You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Philip Martin <ph...@codematters.co.uk> on 2005/08/12 11:26:57 UTC

Re: "svn lock" crash against certain 1.1.x servers

Justin Erenkrantz <ju...@erenkrantz.com> writes:

> On Fri, Aug 12, 2005 at 05:18:04AM +0200, Branko ?ibej wrote:
>> I don't think we can do anything about the repeated authn prompts -- 
>> IMHO they're caused by a server bug. I verified that this doesn't happen 
>> with httpd-2.0.54/svn-1.1.1 on Windows, and IIRC Philip verified 
>> httpd-2.0.54/svn-1.1.4 on *nix.
>
> Nothing particularly jumps out at me that would have changed between httpd
> 2.0.50 and 2.0.54 that would affect this.  -- justin

It's possible the different behaviour is due to a different auth setup
between the tsvn server and my server.  I used httpd to require
read/write auth for all access, tsvn uses mod_auth_svn and probably
allows read-only access as well as read/write access.

-- 
Philip Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: "svn lock" crash against certain 1.1.x servers

Posted by Branko Čibej <br...@xbc.nu>.
Ben Collins-Sussman wrote:

>
> On Aug 12, 2005, at 7:15 AM, Branko Čibej wrote:
>
>>>
>>> It's possible the different behaviour is due to a different auth  setup
>>> between the tsvn server and my server.  I used httpd to require
>>> read/write auth for all access, tsvn uses mod_auth_svn and probably
>>> allows read-only access as well as read/write access.
>>>
>>>
>> D'you know, I think you're right! The damn mod_authz_svn takes it  
>> upon itself to return HTTP_UNAUTHORIZED in the auth_checker hook,  
>> even though it's thecking authorization, not authentication. I  think 
>> this is a bug in mod_authz_svn. It should be returning  
>> HTTP_FORBIDDEN (as in the access_checker hook), and probably  
>> shouldn't provide an auth checker at all.
>
>
>
> CEE (tigris.org) doesn't use mod_authz_svn, but a proprietary module  
> called mod_auth_svn.  The name is deliberately ambiguous, because it  
> registers itself as both an authn and authz hook with apache.  It  
> performs authn by checking the basic-auth-creds against a central CEE  
> database, and performs authz by checking paths against another  
> central CEE database.

Indeed, I was a bit quick off the mark here. I tested with a local 
svn-1.1.1 over http with mod_authz_svn configured, and I still get a 
405, not a 401. So it looks like the bug is in that mod_auth_svn?

Anyway, I've decided to commit Philip's patch, which avoids all auth 
caching in this situation.

-- Brane


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: "svn lock" crash against certain 1.1.x servers

Posted by Ben Collins-Sussman <su...@collab.net>.
On Aug 12, 2005, at 7:15 AM, Branko Čibej wrote:

>>
>> It's possible the different behaviour is due to a different auth  
>> setup
>> between the tsvn server and my server.  I used httpd to require
>> read/write auth for all access, tsvn uses mod_auth_svn and probably
>> allows read-only access as well as read/write access.
>>
>>
> D'you know, I think you're right! The damn mod_authz_svn takes it  
> upon itself to return HTTP_UNAUTHORIZED in the auth_checker hook,  
> even though it's thecking authorization, not authentication. I  
> think this is a bug in mod_authz_svn. It should be returning  
> HTTP_FORBIDDEN (as in the access_checker hook), and probably  
> shouldn't provide an auth checker at all.


CEE (tigris.org) doesn't use mod_authz_svn, but a proprietary module  
called mod_auth_svn.  The name is deliberately ambiguous, because it  
registers itself as both an authn and authz hook with apache.  It  
performs authn by checking the basic-auth-creds against a central CEE  
database, and performs authz by checking paths against another  
central CEE database.



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org


Re: "svn lock" crash against certain 1.1.x servers

Posted by Branko Čibej <br...@xbc.nu>.
Philip Martin wrote:

>Justin Erenkrantz <ju...@erenkrantz.com> writes:
>
>  
>
>>On Fri, Aug 12, 2005 at 05:18:04AM +0200, Branko ?ibej wrote:
>>    
>>
>>>I don't think we can do anything about the repeated authn prompts -- 
>>>IMHO they're caused by a server bug. I verified that this doesn't happen 
>>>with httpd-2.0.54/svn-1.1.1 on Windows, and IIRC Philip verified 
>>>httpd-2.0.54/svn-1.1.4 on *nix.
>>>      
>>>
>>Nothing particularly jumps out at me that would have changed between httpd
>>2.0.50 and 2.0.54 that would affect this.  -- justin
>>    
>>
>
>It's possible the different behaviour is due to a different auth setup
>between the tsvn server and my server.  I used httpd to require
>read/write auth for all access, tsvn uses mod_auth_svn and probably
>allows read-only access as well as read/write access.
>  
>
D'you know, I think you're right! The damn mod_authz_svn takes it upon 
itself to return HTTP_UNAUTHORIZED in the auth_checker hook, even though 
it's thecking authorization, not authentication. I think this is a bug 
in mod_authz_svn. It should be returning HTTP_FORBIDDEN (as in the 
access_checker hook), and probably shouldn't provide an auth checker at all.

-- Brane


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org