My Oracle Support Banner

ODSEE - Very Slow Performance While Deleting Entries From Directory Server Due to Referential Integrity Plugin Operations (Doc ID 1326726.1)

Last updated on AUGUST 14, 2018

Applies to:

Oracle Directory Server Enterprise Edition - Version 5.2 SP4 and later
Information in this document applies to any platform.
***Checked for relevance on 06-Jan-2014***

Symptoms

While deleting entries, at some point, the Directory Server became unresponsive.

For example -

Thirty entries were deleted with no performance problems. Then the next 30 entries also were deleted as expected. But then while deleting the next 30 entries, the Directory Server became unresponsive.

A review of the pstacks show several threads running referential integrity operations.

----------------- lwp# 28 / thread# 28 --------------------
ffffffff7c117dac lwp_park (0, 0, 0)
ffffffff7c11390c mutex_lock_queue (ffffffff7c21baf8, 0, 1002457a0, 0, 0, 0) + 104
ffffffff7c114348 slow_lock (1002457a0, ffffffff40302800, 0, 0, 0, 0) + 58
ffffffff7d12e0cc PR_Lock (1002457a0, 4000, 0, 10, 0, 12) + 4
ffffffff75f0345c ???????? (100355570, 11b825ef0, 0, 0, 1002c3f68, 1001c5181)
ffffffff75f02148 referint_postop_del (0, ffffffff75f01f50, 118539408, 11d495c10, 0, 0) + 1f8
ffffffff7ec84ab0 ???????? (1c00, 1ea8, 118539408, 5400, 5638, 1)
ffffffff7ec83524 plugin_call_plugins (118539408, 1fc, ffffffff3f91f5f4, 1006c9910, 0, 11a6d3df0) + 2d4
ffffffff7ecb4b5c ???????? (118539408, 65, 0, 9f, 1f, 3c00)
ffffffff7ecb4578 delete_core_pb (118539408, 24e, 135904, 0, ffffffff, 0) + 178
ffffffff7ea3a738 dispatch_operation_core_pb (118539408, 118539408, ffffffff7ea27984, 2000, ffffffff7ee1b1d8, ffffffff7ee1d1d8) +
ffffffff7ea26420 ???????? (ffffffff4080cac8, 124baafa0, 94, 4a, 2, 200000)
ffffffff7ea2672c ldap_frontend_main_using_core_api (ffffffff4080cac8, 124baafa0, 2, 10034a1f8, 12c049710, 0) + fc
ffffffff7ea27764 ???????? (220, 100261150, 11169c, 800, 958, ffffffff4080cac8)
ffffffff7d136a7c _pt_root (100261150, 0, 0, 0, 20000, ffffffff3f91ff48) + d4
ffffffff7c117c9c _lwp_start (0, 0, 0, 0, 0, 0)



Error logs show numerous unindexed searches:

[29/May/2011:05:16:02 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed
[29/May/2011:05:16:40 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed
[29/May/2011:05:17:18 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed
[29/May/2011:05:17:55 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed
[29/May/2011:05:18:33 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed
[29/May/2011:05:19:11 -0400] - WARNING<20805> - Backend Database - conn=-1 op=0 msgId=-1 - search is not indexed

Changes

 

Cause

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


In this Document
Symptoms
Changes
Cause
Solution
 Procedure for Capturing pstacks


My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.