it thinks that the html pages, all of which it correctly
 lists
 as https://whatever have:
 
 You get this if ANY of the elements of the page (eg, images) are
 delivered via an insecure method (http://, file://, etc).  Make sure
 that *all* images are delivered via HTTPS along with the HTML documents,
 and the error should go away.  The javascript ought not matter.
 
 Hope this helps,
 
 --Cliff
 
 Cliff Woolley
 Central Systems Software Administrator
 Washington and Lee University
 http://www.wlu.edu/~jwoolley/
 
 Work: (540) 463-8089
 Pager: (540) 462-2303
 

__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: Frames Javascript - insecure?

2000-07-28 Thread Doug Poulin

There are 2 things you can do.

1. Go into the Internet Browser properties (advanced)  Remove the check box
on "Warn if changing between secure and not secure mode.

2. We also had to change all of our URI's to use full pathnames including
the https...  This was especially the case with forms and cgi scripts.

I hope this helps.

Doug Poulin

- Original Message -
From: Lee Feigenbaum [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Friday, July 28, 2000 11:25 AM
Subject: Re: Frames  Javascript - insecure?


 I did check everything, and everything is being served via HTTPS
 (everything is specified relative to the main frameset pages, which are
 all https). In fact, Netscape lists the individual images as being secure,
 but all of the .html components of the page are listed as 'Status
 unknown'.

 Thanks for the suggestion, though,
 Lee

 On Fri, 28 Jul 2000, Cliff Woolley wrote:

   [EMAIL PROTECTED] 07/28/00 12:12PM 
  The outermost frameset is called with a secure URL, and all of the
  frame
  src="..." tags use only relative URIs. Nevertheless, when viewed on IE
  the page gives the warning "this page contains both secure and insecure
  elements" - it's not an IE specific bug, as Netscape's security summary
  says the same thing - listing the individual page elements in Netscape
  shows that it thinks that the html pages, all of which it correctly
  lists
  as https://whatever have:
 
  You get this if ANY of the elements of the page (eg, images) are
  delivered via an insecure method (http://, file://, etc).  Make sure
  that *all* images are delivered via HTTPS along with the HTML documents,
  and the error should go away.  The javascript ought not matter.
 
  Hope this helps,
 
  --Cliff
 
  Cliff Woolley
  Central Systems Software Administrator
  Washington and Lee University
  http://www.wlu.edu/~jwoolley/
 
  Work: (540) 463-8089
  Pager: (540) 462-2303
 

 __
 Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
 User Support Mailing List  [EMAIL PROTECTED]
 Automated List Manager[EMAIL PROTECTED]

__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: : : apache+modssl+... got SEGVs

2000-07-28 Thread Magnus Stenman

On Fri, Jul 28, 2000 at 09:03:57AM -0700, David Rees wrote:
   No one has had good luck when using RPMs to install mod_ssl.  I'm afraid that
   recompiling Apache from scratch is the way to go.
  
  What's the problem with the RPMs?
  
  I roll them, and if you got a RPM specific porblem,
  I'll be happy to look at it.
 
 Maybe if everyone used the RPMs you rolled there wouldn't be any problems,
 but it seems that people want to take one RPM from every site, throw them
 all together, and hope they work.  Then they don't because one package
 wasn't compiled with the right options (like EAPI) There just seems to be
 too many variables if you want to use RPMs to build Apache/mod_ssl,
 especially since a lot of people would also like to use RPMs to install
 other Apache modules I'm sure (php, mod_perl).  It also seems that a lot

Yeah, mixing Apache RPMs is evil, the dependency facilities in RPM
is just not enough to achieve full modularity in apache-mod_ssl

There should probably be a FAQ just for doing this...

And some things just don't run reliably as a loadable module
no matter what you do... mod_perl for example...


But I hear that RedHat 6.2 work fairly well, I haven't confirmed,
though.


 of the people that come to the list reporting Apache/mod_ssl crashes are
 using RPMs.


Yeah, and often mixing non-EAPI/EAPI stuff.

Too bad the apache group didn't include EAPI, but started to discuss
how to do it better, ending up with nothing...


/m


 
   Why will your project manager kick you for recompiling Apache?  You'll only
   be down for a second or two when you install the newly compiled Apache.
  
  Maintenence? It's easier with RPMs...
 
 Can't argue with you there.
 
 -Dave
 __
 Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
 User Support Mailing List  [EMAIL PROTECTED]
 Automated List Manager[EMAIL PROTECTED]

-- 
 Magnus Stenman   mailto:[EMAIL PROTECTED]   http://www.hkust.se

 Get it up, keep it up.  Linux -- Viagra for your PC
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



URG: I've replaced the cert, and I broke it!

2000-07-28 Thread Mark Drummond

Ack! Cert has expired so we got a new one from Verisign. I figured I was
just replacing the old one (just did the same, successfully, on another
server elsewhere) but I must have screwed something up because when I
try to start the server up with the new cert i get:

24381:error:0906406D:PEM routines:DEF_CALLBACK:problems getting
password:/usr/local/covalent/src/SSLeay/crypto/pem/pem_lib.c:110:
24381:error:0906A068:PEM routines:PEM_do_header:bad password
read:/usr/local/covalent/src/SSLeay/crypto/pem/pem_lib.c:387:

Now I have picked up this server from another SA that flew the coop(sp)
so I am feeling a bit lost here. I have a directory structure that looks
like this:

/usr/local/ssl
 |
 |- bin
 |- include
 |- lib
 |- certs
 |- private

This is from when openssl was just ssleay. Apparently, certificates are
in the cert subdir and the private subdir _seems_ to be private keys.
And yet, when I look at the old certs in the cert subdir they have both
a "RSA Private key" component and a "Certificate" component. However, I
seem some private keys that start like this:

-BEGIN RSA PRIVATE KEY-

and others that begin like this:

-BEGIN RSA PRIVATE KEY-
Proc-Type: 4,ENCRYPTED
DEK-Info: DES-EDE3-CBC,01339285FAE6F39B

What's the diff?

So anyway, I generated a CSR, gave that to the person with the credit
card who purchased the cert. They gave me the cert and at that point I
was unsure of what to do. I'd swear that on that other server all I did
was replaced the "Certificate" portion of the file with the new cert and
now it is working fine. So I did the same here only I get the error
above when I try to start the server.

Any ideas?
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: URG: I've replaced the cert, and I broke it!

2000-07-28 Thread Mark Drummond

Mark Drummond wrote:
 
 Ack! Cert has expired so we got a new one from Verisign. I figured I was
 just replacing the old one (just did the same, successfully, on another
 server elsewhere) but I must have screwed something up because when I
 try to start the server up with the new cert i get:
 
 24381:error:0906406D:PEM routines:DEF_CALLBACK:problems getting
 password:/usr/local/covalent/src/SSLeay/crypto/pem/pem_lib.c:110:
 24381:error:0906A068:PEM routines:PEM_do_header:bad password
 read:/usr/local/covalent/src/SSLeay/crypto/pem/pem_lib.c:387:

I just realised .. when I start up the server, should it not ask for the
passphrase? It doesn't.
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: URG: I've replaced the cert, and I broke it!

2000-07-28 Thread Mark Drummond

Mark Drummond wrote:
 

Hello?
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]



Re: URG: I've replaced the cert, and I broke it!

2000-07-28 Thread Mark Drummond

Mark Drummond wrote:
 
 Ack! Cert has expired so we got a new one from Verisign. I figured I was

Forget it. I managed to figure it out myself. I needed to "ssleay rsa 
pem  key" my key file. Um, what exactly does that do?
__
Apache Interface to OpenSSL (mod_ssl)   www.modssl.org
User Support Mailing List  [EMAIL PROTECTED]
Automated List Manager[EMAIL PROTECTED]