You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@directory.apache.org by Markus Pohle <ap...@webunity.de> on 2007/07/26 12:22:59 UTC

Re: [TwixEncoder] :: delivers different PDUs for same search result on different operating systems

Hi Emmanuel,

first of all thx for your reply!

I did all my tests with the same apacheds-1.5.1-snapshot and testet  
ads first with 1.5.0_10 jdk and then did the test again with ads  
running with 1.6.0 jdk to be sure that there is no influence caused by  
the jdk.

What I can say is, from within my tests, that the jdk version seems  
not to be relevant.

What is relevant, from within my tests, seems to be the host operating  
system on which ads-1.5.1-snapshot is running.

Here is, in your reply to me, a little mistake:
On all unix-like operating systems I tested (with jdk 5/6) there are  
never whitespace in the returning DN. But on the windows based  
operating systems I tested there are whitespace in the returning DN.

So the operating system on which ADS runs seems to have influence on  
what the returning PDU is.

My question is: is this a bug in ADS? What can make ADS deliver  
different PDU for same ldap request? What can I do, except for your  
hint to use JNDI methods to do the comparison?

TIA
Markus


Zitat von Emmanuel Lecharny <el...@gmail.com>:

> Hi Markus,
>
> when I analyze both dump, what I can say is that the PDU lengths are
> different. The first one (windows) has this DN :
> uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING
> The second one, Cent OS, contains spaces :
> uid=Jedel, cn=users, dc=VERWALMUNG, dc=DOUGLASHOLDING
>
> This is why you have different lengths (3 bytes, as we have three spaces).
>
> As you may use different JVMs on those two servers, the attributes
> order may vary too. Which JVM are you using ?
>
> Here are the decoded messages :
> on windows :
> 0x30 0x81 0xAD
>  0x02 0x01 0x02
>  0x64 0x81 0xA7
>    0x04 0x32
>      uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING
>    0x30 0x71
>      0x30 0x0E
>        0x04 0x03
>          uid
>        0x31 0x07
>          0x04 0x05
>            Jeder
>      0x30 0x0D
>        0x04 0x02
>          sn
>        0x31 0x07
>          0x04 0x05
>            Jeder
>      0x30 0x0D
>        0x04 0x02
>          cn
>        0x31 0x07
>          0x04 0x05
>            Jeder
>      0x30 0x41
>        0x04 0x0B
>          objectClass
>        0x31 0x32
>          0x04 0x14
>            organizationalPerson
>          0x04 0x06
>            person
>          0x04 0x0D
>            inetOrgPerson
>          0x04 0x03
>            top
>
> on CentOS :
> 0x30 0x81 0xB0
>  0x02 0x01 0x02
>  0x64 0x81 0xAA
>    0x04 0x35
>      uid=Jeder, cn=users, dc=VERWALMUNG, dc=DOUGLASHOLDING
>    0x30 0x71
>      0x30 0x0D
>        0x04 0x02
>          sn
>        0x31 0x07
>          0x04 0x05
>            Jeder
>      0x30 0x41
>        0x04 0x0B
>          objectClass
>        0x31 0x32
>          0x04 0x06
>            Person
>          0x04 0x14
>            organizationalPerson
>          0x04 0x0D
>            inetOrgPerson
>          0x04 0x03
>            top
>      0x30 0x0D
>        0x04 0x02
>          cn
>        0x31 0x07
>          0x04 0x05
>            Jeder
>      0x30 0x0E
>        0x04 0x03
>          uid
>        0x31 0x07
>          0x04 0x05
>            Jeder
>
> One hypothesis is that you injected the last DN with spaces, which
> should not be a pb at all, as spaces around ',' are to be ignored. Of
> course, if you simply compare the resulting DN as strings, then you
> will have a pb. You should use some of the JNDI methods to do those
> comparisons (Name class)
>
> Hope it helps
>
> On 7/25/07, Markus Pohle <ap...@webunity.de> wrote:
>> Hi list-users,
>>
>> I do have problem with my content management system working with
>> apacheds on a windows 2003 server installation.
>>
>> My content management system (coremmedia 2005) tries to identify each
>> user by ldap request against apacheds 1.5.x ldap server.
>>
>> What I found out is, that the ldap request against an apacheds-1.5.x on
>> a windows 2003 server system isnt successful, while the same request
>> against a centos 4.3 or fedora 7 apacheds-1.5.x installation is successful.
>>
>> How did I find that out? Well, I configured my CMS to connect against
>> apacheds-1.5.x on a windows 2003 server, where apacheds logs are in
>> debug mode. Then I tried to load CMS startpage, which is not successful.
>> Then I did the same with CMS connected against apacheds-1.5.x running on
>> centos 4.3 with logfiles in debug mode. Now the startpage was loaded.
>>
>> I then searched thru the logfiles and found out the following:
>>
>> - the search request from CMS is always the following:
>>
>> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixDecoder] -
>> Decoded LdapMessage : LdapMessage
>>     message Id : 2
>>     Search Request
>>         Base Object : 'cn=users,cn=cms,dc=APPLICATIONS,dc=DOUGLASHOLDING'
>>         Scope : whole subtree
>>         Deref Aliases : deref Always
>>         Size Limit : no limit
>>         Time Limit : no limit
>>         Types Only : false
>>         Filter : '(&(objectClass=inetOrgPerson)(uid=Jeder))'
>>         Attributes : uid, mail, sn, cn, givenname, objectClass
>>     Control
>>         Control type : '2.16.840.1.113730.3.4.2'
>>         Criticality : 'false'
>>
>> [19:48:34] DEBUG
>> [org.apache.directory.shared.ldap.codec.TwixTransformer] - Transforming
>> LdapMessage <2, SEARCH_REQUEST> from Twix to Snickers.
>> [19:48:34] DEBUG
>> [org.apache.directory.server.ldap.support.SearchHandler] - Message
>> received:      SearchRequest
>>         baseDn : 'cn=users,cn=cms,dc=APPLICATIONS,dc=DOUGLASHOLDING'
>>         filter : '(& (objectClass=inetOrgPerson) (uid=Jeder) ) '
>>         scope : whole subtree
>>         typesOnly : false
>> no limit
>>         Time Limit : no limit
>>         Deref Aliases : deref Always
>>         attributes : 'uid', 'mail', 'sn', 'cn', 'givenname', 'objectClass'
>>
>> - the search result is the same on both systems:
>>
>> [19:48:34] DEBUG
>> [org.apache.directory.shared.ldap.codec.TwixTransformer] - Transformed
>> message : LdapMessage
>>     message Id : 2
>>     Search Result Entry
>>         Object Name :
>> '0.9.2342.19200300.100.1.1=jeder,2.5.4.3=users,0.9.2342.19200300.100.1.25=verwaltung,0.9.2342.19200300.100.1.25=douglasholding'
>>         Attributes
>>             Attributes
>>             Attribute id : 'uid',  Values : ['Jeder']
>>             Attribute id : 'sn',  Values : ['Jeder']
>>             Attribute id : 'cn',  Values : ['Jeder']
>>             Attribute id : 'objectClass',  Values :
>> ['organizationalPerson', 'person', 'inetOrgPerson', 'top']
>>
>> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
>> Encoding this LdapMessage : LdapMessage
>>     message Id : 2
>>     Search Result Entry
>>         Object Name :
>> '0.9.2342.19200300.100.1.1=jeder,2.5.4.3=users,0.9.2342.19200300.100.1.25=verwaltung,0.9.2342.19200300.100.1.25=douglasholding'
>>         Attributes
>>             Attributes
>>             Attribute id : 'uid',  Values : ['Jeder']
>>             Attribute id : 'sn',  Values : ['Jeder']
>>             Attribute id : 'cn',  Values : ['Jeder']
>>             Attribute id : 'objectClass',  Values :
>> ['organizationalPerson', 'person', 'inetOrgPerson', 'top']
>>
>> But the next line in debug log which contains the Encoded PDU differs
>> from windows 2003 server to centos43.
>>
>> The windows 2003 encoded PDU looks like this:
>>
>> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
>> Encoded PDU : 0x30 0x81 0xAD 0x02 0x01 0x02 0x64 0x81 0xA7 0x04 0x32
>> 0x75 0x69 0x64 0x3D 0x4A 0x65 0x64 0x65 0x72 0x2C 0x63 0x6E 0x3D 0x75
>> 0x73 0x65 0x72 0x
>> 73 0x2C 0x64 0x63 0x3D 0x56 0x45 0x52 0x57 0x41 0x4C 0x54 0x55 0x4E 0x47
>> 0x2C 0x64 0x63 0x3D 0x44 0x4F 0x55 0x47 0x4C 0x41 0x53 0x48 0x4F 0x4C
>> 0x44 0x49 0x4E 0x47 0x30 0x71 0x30 0x0E 0x04 0x03 0x75 0x69 0x64 0x31
>> 0x07 0x04 0x05
>> 0x4A 0x65 0x64 0x65 0x72 0x30 0x0D 0x04 0x02 0x73 0x6E 0x31 0x07 0x04
>> 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x0D 0x04 0x02 0x63 0x6E 0x31 0x07
>> 0x04 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x41 0x04 0x0B 0x6F 0x62 0x6A
>> 0x65 0x63 0x74 0x4
>> 3 0x6C 0x61 0x73 0x73 0x31 0x32 0x04 0x14 0x6F 0x72 0x67 0x61 0x6E 0x69
>> 0x7A 0x61 0x74 0x69 0x6F 0x6E 0x61 0x6C 0x50 0x65 0x72 0x73 0x6F 0x6E
>> 0x04 0x06 0x70 0x65 0x72 0x73 0x6F 0x6E 0x04 0x0D 0x69 0x6E 0x65 0x74
>> 0x4F 0x72 0x67 0
>> x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x03 0x74 0x6F 0x70
>>
>> The centos43 encoded PDU looks like this:
>>
>> [10:26:02] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
>> Encoded PDU : 0x30 0x81 0xB0 0x02 0x01 0x02 0x64 0x81 0xAA 0x04 0x35
>> 0x75 0x69 0x64 0x3D 0x4A 0x65 0x64 0x65 0x72 0x2C 0x20 0x63 0x6E 0x3D
>> 0x75 0x73 0x65 0x
>> 72 0x73 0x2C 0x20 0x64 0x63 0x3D 0x56 0x45 0x52 0x57 0x41 0x4C 0x54 0x55
>> 0x4E 0x47 0x2C 0x20 0x64 0x63 0x3D 0x44 0x4F 0x55 0x47 0x4C 0x41 0x53
>> 0x48 0x4F 0x4C 0x44 0x49 0x4E 0x47 0x30 0x71 0x30 0x0D 0x04 0x02 0x73
>> 0x6E 0x31 0x07
>> 0x04 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x41 0x04 0x0B 0x6F 0x62 0x6A
>> 0x65 0x63 0x74 0x43 0x6C 0x61 0x73 0x73 0x31 0x32 0x04 0x06 0x70 0x65
>> 0x72 0x73 0x6F 0x6E 0x04 0x14 0x6F 0x72 0x67 0x61 0x6E 0x69 0x7A 0x61
>> 0x74 0x69 0x6F 0x6
>> E 0x61 0x6C 0x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x0D 0x69 0x6E 0x65 0x74
>> 0x4F 0x72 0x67 0x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x03 0x74 0x6F 0x70
>> 0x30 0x0D 0x04 0x02 0x63 0x6E 0x31 0x07 0x04 0x05 0x4A 0x65 0x64 0x65
>> 0x72 0x30 0x0E 0
>> x04 0x03 0x75 0x69 0x64 0x31 0x07 0x04 0x05 0x4A 0x65 0x64 0x65 0x72
>>
>>
>> Putting this two PDU strings in a small program that converts that hex
>> strings to readable char string delivers the following:
>>
>> Windows 2003 Server:
>> 0 °   d ª 5uid=Jeder, cn=users, dc=VERWALTUNG,
>> dc=DOUGLASHOLDING0q0
>>  sn1   Jeder0A  objectClass12  person  organizationalPerson
>> inetOrgPerson  top0   cn1   Jeder0   uid1   Jeder
>>
>> CentOs 4.3:
>> 0 ­   d §   
>> 2uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING0q0   uid1     
>> Jeder0
>>  sn1   Jeder0
>>  cn1   Jeder0A  objectClass12  organizationalPerson  person
>> inetOrgPerson  top
>>
>>
>> My question is: what makes the difference? Where is the mistake? Where
>> is my mistake? What can I do? Is it a bug?
>>
>> Any tips and hints are really appreciated. If neccessary, I can send
>> server.xml and logfiles from both servers, or whatever is needed.
>>
>> Thanks in advance,
>> Markus
>>
>>
>
>
> -- 
> Regards,
> Cordialement,
> Emmanuel Lécharny
> www.iktek.com






Re: [TwixEncoder] :: delivers different PDUs for same search result on different operating systems

Posted by Emmanuel Lecharny <el...@gmail.com>.
Hi Markus,

the underlying OS won't add spaces in the DN ! Even if you dislike
Windows, this OS is not bad up to the point to add spaces just after a
comma... There should be something different.

I need some more info : can you try to inject a LDIF file into both
LDAP server with a new entry, and check if the result is the same when
requesting this entry ?

Now, the LDAP spec  says that DN are considered equals if each or its
RDN is equal, using its specific MatchingRule . But first, the DN must
be normalized, which means all spaces before or after ',' and '+' are
to be removed, and all special chars are to be escaped.
Doing a straight string comparizon is not good at all, because we keep
the user provided data into the server (meaning that if you inject the
DN
"Dc=My     OwN      Do MaIn   ,   dC = ORg", you will get it as is
when searching for this entry). This is a standard behavior, and you
can expect every LDAP server to give you back the DN you have
injected. However, if you do a search with 'dc=my won domain, dc=org",
the server will return the same entry.

Emmanuel

On 7/26/07, Markus Pohle <ap...@webunity.de> wrote:
>
> Hi Emmanuel,
>
> first of all thx for your reply!
>
> I did all my tests with the same apacheds-1.5.1-snapshot and testet
> ads first with 1.5.0_10 jdk and then did the test again with ads
> running with 1.6.0 jdk to be sure that there is no influence caused by
> the jdk.
>
> What I can say is, from within my tests, that the jdk version seems
> not to be relevant.
>
> What is relevant, from within my tests, seems to be the host operating
> system on which ads-1.5.1-snapshot is running.
>
> Here is, in your reply to me, a little mistake:
> On all unix-like operating systems I tested (with jdk 5/6) there are
> never whitespace in the returning DN. But on the windows based
> operating systems I tested there are whitespace in the returning DN.
>
> So the operating system on which ADS runs seems to have influence on
> what the returning PDU is.
>
> My question is: is this a bug in ADS? What can make ADS deliver
> different PDU for same ldap request? What can I do, except for your
> hint to use JNDI methods to do the comparison?
>
> TIA
> Markus
>
>
> Zitat von Emmanuel Lecharny <el...@gmail.com>:
>
> > Hi Markus,
> >
> > when I analyze both dump, what I can say is that the PDU lengths are
> > different. The first one (windows) has this DN :
> > uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING
> > The second one, Cent OS, contains spaces :
> > uid=Jedel, cn=users, dc=VERWALMUNG, dc=DOUGLASHOLDING
> >
> > This is why you have different lengths (3 bytes, as we have three spaces).
> >
> > As you may use different JVMs on those two servers, the attributes
> > order may vary too. Which JVM are you using ?
> >
> > Here are the decoded messages :
> > on windows :
> > 0x30 0x81 0xAD
> >  0x02 0x01 0x02
> >  0x64 0x81 0xA7
> >    0x04 0x32
> >      uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING
> >    0x30 0x71
> >      0x30 0x0E
> >        0x04 0x03
> >          uid
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >      0x30 0x0D
> >        0x04 0x02
> >          sn
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >      0x30 0x0D
> >        0x04 0x02
> >          cn
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >      0x30 0x41
> >        0x04 0x0B
> >          objectClass
> >        0x31 0x32
> >          0x04 0x14
> >            organizationalPerson
> >          0x04 0x06
> >            person
> >          0x04 0x0D
> >            inetOrgPerson
> >          0x04 0x03
> >            top
> >
> > on CentOS :
> > 0x30 0x81 0xB0
> >  0x02 0x01 0x02
> >  0x64 0x81 0xAA
> >    0x04 0x35
> >      uid=Jeder, cn=users, dc=VERWALMUNG, dc=DOUGLASHOLDING
> >    0x30 0x71
> >      0x30 0x0D
> >        0x04 0x02
> >          sn
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >      0x30 0x41
> >        0x04 0x0B
> >          objectClass
> >        0x31 0x32
> >          0x04 0x06
> >            Person
> >          0x04 0x14
> >            organizationalPerson
> >          0x04 0x0D
> >            inetOrgPerson
> >          0x04 0x03
> >            top
> >      0x30 0x0D
> >        0x04 0x02
> >          cn
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >      0x30 0x0E
> >        0x04 0x03
> >          uid
> >        0x31 0x07
> >          0x04 0x05
> >            Jeder
> >
> > One hypothesis is that you injected the last DN with spaces, which
> > should not be a pb at all, as spaces around ',' are to be ignored. Of
> > course, if you simply compare the resulting DN as strings, then you
> > will have a pb. You should use some of the JNDI methods to do those
> > comparisons (Name class)
> >
> > Hope it helps
> >
> > On 7/25/07, Markus Pohle <ap...@webunity.de> wrote:
> >> Hi list-users,
> >>
> >> I do have problem with my content management system working with
> >> apacheds on a windows 2003 server installation.
> >>
> >> My content management system (coremmedia 2005) tries to identify each
> >> user by ldap request against apacheds 1.5.x ldap server.
> >>
> >> What I found out is, that the ldap request against an apacheds-1.5.x on
> >> a windows 2003 server system isnt successful, while the same request
> >> against a centos 4.3 or fedora 7 apacheds-1.5.x installation is successful.
> >>
> >> How did I find that out? Well, I configured my CMS to connect against
> >> apacheds-1.5.x on a windows 2003 server, where apacheds logs are in
> >> debug mode. Then I tried to load CMS startpage, which is not successful.
> >> Then I did the same with CMS connected against apacheds-1.5.x running on
> >> centos 4.3 with logfiles in debug mode. Now the startpage was loaded.
> >>
> >> I then searched thru the logfiles and found out the following:
> >>
> >> - the search request from CMS is always the following:
> >>
> >> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixDecoder] -
> >> Decoded LdapMessage : LdapMessage
> >>     message Id : 2
> >>     Search Request
> >>         Base Object : 'cn=users,cn=cms,dc=APPLICATIONS,dc=DOUGLASHOLDING'
> >>         Scope : whole subtree
> >>         Deref Aliases : deref Always
> >>         Size Limit : no limit
> >>         Time Limit : no limit
> >>         Types Only : false
> >>         Filter : '(&(objectClass=inetOrgPerson)(uid=Jeder))'
> >>         Attributes : uid, mail, sn, cn, givenname, objectClass
> >>     Control
> >>         Control type : '2.16.840.1.113730.3.4.2'
> >>         Criticality : 'false'
> >>
> >> [19:48:34] DEBUG
> >> [org.apache.directory.shared.ldap.codec.TwixTransformer] - Transforming
> >> LdapMessage <2, SEARCH_REQUEST> from Twix to Snickers.
> >> [19:48:34] DEBUG
> >> [org.apache.directory.server.ldap.support.SearchHandler] - Message
> >> received:      SearchRequest
> >>         baseDn : 'cn=users,cn=cms,dc=APPLICATIONS,dc=DOUGLASHOLDING'
> >>         filter : '(& (objectClass=inetOrgPerson) (uid=Jeder) ) '
> >>         scope : whole subtree
> >>         typesOnly : false
> >> no limit
> >>         Time Limit : no limit
> >>         Deref Aliases : deref Always
> >>         attributes : 'uid', 'mail', 'sn', 'cn', 'givenname', 'objectClass'
> >>
> >> - the search result is the same on both systems:
> >>
> >> [19:48:34] DEBUG
> >> [org.apache.directory.shared.ldap.codec.TwixTransformer] - Transformed
> >> message : LdapMessage
> >>     message Id : 2
> >>     Search Result Entry
> >>         Object Name :
> >> '0.9.2342.19200300.100.1.1=jeder,2.5.4.3=users,0.9.2342.19200300.100.1.25=verwaltung,0.9.2342.19200300.100.1.25=douglasholding'
> >>         Attributes
> >>             Attributes
> >>             Attribute id : 'uid',  Values : ['Jeder']
> >>             Attribute id : 'sn',  Values : ['Jeder']
> >>             Attribute id : 'cn',  Values : ['Jeder']
> >>             Attribute id : 'objectClass',  Values :
> >> ['organizationalPerson', 'person', 'inetOrgPerson', 'top']
> >>
> >> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
> >> Encoding this LdapMessage : LdapMessage
> >>     message Id : 2
> >>     Search Result Entry
> >>         Object Name :
> >> '0.9.2342.19200300.100.1.1=jeder,2.5.4.3=users,0.9.2342.19200300.100.1.25=verwaltung,0.9.2342.19200300.100.1.25=douglasholding'
> >>         Attributes
> >>             Attributes
> >>             Attribute id : 'uid',  Values : ['Jeder']
> >>             Attribute id : 'sn',  Values : ['Jeder']
> >>             Attribute id : 'cn',  Values : ['Jeder']
> >>             Attribute id : 'objectClass',  Values :
> >> ['organizationalPerson', 'person', 'inetOrgPerson', 'top']
> >>
> >> But the next line in debug log which contains the Encoded PDU differs
> >> from windows 2003 server to centos43.
> >>
> >> The windows 2003 encoded PDU looks like this:
> >>
> >> [19:48:34] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
> >> Encoded PDU : 0x30 0x81 0xAD 0x02 0x01 0x02 0x64 0x81 0xA7 0x04 0x32
> >> 0x75 0x69 0x64 0x3D 0x4A 0x65 0x64 0x65 0x72 0x2C 0x63 0x6E 0x3D 0x75
> >> 0x73 0x65 0x72 0x
> >> 73 0x2C 0x64 0x63 0x3D 0x56 0x45 0x52 0x57 0x41 0x4C 0x54 0x55 0x4E 0x47
> >> 0x2C 0x64 0x63 0x3D 0x44 0x4F 0x55 0x47 0x4C 0x41 0x53 0x48 0x4F 0x4C
> >> 0x44 0x49 0x4E 0x47 0x30 0x71 0x30 0x0E 0x04 0x03 0x75 0x69 0x64 0x31
> >> 0x07 0x04 0x05
> >> 0x4A 0x65 0x64 0x65 0x72 0x30 0x0D 0x04 0x02 0x73 0x6E 0x31 0x07 0x04
> >> 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x0D 0x04 0x02 0x63 0x6E 0x31 0x07
> >> 0x04 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x41 0x04 0x0B 0x6F 0x62 0x6A
> >> 0x65 0x63 0x74 0x4
> >> 3 0x6C 0x61 0x73 0x73 0x31 0x32 0x04 0x14 0x6F 0x72 0x67 0x61 0x6E 0x69
> >> 0x7A 0x61 0x74 0x69 0x6F 0x6E 0x61 0x6C 0x50 0x65 0x72 0x73 0x6F 0x6E
> >> 0x04 0x06 0x70 0x65 0x72 0x73 0x6F 0x6E 0x04 0x0D 0x69 0x6E 0x65 0x74
> >> 0x4F 0x72 0x67 0
> >> x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x03 0x74 0x6F 0x70
> >>
> >> The centos43 encoded PDU looks like this:
> >>
> >> [10:26:02] DEBUG [org.apache.directory.shared.ldap.codec.TwixEncoder] -
> >> Encoded PDU : 0x30 0x81 0xB0 0x02 0x01 0x02 0x64 0x81 0xAA 0x04 0x35
> >> 0x75 0x69 0x64 0x3D 0x4A 0x65 0x64 0x65 0x72 0x2C 0x20 0x63 0x6E 0x3D
> >> 0x75 0x73 0x65 0x
> >> 72 0x73 0x2C 0x20 0x64 0x63 0x3D 0x56 0x45 0x52 0x57 0x41 0x4C 0x54 0x55
> >> 0x4E 0x47 0x2C 0x20 0x64 0x63 0x3D 0x44 0x4F 0x55 0x47 0x4C 0x41 0x53
> >> 0x48 0x4F 0x4C 0x44 0x49 0x4E 0x47 0x30 0x71 0x30 0x0D 0x04 0x02 0x73
> >> 0x6E 0x31 0x07
> >> 0x04 0x05 0x4A 0x65 0x64 0x65 0x72 0x30 0x41 0x04 0x0B 0x6F 0x62 0x6A
> >> 0x65 0x63 0x74 0x43 0x6C 0x61 0x73 0x73 0x31 0x32 0x04 0x06 0x70 0x65
> >> 0x72 0x73 0x6F 0x6E 0x04 0x14 0x6F 0x72 0x67 0x61 0x6E 0x69 0x7A 0x61
> >> 0x74 0x69 0x6F 0x6
> >> E 0x61 0x6C 0x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x0D 0x69 0x6E 0x65 0x74
> >> 0x4F 0x72 0x67 0x50 0x65 0x72 0x73 0x6F 0x6E 0x04 0x03 0x74 0x6F 0x70
> >> 0x30 0x0D 0x04 0x02 0x63 0x6E 0x31 0x07 0x04 0x05 0x4A 0x65 0x64 0x65
> >> 0x72 0x30 0x0E 0
> >> x04 0x03 0x75 0x69 0x64 0x31 0x07 0x04 0x05 0x4A 0x65 0x64 0x65 0x72
> >>
> >>
> >> Putting this two PDU strings in a small program that converts that hex
> >> strings to readable char string delivers the following:
> >>
> >> Windows 2003 Server:
> >> 0 °   d ª 5uid=Jeder, cn=users, dc=VERWALTUNG,
> >> dc=DOUGLASHOLDING0q0
> >>  sn1   Jeder0A  objectClass12  person  organizationalPerson
> >> inetOrgPerson  top0   cn1   Jeder0   uid1   Jeder
> >>
> >> CentOs 4.3:
> >> 0 ­   d §
> >> 2uid=Jeder,cn=users,dc=VERWALTUNG,dc=DOUGLASHOLDING0q0   uid1
> >> Jeder0
> >>  sn1   Jeder0
> >>  cn1   Jeder0A  objectClass12  organizationalPerson  person
> >> inetOrgPerson  top
> >>
> >>
> >> My question is: what makes the difference? Where is the mistake? Where
> >> is my mistake? What can I do? Is it a bug?
> >>
> >> Any tips and hints are really appreciated. If neccessary, I can send
> >> server.xml and logfiles from both servers, or whatever is needed.
> >>
> >> Thanks in advance,
> >> Markus
> >>
> >>
> >
> >
> > --
> > Regards,
> > Cordialement,
> > Emmanuel Lécharny
> > www.iktek.com
>
>
>
>
>
>


-- 
Regards,
Cordialement,
Emmanuel Lécharny
www.iktek.com