Re: [389-users] dirsrv won't start

Sorry I did not see the last time that error is in schema. I don't know how you did it but you break it. You don't need to panic. Because when you install a dirsrv instance the setup script copies the schema from somewhere... I don't remember from where... it was somewhere in /var or /usr... here are some ideas to get the good copy: - Red Hat docs will have this info also you can use find command to find this file. - Installing other dirsrv instance for a moment will create a good copy of this file in its directory. There can be many dirsrv servers on the system. You need to specify diferent port and DN. - you can read setup-ds and find from where it copies schema files. I'm traweling at the moment and I cannot give you more details at the moment. I am writing from my phone. I hope this will help. Greg. 11 sty 2013 21:06, "Doug Tucker" <tuckerd(a)lyle.smu.edu <javascript:_e({}, 'cvml', &#39;tuckerd(a)lyle.smu.edu&#39;);&gt;&gt; napisał(a): I was excited to see this reply, thanks so much. Unfortunately, I copied that to the dse.ldif and the results are the same. It won't start and with the same error. Sincerely, Doug Tucker On 01/11/2013 11:14 AM, Grzegorz Dwornicki wrote: For the record dirsrv creates file in its directory with the last good configuration. I believe it was called dse.ldif.startok Greg. 11 sty 2013 18:06, "Chandan Kumar" <chandank.kumar(a)gmail.com <mailto: chandank.kumar(a)gmail.**com&gt;&gt; napisał(a): You may not need to re-install it. If you could just replace the file that you changed, I hope you took a backup before experimenting with the file. Same thing happened with me too, I restored the directory server from the bakcup files. On Friday, January 11, 2013, Doug Tucker wrote: Well, I give up. I can find nothing in the docs or on google to get me around this. I'm see no way other than to uninstall 389 and reinstall from scatch so no need to respond to this. Sincerely, Doug Tucker On 01/10/2013 11:19 AM, Doug Tucker wrote: So I've gone from bad to worse. Googling and googling and no response on my auth issue from the list yesterday, I coudn't stand doing nothing. The only thing I saw that made me curious was some thread where a guy could not auth and he changed the password hash to something else and it worked. I looked at our current password hash in openldap and it was ssha. For the life of me I could not find how to see what the current one was in 389. The only thing I could find in the docs was how to set a password policy which allowed you to set the hash. So I did so according to the documentation on the Users cn. The only thing I did was turn it on, and make sure password hash was set to ssha. I left the rest default which was no expiration, etc. I saved, and tried to restart according to the docs, it woudn't restart. I shut down with the init script instead, and tried to start, and now I get this: [root@lyleauth1 schema]# /etc/init.d/dirsrv start Starting dirsrv: lyleauth1...[09/Jan/2013:16:**23:05 -0600] dse_read_one_file - The entry cn=schema in file /etc/dirsrv/slapd-lyleauth1/**schema/99user.ldif (lineno: 1) is invalid, error code 21 (Invalid syntax) - attribute type olcOverlay: Missing parent attribute syntax OID [09/Jan/2013:16:23:05 -0600] dse - Please edit the file to correct the reported problems and then restart the server. [FAILED] *** Warning: 1 instance(s) failed to start Looking at the time stamp on that file, it is: Dec 20 16:36 99user.ldif . So what I did yesterday did not touch it. Anyone have any idea how to fix this? -- 389 users mailing list 389-users(a)lists.fedoraproject.**org https://admin.fedoraproject.**org/mailman/listinfo/389-users<https://a... -- -- http://about.me/chandank -- 389 users mailing list 389-users(a)lists.fedoraproject.**org <mailto:389-users@lists.**fedoraproject.org> https://admin.fedoraproject.**org/mailman/listinfo/389-users<https://a... -- 389 users mailing list 389-users(a)lists.fedoraproject.**org <https://admin.fedoraproject.org/mailman/listinfo/389-users>