On 03/10/2014 08:33 PM, Timothy Pollard wrote:
On Mon, 10 Mar 2014 18:41:35 -0600
Rich Megginson <rmeggins@redhat.com> wrote:
On 03/10/2014 04:14 PM, Timothy Pollard wrote:
Hi,

We received this strange error a few hours ago for one of our DNS entries:

[10/Mar/2014:15:56:01 +0000] - str2entry_fast: Error. Non-contiguous
attribute values for tXTRecord [10/Mar/2014:15:56:01 +0000] - Entry
"relativeDomainName=testingstatus,zoneName=OURORG.com,dc=DNS,o=Internet"
required attribute "objectclass" missing

This is a monitor DNS entry that we update every four minutes to verify
that our Replication is working across the board, so that next error
repeats every four minutes as we try to update the entry.

The entry looks like it's had everything except for the tXTRecord field
deleted as well. Fortunately it's a testing entry, so we can easily rebuild
it manually.

This error occurred a few minutes after a bad restart which required a
changelog regenerate, so that may be related.

Anyone know what might cause this?
No.  Let's take a look at the entry with state information:

ldapsearch -x -D "cn=directory manager" -W -s base -b
"relativeDomainName=testingstatus,zoneName=OURORG.com,dc=DNS,o=Internet"
"objectclass=*" nscpEntryWsi

If that doesn't work, let's take a look at the raw database entry:

dbscan -f /var/lib/dirsrv/slapd-INST/db/userRoot/entryrdn.* | grep -i
relativeDomainName=testingstatus

That will tell you the ID e.g. ID: 10

Then use that ID to look at the raw database entry:

dbscan -f /var/lib/dirsrv/slapd-INST/db/userRoot/id2entry.* -K 10

I already manually re-added the missing attributes (such as the objectClass,
etc), but I've attached the ldapsearch output to this email.

Ok.  It would be useful to see it (and the db dump) when you can reproduce the problem.


I decided to have a look at the raw database entry as well, the id2entry.db4
entry looked to be the same. One thing I noticed was that it was listed twice
in the entryrdn.db4 DB though:

$ dbscan -f db/dnsRoot/entryrdn.db4 | grep -i testingstatus
  ID: 250737; RDN: "relativeDomainName=testingstatus"; NRDN: "relativedomainname=testingstatus"
  ID: 250737; RDN: "relativeDomainName=testingstatus"; NRDN: "relativedomainname=testingstatus"

That's normal.  One entry for the actual node, one entry for its role as the child of its parent.

$ dbscan -f db/dnsRoot/id2entry.db4 -K 250737

To dump the raw database entry.

Thanks,


--
389 users mailing list
389-users@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/389-users