From 891e9c7cb924830334a42864ef2582e545f42723 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabiano=20Fid=C3=AAncio?= <fidencio@redhat.com>
Date: Mon, 29 May 2017 13:32:59 +0200
Subject: [PATCH 178/181] SYSDB_OPS: Mark an entry as expired also in the
 timestamp cache
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As the cleanup task will start using new methods for searching the users
and groups which have to be cleaned up, SSSD starts relying more in a
more consistent state of the timestamp cache on pretty much everything
related to the cleanup task.

One of the things that would cause SSSD some problems is not having the
ghost user expired in the persistent cache but not in the timestamp
cache.

With this patch, the entry is also expired in the timestamp cache when
it's present.

Related:
https://pagure.io/SSSD/sssd/issue/3369

Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>

Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
(cherry picked from commit 9883d1e2913ff0c1db479f1ece8148e03155c7f3)
---
 src/db/sysdb_ops.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/src/db/sysdb_ops.c b/src/db/sysdb_ops.c
index 12f8095d2edc60ffab09c92d64f968892c577bbf..ae26470487f859fe1de1dc364b6a05b9793a0545 100644
--- a/src/db/sysdb_ops.c
+++ b/src/db/sysdb_ops.c
@@ -5065,6 +5065,15 @@ errno_t sysdb_mark_entry_as_expired_ldb_dn(struct sss_domain_info *dom,
         goto done;
     }
 
+    if (dom->sysdb->ldb_ts != NULL) {
+        ret = ldb_modify(dom->sysdb->ldb_ts, msg);
+        if (ret != LDB_SUCCESS) {
+            DEBUG(SSSDBG_MINOR_FAILURE,
+                  "Could not mark an entry as expired in the timestamp cache\n");
+            /* non-fatal */
+        }
+    }
+
     ret = EOK;
 
 done:
-- 
2.9.4