Deadlock condition in Directory Server Retro Changelog plug-in causes IPA to become unresponsive
Issue
A customer executed a host deletion operation (ipa host-del) which subsequently caused the IPA server to become unresponsive. An inspection of the Directory Server error log to determine the root cause of the problem revealed a deadlock condition in the Retro changelog plug-in (retrocl-plugin) :
[25/Apr/2016:12:55:39 +0000] - Retry count exceeded in delete
[25/Apr/2016:12:55:39 +0000] DSRetroclPlugin - delete_changerecord: could not delete change record 2107045 (rc: 51)
[25/Apr/2016:12:57:03 +0000] - Operation error fetching Null DN (3424e306-0ae511e6-a817c0c2-8e7cec37), error -30993.
[25/Apr/2016:12:57:03 +0000] - dn2entry_ext: Failed to get id for changenumber=2137810,cn=changelog from entryrdn index (-30993)
[25/Apr/2016:12:57:03 +0000] - Operation error fetching changenumber=2137810,cn=changelog (null), error -30993.
[25/Apr/2016:12:57:03 +0000] DSRetroclPlugin - replog: an error occured while adding change number 2137810, dn = changenumber=2137810,cn=changelog: Operations error.
[25/Apr/2016:12:57:03 +0000] retrocl-plugin - retrocl_postob: operation failure [1]
[25/Apr/2016:12:57:05 +0000] - ldbm_back_seq deadlock retry BAD 1601, err=0 BDB0062 Successful return: 0
[25/Apr/2016:12:57:24 +0000] - slapd shutting down - signaling operation threads - op stack size 11 max work q size 9 max work q stack size 9
[25/Apr/2016:12:57:24 +0000] - slapd shutting down - waiting for 29 threads to terminate
Environment
- Red Hat Enterprise Linux 7
- Red Hat Identity Management (IPA)
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
