wrappers/systemd.template.sysconfig | 3 +++
1 file changed, 3 insertions(+)
New commits:
commit 5e810f82fd36d606f48d9a1218d2b8937bdf04ef
Author: Mark Reynolds <mreynolds(a)redhat.com>
Date: Tue Jul 26 15:31:32 2016 -0400
Ticket 48935 - Update dirsrv.systemd file
BUg Description: Two issues here. First the default system startup timeout
is set to 1 minute 30 seconds. But in the start-dirsrv script
it attempts to use a 10 minute timeout.
Second, starting in F23 systemd does not work well with valgrind.
systemd does not accept the notification once the DS starts.
So the start command actually fails after it times out.
Fix Description: For the first issue set the system startup timeout to match the
start-dirsrv script by setting:
TimeoutStartSecs=10min
Second, allow valgrind's startup success message to be
recognized/accepted we set:
NotifyAccess=all
https://fedorahosted.org/389/ticket/48935
Reviewed by: nhosoi(Thanks!)
(cherry picked from commit ce44176803aa52ab8001113136bfbb7ff4a50972)
diff --git a/wrappers/systemd.template.sysconfig b/wrappers/systemd.template.sysconfig
index d88bdcd..e653c86 100644
--- a/wrappers/systemd.template.sysconfig
+++ b/wrappers/systemd.template.sysconfig
@@ -1,3 +1,6 @@
[Service]
+TimeoutStartSec=10m
+NotifyAccess=all
+
# uncomment this line to raise the file descriptor limit
# LimitNOFILE=8192
wrappers/systemd.template.sysconfig | 3 +++
1 file changed, 3 insertions(+)
New commits:
commit ce44176803aa52ab8001113136bfbb7ff4a50972
Author: Mark Reynolds <mreynolds(a)redhat.com>
Date: Tue Jul 26 15:31:32 2016 -0400
Ticket 48935 - Update dirsrv.systemd file
BUg Description: Two issues here. First the default system startup timeout
is set to 1 minute 30 seconds. But in the start-dirsrv script
it attempts to use a 10 minute timeout.
Second, starting in F23 systemd does not work well with valgrind.
systemd does not accept the notification once the DS starts.
So the start command actually fails after it times out.
Fix Description: For the first issue set the system startup timeout to match the
start-dirsrv script by setting:
TimeoutStartSecs=10min
Second, allow valgrind's startup success message to be
recognized/accepted we set:
NotifyAccess=all
https://fedorahosted.org/389/ticket/48935
Reviewed by: nhosoi(Thanks!)
diff --git a/wrappers/systemd.template.sysconfig b/wrappers/systemd.template.sysconfig
index d88bdcd..e653c86 100644
--- a/wrappers/systemd.template.sysconfig
+++ b/wrappers/systemd.template.sysconfig
@@ -1,3 +1,6 @@
[Service]
+TimeoutStartSec=10m
+NotifyAccess=all
+
# uncomment this line to raise the file descriptor limit
# LimitNOFILE=8192
ldap/servers/slapd/opshared.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
New commits:
commit c8e7fc567ca9b00969e3911bf3aa6245ebdffe5b
Author: Mark Reynolds <mreynolds(a)redhat.com>
Date: Tue Jul 19 18:28:06 2016 -0400
Ticket 48930 - Paged result search can hang the server
Bug Description: If a pages result has sent all its entry it's possible that
the connection lock/monitor is not exited. This can lead
to a deadlock when a new connection comes in.
Fix Description: Move the monitor exit to the proper location, and also set
the result to success if all the entries have been sent.
https://fedorahosted.org/389/ticket/48930
Reviewed by: nhosoi(Thanks!)
(cherry picked from commit bff1dd433f5de807ded1114e5961d66e2c0494e2)
diff --git a/ldap/servers/slapd/opshared.c b/ldap/servers/slapd/opshared.c
index 84e4c71..93334a1 100644
--- a/ldap/servers/slapd/opshared.c
+++ b/ldap/servers/slapd/opshared.c
@@ -694,11 +694,12 @@ op_shared_search (Slapi_PBlock *pb, int send_result)
/* search result could be reset in the backend/dse */
slapi_pblock_get(pb, SLAPI_SEARCH_RESULT_SET, &sr);
pagedresults_set_search_result(pb->pb_conn, operation, sr, 1/*locked*/, pr_idx);
- PR_ExitMonitor(pb->pb_conn->c_mutex);
}
} else {
pr_stat = PAGEDRESULTS_SEARCH_END;
+ rc = LDAP_SUCCESS;
}
+ PR_ExitMonitor(pb->pb_conn->c_mutex);
pagedresults_unlock(pb->pb_conn, pr_idx);
if ((PAGEDRESULTS_SEARCH_END == pr_stat) || (0 == pnentries)) {