Should only update modifyTimestamp/modifiersName on MODIFY ops

[389-ds-bot]

Cloned from Pagure issue: https://pagure.io/389-ds-base/issue/129

• Created at 2012-01-04 19:46:25 by mkosek
• Closed as Fixed
• Assigned to mreynolds (@mreynolds389)
• Associated bugzillas
  • https://bugzilla.redhat.com/show_bug.cgi?id=249327
  • https://bugzilla.redhat.com/show_bug.cgi?id=834054

---

https://bugzilla.redhat.com/show_bug.cgi?id=249327

The directory server updates the modifyTimestamp on operations other than
Modify
operations.  See http://tools.ietf.org/html/rfc4512.  However, attributes can
be
modified by operations other than Modify.  For example, with password policy
enabled, a Bind operation may update one or more operational attributes in a
user's entry.  Admin's do not expect modifyTimestamp to be updated under this
and similar circumstances.  See also
http://www.openldap.org/its/index.cgi/Software%20Bugs?id=5040;page=2

[389-ds-bot]

Comment from rmeggins (@richm) at 2012-01-10 06:16:59

batch move to milestone 1.3

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2012-02-08 03:22:11

This was fixed for password policy after the bug was filed.

I expanded the original fix with a "set" pblock function, so now plugins that are doing internal modifies have the option of not setting the "modify attrs". Currently this only effects the Account Policy plugin.

[389-ds-bot]

Comment from nhosoi (@nhosoi) at 2012-02-08 03:33:29

2826	        case SLAPI_SKIP_MODIFIED_ATTRS: 
2827	                if(value == 0){ 
2828	                        pblock->pb_op->o_flags &= OP_FLAG_SKIP_MODIFIED_ATTRS; 
2829	                } else { 

Should the line 2828 be something like this (not to affect the other bits)???
2828 pblock->pb_op->o_flags &= ~OP_FLAG_SKIP_MODIFIED_ATTRS;

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2012-02-08 03:39:04

Nice catch!

I think we might also want to use this new flag in all of our plugins that do internal modifies. Look at ticket 111 to see what I'm talking about.

[389-ds-bot]

Comment from rmeggins (@richm) at 2012-02-08 04:10:00

The pblock code for case SLAPI_SKIP_MODIFIED_ATTRS should first check to see if pblock->pb_op is not NULL. Although I can't foresee anyone using SLAPI_SKIP_MODIFIED_ATTRS outside the context of an operation, I've still seen similar cases where we've been bitten by segfaults.

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2012-02-08 22:04:34

attachment
0001-Ticket-129-Should-only-update-modifyTimestamp-modifi.patch

[389-ds-bot]

Comment from mreynolds (@mreynolds389) at 2012-02-09 00:10:21

[mareynol@localhost slapd]$ git merge ticket129
Updating ee32016..3e60f7f
Fast-forward
ldap/servers/plugins/acctpolicy/acct_plugin.c | 1 +
ldap/servers/slapd/pblock.c | 10 +++++++++-
2 files changed, 10 insertions(+), 1 deletions(-)

[mareynol@localhost slapd]$ git push origin master
Counting objects: 17, done.
Delta compression using up to 4 threads.
Compressing objects: 100% (9/9), done.
Writing objects: 100% (9/9), 1.10 KiB, done.
Total 9 (delta 7), reused 0 (delta 0)
To ssh://git.fedorahosted.org/git/389/ds.git
ee32016..3e60f7f master -> master

[389-ds-bot]

Comment from rmeggins (@richm) at 2012-03-31 03:36:34

originally targeted for 1.2.11.rc1, but actually in the 1.2.11.a1 release

[389-ds-bot]

Comment from nkinder (@nkinder) at 2012-06-20 23:59:48

Ticket has been cloned to Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=834054

[389-ds-bot]

Comment from nkinder (@nkinder) at 2012-08-28 04:14:30

Added initial screened field value.

[389-ds-bot]

Comment from nhosoi (@nhosoi) at 2017-02-11 22:57:03

Metadata Update from @nhosoi:

• Issue assigned to mreynolds389
• Issue set to the milestone: 1.2.11.a1