testing ebox 1.3

[jonny2k]

I'm testing ebox 1.3 at the moment.
First I have to say, you are doing a great job.
I don't know if this is the right place to post the bugs i've found but I hope I can help you developing by posting them here.

1) If I leave the ebox-webinterface password blank in the first configuration i'm unable to change it. If I'm going to change it in the webfrontend and leave current password blank (because the isn't one) it says "Password is empty."!

2) I'v configured the server to be the master-ldap server and set the password. After that i created a user without any problem. But activating the "File Sharing" in the "Module Status" takes very long.
There seems to be an problem with ldap.
Output of /var/log/syslog (while adding the module filesharing in module status):

Code: [Select]

Jul 19 20:05:55 filetest last message repeated 3 times                                                                                                                                               
Jul 19 20:05:55 filetest slapd[6370]: connection_read(17): no connection!                                                                                                                           
Jul 19 20:06:02 filetest slapd[6370]: <= bdb_equality_candidates: (sambaDomainName) not indexed                                                                                                     
Jul 19 20:06:07 filetest slapd[6370]: <= bdb_equality_candidates: (sambaDomainName) not indexed                                                                                                     
Jul 19 20:06:07 filetest slapd[6370]: connection_read(18): no connection!                                                                                                                           
Jul 19 20:06:11 filetest slapd[6370]: <= bdb_equality_candidates: (cn) not indexed                                                                                                                   
Jul 19 20:06:11 filetest slapd[6370]: <= bdb_equality_candidates: (cn) not indexed                                                                                                                   
Jul 19 20:06:23 filetest slapd[6370]: connection_read(17): no connection!                                                                                                                           
Jul 19 20:06:23 filetest slapd[6370]: <= bdb_equality_candidates: (sambaDomainName) not indexed                                                                                                     
Jul 19 20:06:23 filetest slapd[6370]: connection_read(18): no connection!                                                                                                                           
Jul 19 20:06:24 filetest slapd[6370]: connection_read(17): no connection!                                                                                                                           
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaGroupType) not indexed                                                                                                       
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaSIDList) not indexed                                                                                                         
Jul 19 20:06:24 filetest last message repeated 4 times                                                                                                                                               
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaGroupType) not indexed                                                                                                       
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaSIDList) not indexed                                                                                                         
Jul 19 20:06:24 filetest last message repeated 4 times                                                                                                                                               
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaGroupType) not indexed                                                                                                       
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaSIDList) not indexed                                                                                                         
Jul 19 20:06:24 filetest last message repeated 4 times                                                                                                                                               
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaGroupType) not indexed                                                                                                       
Jul 19 20:06:24 filetest slapd[6370]: <= bdb_equality_candidates: (sambaSIDList) not indexed                                                                                                         
Jul 19 20:06:24 filetest last message repeated 4 times
Jul 19 20:06:34 filetest slapd[6370]: <= bdb_equality_candidates: (cn) not indexed
Jul 19 20:06:34 filetest slapd[6370]: <= bdb_equality_candidates: (cn) not indexed
Jul 19 20:06:35 filetest slapd[6370]: <= bdb_equality_candidates: (sambaDomainName) not indexed
Jul 19 20:06:35 filetest slapd[6370]: connection_read(19): no connection!                       
Jul 19 20:06:38 filetest slapd[6370]: connection_read(17): no connection!


[J. A. Calvo]

Hi! thanks for testing our development version, you are of great help with that!

1) Yeah, that's a know bug, when using the official eBox installer it doesn't allow you to choose a blank password, but if you install the packages there is that bug. You can fix it by simply executing "dpkg-reconfigure ebox" and choosing a non-blank password.

2) How much time is "very long"? Is it working correctly at the end of the enable? I know that it takes time to enable but it works ok after that. Take into account that you only have to enable the module one time so that's not very critical.

Thanks again and let us know any bug you encounter!

Regards,

J. A. Calvo

[jonny2k]

hi

1)thanks for the tip with the password.

2) yes you are right. after the activation of the modul it worked perfect.

3)Now I wanted to add a slave machine which use the ldap of the first machine. I'v done this like it is described here:
http://trac.ebox-platform.com/wiki/Document/HowTo/EBoxMasterSlaveSetup

When I'm adding the second machine (activate the module "users and groups" on the configured slave machine) I got these messages on the master (syslog):

Code: [Select]

Jul 21 13:00:26 filetest last message repeated 2 times                               
Jul 21 13:00:27 filetest slapd[5014]: send_ldap_response: ber write failed
Jul 21 13:00:28 filetest slapd[5014]: connection_read(24): no connection!
Jul 21 13:00:29 filetest slapd[5014]: connection_read(19): no connection!
Jul 21 13:00:29 filetest slapd[5014]: connection_read(24): no connection!
Jul 21 13:00:30 filetest slapd[5014]: connection_read(19): no connection!
Jul 21 13:00:31 filetest slapd[5014]: connection_read(19): no connection!
Jul 21 13:00:31 filetest slapd[5014]: connection_read(24): no connection!
on the slave it got these syslog:

Code: [Select]

Jul 21 13:00:56 inettest slapd[7144]: syncrepl_message_to_entry: rid=111 mods check (objectClass: value #1 invalid per syntax)
Jul 21 13:00:56 inettest slapd[7144]: do_syncrepl: rid=111 retrying (4 retries left)                                           
Jul 21 13:00:57 inettest slapd[7144]: syncrepl_message_to_entry: rid=110 mods check (objectClass: value #3 invalid per syntax)
Jul 21 13:00:57 inettest slapd[7144]: do_syncrepl: rid=110 retrying (4 retries left)                                           
Jul 21 13:00:57 inettest slapd[7144]: syncrepl_message_to_entry: rid=111 mods check (objectClass: value #1 invalid per syntax)
Jul 21 13:00:57 inettest slapd[7144]: do_syncrepl: rid=111 retrying (4 retries left)                                         
and the webinterface stops with:

Code: [Select]

A really nasty bug has occurred
Exception
Failed to enable: Replication failed
Trace
Failed to enable: Replication failed at /usr/share/perl5/EBox/CGI/ServiceModule/ConfigureModuleController.pm line 74
EBox::CGI::ServiceModule::ConfigureModuleController::_process('EBox::CGI::ServiceModule::ConfigureModuleController=HASH(0xaf...') called at /usr/share/perl5/EBox/CGI/Base.pm line 261
EBox::CGI::Base::run('EBox::CGI::ServiceModule::ConfigureModuleController=HASH(0xaf...') called at /usr/share/perl5/EBox/CGI/Run.pm line 120
EBox::CGI::Run::run('EBox::CGI::Run', 'ServiceModule/ConfigureModuleController', 'EBox') called at /usr/share/ebox/cgi/ebox.cgi line 19
ModPerl::ROOT::ModPerl::Registry::usr_share_ebox_cgi_ebox_2ecgi::handler('Apache2::RequestRec=SCALAR(0xafbf160)') called at /usr/lib/perl5/ModPerl/RegistryCooker.pm line 204
eval {...} called at /usr/lib/perl5/ModPerl/RegistryCooker.pm line 204
ModPerl::RegistryCooker::run('ModPerl::Registry=HASH(0xafbef8c)') called at /usr/lib/perl5/ModPerl/RegistryCooker.pm line 170
ModPerl::RegistryCooker::default_handler('ModPerl::Registry=HASH(0xafbef8c)') called at /usr/lib/perl5/ModPerl/Registry.pm line 31
ModPerl::Registry::handler('ModPerl::Registry', 'Apache2::RequestRec=SCALAR(0xafbf160)') called at -e line 0
eval {...} called at -e line 0

[jonny2k]

hi,

that's no bug but a feature ...

4) it would be nice, if the machinename is displayed in the top or/and title of the frontend.
that would make it more easy to see at which machine you are working, when you have to adminitrate more than one ebox.

[isaac]

Hi Jonny2k!

Thank you for reporting these issues. It's the first time I see that kind of problem with the master-slave setup (and I can tell you I have seen loads of different ones while developing it ).

In any case it somehow seems like you are lacking some schemas in the replicated server. Can you run:

Code: [Select]

slapcat -F /etc/ldap/slapd-replica.d/ -b 'cn=config' | grep schema
and paste here the output?

Thanks!

[jonny2k]

hi, here is the output of the command ran on the slave machine:

Code: [Select]

root@inettest:~# slapcat -F /etc/ldap/slapd-replica.d/ -b 'cn=config' | grep schema
dn: cn=schema,cn=config
cn: schema
dn: cn={0}core,cn=schema,cn=config
dn: cn={1}cosine,cn=schema,cn=config
dn: cn={2}nis,cn=schema,cn=config
dn: cn={3}inetorgperson,cn=schema,cn=config
dn: cn={4}passwords,cn=schema,cn=config
dn: cn={5}samba,cn=schema,cn=config
dn: cn={6}ebox,cn=schema,cn=config
dn: cn={7}authldap,cn=schema,cn=config
dn: cn={8}eboxmail,cn=schema,cn=config
 eidt" scope=sub schemachecking=off type=refreshAndPersist retry="1 5 60 +"
 heidt" scope=sub schemachecking=off type=refreshAndPersist retry="1 5 60 +"
i hope it helps to understand that problem

greets jonny2k

[jonny2k]

hi,
i retried the master/slave setup by doing following on both machines

Code: [Select]

apt-get --purge remove ebox*
apt-get --purge remove slapd*
apt-get install ebox-usersadngroups
then i activated the usermodules with no users or groups created. and the moduleactivation worked fine...
then i added a user on the master. and the same problem occured. The users are not or just partly synchronised.
in the "slave status" there are alway some "pending operations" listet.
But with a look in the /var/log/ebox/ebox.log when i clicked on "force sync" in "slave status" i've found some interesting thing:

Code: [Select]

2009/07/24 14:01:35 DEBUG> SOAPClient.pm:108 EBox::SOAPClient::__ANON__ - 500 Can't connect to inettest:443 (Bad hostname 'inettest')
2009/07/24 14:01:35 DEBUG> slave-sync:65 main::__ANON__ - Trying to run method addUser with param test1 on slave inettest failed
2009/07/24 14:01:35 DEBUG> SOAPClient.pm:108 EBox::SOAPClient::__ANON__ - 500 Can't connect to inettest:443 (Bad hostname 'inettest')
2009/07/24 14:01:35 DEBUG> slave-sync:65 main::__ANON__ - Trying to run method addUser with param test2 on slave inettest failed
2009/07/24 14:01:35 DEBUG> SOAPClient.pm:108 EBox::SOAPClient::__ANON__ - 500 Can't connect to inettest:443 (Bad hostname 'inettest')
2009/07/24 14:01:35 DEBUG> slave-sync:65 main::__ANON__ - Trying to run method modifyUser with param test2 on slave inettest failed
2009/07/24 14:01:35 ERROR> Sudo.pm:214 EBox::Sudo::_rootError - root command /usr/share/ebox-usersandgroups/slave-sync failed.
Error output: Can't open the journal dir: /var/lib/ebox-usercorner/userjournal/
so i installed ebox-dns and configured the hosts, so that domainlookup worked:

Code: [Select]

root@filetest:~# nslookup inettest
Server:         192.168.1.101
Address:        192.168.1.101#53

Name:   inettest.lingscheidt
Address: 192.168.1.101

and great after that a "force sync" worked and all users and groups worked on the slave. and "There are no pending operations at the moment" in the slave status .

maybe these was the problem from the beginning. but i'm not sure.
hope it helps

[isaac]

The problem that prevented the replication in the first place had to be something different, but I am glad it's already working.

And yeah, when the slave registers in the master it tells the master its hostname and that hostname must be resolvable by the master to perform some operations in the slave.

I'll add that to the documentation.

Thank you