Loading...

Linking RHIVOS CVEs to...

Migration: Automation ...

SWIFT: POC Conversion

Sync from "Extern...

XML

Word

Printable

Type: Bug
Resolution: Done-Errata
Priority: Major
Fix Version/s: rhel-9.6
Affects Version/s: rhel-9.4
Component/s: 389-ds-base
Labels:
- CEE.neXT
- CEE.nxt

Fixed in Build:
389-ds-base-2.6.1-1.el9
Regression:
No
Severity:
Important
Keywords:

ZStream

AssignedTeam:
rhel-idm-ds
Sub-System Group:

ssg_idm

Internal Target Milestone:
26
Story Points:
0
Blocked:
False
Ready:
False
Blocked Reason:

Hide

None

Show
None
Product Documentation Required:
Yes
Sprint:
None
Release Blocker:
Approved Blocker

Preliminary Testing:
Pass
Errata Link:
https://errata.engineering.redhat.com/advisory/144130
Test Coverage:

RegressionOnly

Release Note Type:
Bug Fix
Release Note Text:

Hide
.`cleanAllRUV` no longer blocks itself

Before this update, when you ran the `cleanAllRUV` task after a replica deletion from replication topology, the task was trying to update the replication configuration entry while the same task was purging the replication changelog of the old replica ID (`rid`). As a result, the server was unresponsive.

With this update, `cleanAllRUV` cleans up the replication configuration only after the changelog purging is complete.

Show
.`cleanAllRUV` no longer blocks itself Before this update, when you ran the `cleanAllRUV` task after a replica deletion from replication topology, the task was trying to update the replication configuration entry while the same task was purging the replication changelog of the old replica ID (`rid`). As a result, the server was unresponsive. With this update, `cleanAllRUV` cleans up the replication configuration only after the changelog purging is complete.
Release Note Status:
Done

Experience:
Architecture:

x86_64

PX Impact Score:
PX Review Complete:
PX Scheduling Request:
SFDC Cases Counter:
SFDC Cases Open:
SFDC Cases Links:

Planning:
None

What were you trying to do that didn't work?

This issue typically happens right after an IPA replica is deleted ( ipa server-del ).
The deletion process triggers the purging of replicaID(s) used by the removed replica.
The purging thread uses a high amount of CPU and the LDAP won't respond to requests.
Killing the LDAP server is sometimes the only option to recover.
A few customers are noticing this behaviour ( a common pattern is the RHEL version that is 9.4 ).

A couple of stacktraces from 2 different servers:

   PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
1033655 dirsrv    20   0 1805076 546380 327680 R  99.9   9.4   2:25.51 ns-slapdThread 43 (Thread 0x7f0ba44af640 (LWP 1033655) "ns-slapd"):
#0  0x00007f0bc6015c75 in __db_tas_mutex_lock_int () at target:/lib64/libdb-5.3.so
#1  0x00007f0bc60d07fc in __db_cursor_int () at target:/lib64/libdb-5.3.so
#2  0x00007f0bc60d3481 in __dbc_idup () at target:/lib64/libdb-5.3.so
#3  0x00007f0bc60d3cb6 in __dbc_iget () at target:/lib64/libdb-5.3.so
#4  0x00007f0bc60e25e1 in __dbc_get_pp () at target:/lib64/libdb-5.3.so
#5  0x00007f0bc623e304 in bdb_dblayer_cursor_iterate () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#6  0x00007f0bc815995c in _cl5Iterate () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#7  0x00007f0bc8159e3c in _cl5PurgeRID () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#8  0x00007f0bc815c98e in trigger_cl_purging_thread () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#9  0x00007f0bca76ebd4 in _pt_root () at target:/lib64/libnspr4.so
#10 0x00007f0bca089c02 in start_thread () at target:/lib64/libc.so.6
#11 0x00007f0bca10ec40 in clone3 () at target:/lib64/libc.so.6


    PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
 994148 dirsrv    20   0 1668720 251560 137984 R  99.9   4.3   1:36.64 ns-slapdThread 42 (Thread 0x7fe57ba61640 (LWP 994148) "ns-slapd"):
#0  0x00007fe59e01db74 in __bamc_next () at target:/lib64/libdb-5.3.so
#1  0x00007fe59e0220dc in __bamc_get () at target:/lib64/libdb-5.3.so
#2  0x00007fe59e0d3d4d in __dbc_iget () at target:/lib64/libdb-5.3.so
#3  0x00007fe59e0e25e1 in __dbc_get_pp () at target:/lib64/libdb-5.3.so
#4  0x00007fe59e23e304 in bdb_dblayer_cursor_iterate () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#5  0x00007fe59df5195c in _cl5Iterate () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#6  0x00007fe59df51e3c in _cl5PurgeRID () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#7  0x00007fe59df5498e in trigger_cl_purging_thread () at target:/usr/lib64/dirsrv/plugins/libreplication-plugin.so
#8  0x00007fe5a2761bd4 in _pt_root () at target:/lib64/libnspr4.so
#9  0x00007fe5a2089c02 in start_thread () at target:/lib64/libc.so.6
#10 0x00007fe5a210ec40 in clone3 () at target:/lib64/libc.so.6

What is the impact of this issue to you?

The LDAP server becomes unresponsive.

Please provide the package NVR for which the bug is seen:

  cat etc/redhat-release 
Red Hat Enterprise Linux release 9.4 (Plow)

  grep ^ipa installed-rpms 
ipa-client-4.11.0-15.el9_4.x86_64                           Mon Jun 24 12:26:48 2024
ipa-client-common-4.11.0-15.el9_4.noarch                    Mon Jun 24 12:26:32 2024
ipa-common-4.11.0-15.el9_4.noarch                           Mon Jun 24 12:26:48 2024
ipa-healthcheck-0.16-3.el9.noarch                           Mon May  6 11:54:23 2024
ipa-healthcheck-core-0.16-3.el9.noarch                      Mon May  6 11:53:01 2024
ipa-selinux-4.11.0-15.el9_4.noarch                          Mon Jun 24 12:26:40 2024
ipa-server-4.11.0-15.el9_4.x86_64                           Mon Jun 24 12:27:10 2024
ipa-server-common-4.11.0-15.el9_4.noarch                    Mon Jun 24 12:26:32 2024
ipa-server-dns-4.11.0-15.el9_4.noarch                       Mon Jun 24 12:27:12 2024

How reproducible is this bug?:

Quite often on RHEL 9.4.

Steps to reproduce

Delete an IPA server
Check the CPU usage of the LDAP server on other replicas in the topology
Check for the pattern "CleanAllRUV" in the LDAP errors log

Expected results

Working LDAP server.

Actual results

Unresponsive LDAP server.

relates to

RHEL-64854 cleanallruv consums CPU and is slow

Closed

links to

https://github.com/389ds/389-ds-base/issues/6381

RHBA-2024:144130 389-ds-base bug fix and enhancement update

Assignee:: Mark Reynolds

Reporter:: Têko Mihinto

Developer:: IdM DS Dev

QA Contact:: Viktor Ashirov

Doc Contact:: Evgenia Martyniuk

Votes:: 1 Vote for this issue

Watchers:: 22 Start watching this issue

Created:: 2024/09/25 5:33 PM

Updated:: 2025/09/14 12:29 AM

Resolved:: 2025/05/13 11:47 AM

Target end:: 2025/02/17

Next Planned Release Date:: 2025/05/13

Release Date:: 2025/05/13

Details

Description

What were you trying to do that didn't work?

What is the impact of this issue to you?

Please provide the package NVR for which the bug is seen:

How reproducible is this bug?:

Steps to reproduce

Expected results

Actual results

Attachments

Issue Links

Easy Agile Planning Poker

Activity

People

Dates

Hide