Hello,
Last evening we experienced server corruption inside our replicated prefix.
Three users got moved from ou=People,dc=scripts,dc=mit,dc=edu to dc=scripts, and another user got moved from ou=People,dc=scripts to ou=People,ou=People,dc=scripts. We also couldn't fix it because it claimed that the correct DN existed.
We also saw replication conflicts using , as the thing after nsuniqueid instead of +.
Please advise.
Cheers, Edward
P.S. Our dirsrvs have been crashing and segfaulting because of the GSSAPI bugs, so if you want to blame it on that, ok, though dirsrv should not be corrupting data when it gets interrupted.
Edward Z. Yang wrote:
Hello,
Last evening we experienced server corruption inside our replicated prefix.
Three users got moved from ou=People,dc=scripts,dc=mit,dc=edu to dc=scripts, and another user got moved from ou=People,dc=scripts to ou=People,ou=People,dc=scripts. We also couldn't fix it because it claimed that the correct DN existed.
We also saw replication conflicts using , as the thing after nsuniqueid instead of +.
Please advise.
This could be https://bugzilla.redhat.com/show_bug.cgi?id=643532 which was fixed in 389-ds-base 1.2.7.a3.
Cheers, Edward
P.S. Our dirsrvs have been crashing and segfaulting because of the GSSAPI bugs, so if you want to blame it on that, ok, though dirsrv should not be corrupting data when it gets interrupted. -- 389 users mailing list 389-users@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/389-users
On 11/01/2010 06:34 AM, Rich Megginson wrote:
Edward Z. Yang wrote:
Hello,
Last evening we experienced server corruption inside our replicated prefix.
Three users got moved from ou=People,dc=scripts,dc=mit,dc=edu to dc=scripts, and another user got moved from ou=People,dc=scripts to ou=People,ou=People,dc=scripts. We also couldn't fix it because it claimed that the correct DN existed.
We also saw replication conflicts using , as the thing after nsuniqueid instead of +.
Please advise.
This could be https://bugzilla.redhat.com/show_bug.cgi?id=643532 which was fixed in 389-ds-base 1.2.7.a3.
Edward,
If you have hit the bug 643532, restarting the server temporarily "fixes" the problem. The DN is corrupted in the entry cache in memory. The entries in the backend db should be okay. (That's why you could not "fix it".) But of course, the corruption would be observed again. So, we recommend you to upgrade to 1.2.7.a3.
Thanks, --noriko
Cheers, Edward
P.S. Our dirsrvs have been crashing and segfaulting because of the GSSAPI bugs, so if you want to blame it on that, ok, though dirsrv should not be corrupting data when it gets interrupted. -- 389 users mailing list 389-users@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/389-users
-- 389 users mailing list 389-users@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/389-users
389-users@lists.fedoraproject.org