Hi Laurent,
389-Directory/3.1.1 B2024.289.0000
Looking at the stacktrace, i wonder if there is a possible deadlock in retroCL triming. Where thread 4 acquired the cn=changelog backend lock then hang for a write TXN and thread 20 created a write TXN (blocking thread 4) and wait for cn=changelog backend lock.
Thread 4 (Thread 0x7f680c2b56c0 (LWP 3980786) "ns-slapd"):
#0 0x00007f6d14081332 in __pthread_mutex_lock_full () at target:/lib64/libc.so.6
#1 0x00007f6d0f6b6bd2 in mdb_txn_renew0 () at target:/lib64/liblmdb.so.0.0.0
#2 0x00007f6d0f6b73c4 in mdb_txn_begin () at target:/lib64/liblmdb.so.0.0.0
#3 0x00007f6d0ef88b55 in dbmdb_start_txn () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#4 0x00007f6d0ef8c745 in dbmdb_txn_begin () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#5 0x00007f6d0ef113bd in dblayer_txn_begin () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#6 0x00007f6d0ef3be7b in ldbm_back_delete () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#7 0x00007f6d142297b4 in op_shared_delete.lto_priv () at target:/usr/lib64/dirsrv/libslapd.so.0
#8 0x00007f6d142ccd9d in delete_internal_pb.isra () at target:/usr/lib64/dirsrv/libslapd.so.0
#9 0x00007f6d14223f56 in slapi_delete_internal_pb () at target:/usr/lib64/dirsrv/libslapd.so.0
#10 0x00007f6d0ec095be in delete_changerecord () at target:/usr/lib64/dirsrv/plugins/libretrocl-plugin.so
#11 0x00007f6d0ec0a853 in changelog_trim_thread_fn () at target:/usr/lib64/dirsrv/plugins/libretrocl-plugin.so
#12 0x00007f6d13e4d3d7 in _pt_root () at target:/lib64/libnspr4.so
#13 0x00007f6d1407e168 in start_thread () at target:/lib64/libc.so.6
#14 0x00007f6d1410214c in __clone3 () at target:/lib64/libc.so.6Thread 20 (Thread 0x7f68025fe6c0 (LWP 3980449) "ns-slapd"):
#0 0x00007f6d1407a7e9 in __futex_abstimed_wait_common () at target:/lib64/libc.so.6
#1 0x00007f6d1407d239 in pthread_cond_wait@@GLIBC_2.3.2 () at target:/lib64/libc.so.6
#2 0x00007f6d13e467db in PR_EnterMonitor () at target:/lib64/libnspr4.so
#3 0x00007f6d0ef11405 in dblayer_txn_begin () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#4 0x00007f6d0ef2bced in ldbm_back_add () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#5 0x00007f6d142190f0 in op_shared_add.lto_priv () at target:/usr/lib64/dirsrv/libslapd.so.0
#6 0x00007f6d142cce7c in add_internal_pb.isra () at target:/usr/lib64/dirsrv/libslapd.so.0
#7 0x00007f6d142151d5 in slapi_add_internal_pb () at target:/usr/lib64/dirsrv/libslapd.so.0
#8 0x00007f6d0ec0bfd4 in retrocl_postob () at target:/usr/lib64/dirsrv/plugins/libretrocl-plugin.so
#9 0x00007f6d1427b5c0 in plugin_call_func.lto_priv () at target:/usr/lib64/dirsrv/libslapd.so.0
#10 0x00007f6d1427b931 in plugin_call_plugins () at target:/usr/lib64/dirsrv/libslapd.so.0
#11 0x00007f6d0ef441cd in ldbm_back_modify () at target:/usr/lib64/dirsrv/plugins/libback-ldbm.so
#12 0x00007f6d14267360 in op_shared_modify.lto_priv () at target:/usr/lib64/dirsrv/libslapd.so.0
#13 0x00007f6d1426910e in do_modify () at target:/usr/lib64/dirsrv/libslapd.so.0
#14 0x000056425298e8bb in connection_threadmain ()
#15 0x00007f6d13e4d3d7 in _pt_root () at target:/lib64/libnspr4.so
#16 0x00007f6d1407e168 in start_thread () at target:/lib64/libc.so.6
#17 0x00007f6d1410214c in __clone3 () at target:/lib64/libc.so.6
At the same time, I wonder if it could be related to https://github.com/389ds/389-ds-base/issues/6644.
Could you try to disable retroCL trimming to see if can give a relief.
best regards
thierry
Hi,
you can reach out the directory server developers at 389-users@lists.fedoraproject.org.They will ask you to provide logs obtained as described here (Debugging Hangs), with the exact version and OS you have installed on your machines.
flo
On Sat, Mar 1, 2025 at 7:33 PM ARNAL Laurent via FreeIPA-users <freeipa-users@lists.fedorahosted.org> wrote:
Hello,
Some more infos : I've reinstalled the replica this afternoon.
Now the replication seems to work ok again.
But I've still have the deadlock after that.
Laurent.
--
_______________________________________________
FreeIPA-users mailing list -- freeipa-users@lists.fedorahosted.org
To unsubscribe send an email to freeipa-users-leave@lists.fedorahosted.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedorahosted.org/archives/list/freeipa-users@lists.fedorahosted.org
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
No comments:
Post a Comment