From 653df698a7a04c40df13eb4217c7d598aba8f8f8 Mon Sep 17 00:00:00 2001
From: Alexey Tikhonov <atikhono@redhat.com>
Date: Mar 18 2020 12:24:26 +0000
Subject: Watchdog: fixes "off-by-one" error


'man sssd.conf': timeout: "Note that after three missed heartbeats
the process will terminate itself."

But implementation was:
```
\#define WATCHDOG_MAX_TICKS 3
...
    if (__sync_add_and_fetch(&watchdog_ctx.ticks, 1) > WATCHDOG_MAX_TICKS) {
        ...
        _exit(1);
```
  -- since after reset ticks start from 0 effectively this was 4 heartbeats.

Fixed to match man page.

Resolves: https://pagure.io/SSSD/sssd/issue/4169

Reviewed-by: Pavel Březina <pbrezina@redhat.com>

---

diff --git a/src/util/util_watchdog.c b/src/util/util_watchdog.c
index 38c2482..a82014c 100644
--- a/src/util/util_watchdog.c
+++ b/src/util/util_watchdog.c
@@ -71,7 +71,7 @@ static void watchdog_handler(int sig)
     watchdog_detect_timeshift();
 
     /* if a pre-defined number of ticks passed by kills itself */
-    if (__sync_add_and_fetch(&watchdog_ctx.ticks, 1) > WATCHDOG_MAX_TICKS) {
+    if (__sync_add_and_fetch(&watchdog_ctx.ticks, 1) >= WATCHDOG_MAX_TICKS) {
         if (getpid() == getpgrp()) {
             kill(-getpgrp(), SIGTERM);
         }