[Date Prev][Date Next] [Chronological] [Thread] [Top]
Re: (ITS#4394) Fix Segfault in auditlog.c

To: openldap-its@OpenLDAP.org
Subject: Re: (ITS#4394) Fix Segfault in auditlog.c
From: richter@ecos.de
Date: Fri, 10 Feb 2006 17:19:21 GMT
> >
> > Nevertheless the audit log shows deletion for all attributes of the
> > entry, followed by replace operation with the current values for all
> > attributes of entry. Maybe this is what you mean.
>
> Yes. I recently encountered this same problem in another section of
> modify.c. Please try the patch I've committed to CVS HEAD, it should
> work correctly.
>

I have taken modiy.c from HEAD, but it looks the same. Here is the log from 
the consumer:

Feb 10 18:01:53 bb154 slapd_2.3[30658]: syncrepl_entry: be_search (0)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: syncrepl_entry: 
ou=Dienste,cn=bb154,dc=testuml,dc=test
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify: 
ou=Dienste,cn=bb154,dc=testuml,dc=test
Feb 10 18:01:53 bb154 slapd_2.3[30658]: 
bdb_dn2entry("ou=dienste,cn=bb154,dc=testuml,dc=test")
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: 0x0000010c: 
ou=Dienste,cn=bb154,dc=testuml,dc=test
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= acl_access_allowed: granted to 
database root
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete sfbase-id
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
structuralObjectClass
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
creatorsName
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
createTimestamp
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
description
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete ou
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
objectClass
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete entryCSN
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
modifiersName
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete 
modifyTimestamp
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: delete entryUUID
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace sfbase-id
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
structuralObjectClass
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace entryUUID
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
creatorsName
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
createTimestamp
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
description
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace ou
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
objectClass
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace entryCSN
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
modifiersName
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace 
modifyTimestamp
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_modify_internal: replace entryUUID
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [04d92507]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [e1bc49b7]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [ef5a4630]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [fc32e8ab]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [e72f2b22]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [9bee355f]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [0096defd]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(DELETE,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_delete_key: 10c [473e72b9]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(ADD,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_insert_key: 10c [04d92507]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(ADD,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_insert_key: 10c [e1bc49b7]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(ADD,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_insert_key: 10c [ef5a4630]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0
Feb 10 18:01:53 bb154 slapd_2.3[30658]: => key_change(ADD,10c)
Feb 10 18:01:53 bb154 slapd_2.3[30658]: bdb_idl_insert_key: 10c [fc32e8ab]
Feb 10 18:01:53 bb154 slapd_2.3[30658]: <= key_change 0

This operation, should only replace attributes, but not delete any. On the 
master I only see the "bdb_modify_internal: replace", but no delete.

> > Additionals hints are, that I have disabled Sessionlog on the provider,
> > because I had problemes synchronising bursts of modification (this is
> > another issue, that I need to look deeper at) and I am currently testing
> > the replication with rootdn as replication user.
>
> The particular crash you saw is independent of the Sessionlog. But if
> you're seeing other problems there, let us know when you have more info.

At the moment I can't reproduce the problem with session log. I will send 
another message if it occurs again. 

I have other trouble with replication, but I will send a separate message.

Gerald



 
** Virus checked by BB-5000 Mailfilter **
Prev by Date: RE: (ITS#4394) Fix Segfault in auditlog.c
Next by Date: (ITS#4396) doc update for synrepl
Index(es):
- Chronological
- Thread