Red Hat Bugzilla – Bug 830342
cannot delete tombstone entries
Last modified: 2013-12-18 10:56:59 EST
This bug is created as a clone of upstream ticket: https://fedorahosted.org/389/ticket/376 An attempt to delete a tombstone entry gives err=32 - no such entry: {{{ 17/May/2012:01:41:35 +0000] conn=947 fd=80 slot=80 SSL connection from ::1 to ::1 [17/May/2012:01:41:35 +0000] conn=947 SSL 256-bit AES [17/May/2012:01:41:35 +0000] conn=947 op=0 BIND dn="cn=Directory Manager" method=128 version=3 [17/May/2012:01:41:35 +0000] conn=947 op=0 RESULT err=0 tag=97 nentries=0 etime=0 dn="cn=directory manager" [17/May/2012:01:41:35 +0000] conn=947 op=1 DEL dn="nsuniqueid=14403301-1dd211b2-9635fc8a-5f520000,uid=username,ou=people,dc=example,dc=com" [17/May/2012:01:41:35 +0000] conn=947 op=1 RESULT err=32 tag=107 nentries=0 etime=0 [17/May/2012:01:41:35 +0000] conn=947 op=1 RESULT err=32 tag=107 nentries=0 etime=0 [17/May/2012:01:41:35 +0000] conn=947 op=2 UNBIND [17/May/2012:01:41:35 +0000] conn=947 op=2 fd=80 closed - U1 }}} Note there are two result lines - this is bad.
please add steps to reproduce/verify this bug
The target milestone is 389-ds-base-1.3/RHEL7 or FUTURE.
This is not going to be implemented for RHEL 7.0. Pushing it out to RHEL 7.1 for now, but we might push it out even further than that.
The upstream ticket for this was closed as WONTFIX. Here is the explanation from the upstream ticket: Before I forget, here was the latest status... If a tombstone entry is in the cache, you can delete it, but when it's not, you can not delete the tombstone(error 32). Stack trace: ldbm_entryrdn.c:3245: _entryrdn_get_elem returns -30988 #0 _entryrdn_index_read (be=0xfef0a0, cursor=0x108ea60, srdn=0x7f04a25f3700, elem=0x7f04a25f3740, parentelem=0x0, childelems=0x0, flags=0, db_txn=0x0) at ../ds/ldap/servers/slapd/back-ldbm/ldbm_entryrdn.c:3247 #1 entryrdn_index_read_ext (be=0xfef0a0, sdn=0x7f04680008c0, id=0x7f04a25f380c, flags=0, txn=0x7f04a25f3b30) at ../ds/ldap/servers/slapd/back-ldbm/ldbm_entryrdn.c:439 #2 dn2entry_ext (be=0xfef0a0, sdn=0x7f04680008c0, txn=0x7f04a25f3b30, flags=0, err=0x7f04a25f38a8) at ../ds/ldap/servers/slapd/back-ldbm/dn2entry.c:92 #3 find_entry_internal_dn (pb=0x11c8660, be=0xfef0a0, sdn=0x7f04680008c0, lock=1, txn=0x7f04a25f3b30, flags=0) at ../ds/ldap/servers/slapd/back-ldbm/findentry.c:130 #4 find_entry_internal (pb=0x11c8660, be=0xfef0a0, addr=0x11c89e8, lock=1, txn=0x7f04a25f3b30, flags=0) at ../ds/ldap/servers/slapd/back-ldbm/findentry.c:293 #5 find_entry2modify (pb=0x11c8660, be=0xfef0a0, addr=0x11c89e8, txn=0x7f04a25f3b30) at ../ds/ldap/servers/slapd/back-ldbm/findentry.c:324 #6 ldbm_back_delete (pb=0x11c8660) at ../ds/ldap/servers/slapd/back-ldbm/ldbm_delete.c:185 The problem why it fails is that entryrdn_index_read_ext() strips off the tombstone flag so it ignores the tombstone rdns: dn2entry.c:92 *err = entryrdn_index_read_ext(be, sdn, &id, flags & TOMBSTONE_INCLUDED, txn); This was apparently stripped off to improve performance. Needs more investigation... Manual tombstone deletion is not recommended/supported. . As described in comment 6, the operation is only allowed when the tombstone entry is in entry cache. . Even if a tombstone entry is deleted on a master, the operation is not replicated to the consumers. . The same symptom is observed on RHDS8.2, as well. . We strongly recommend to use tombstone reap to clean up tombstone entries. A doc bug was filed for that: Bug 974695 - Adding tombstone entry and how to delete it to Admin Guide.