[shepherd] 13/13: herd: Report startup failure.

[Ludovic Courtès]

civodul pushed a commit to branch master
in repository shepherd.

commit fbca4e20a9e14a3024a02ee9d3bf6292e3fb99c2
Author: Ludovic Courtès <ludo@gnu.org>
AuthorDate: Sun Apr 16 22:48:45 2023 +0200

    herd: Report startup failure.
    
    * modules/shepherd/scripts/herd.scm (display-service-status): Check
    'startup-failures'.  When the service is stopped and STARTUP-FAILURES is
    a pair, mark it as failing and report the startup failure time.
    * tests/startup-failure.sh: New file.
    * Makefile.am (TESTS): Add it.
---
 Makefile.am                       |  1 +
 modules/shepherd/scripts/herd.scm | 16 ++++++++--
 tests/startup-failure.sh          | 61 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 75 insertions(+), 3 deletions(-)

diff --git a/Makefile.am b/Makefile.am
index 6eba8d9..c6f0c70 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -242,6 +242,7 @@ TESTS =                                             \
   tests/basic.sh                               \
   tests/starting-status.sh                     \
   tests/stopping-status.sh                     \
+  tests/startup-failure.sh                     \
   tests/replacement.sh                         \
   tests/respawn.sh                             \
   tests/respawn-throttling.sh                  \
diff --git a/modules/shepherd/scripts/herd.scm 
b/modules/shepherd/scripts/herd.scm
index cc82f61..998b5d7 100644
--- a/modules/shepherd/scripts/herd.scm
+++ b/modules/shepherd/scripts/herd.scm
@@ -91,7 +91,8 @@ of pairs."
   (match service
     (('service ('version 0 _ ...) properties ...)
      (alist-let* properties (provides requires status running respawn? enabled?
-                             last-respawns one-shot? transient?)
+                             last-respawns startup-failures
+                             one-shot? transient?)
        (format #t (l10n "Status of ~a:~%") (first provides))
 
        ;; Note: Shepherd up to 0.9.x included did not provide 'status', hence
@@ -109,7 +110,9 @@ of pairs."
          ('stopped
           (if one-shot?
               (format #t (l10n "  It is stopped (one-shot).~%"))
-              (format #t (l10n "  It is stopped.~%"))))
+              (if (pair? startup-failures)
+                  (format #t (l10n "  It is stopped (failing).~%"))
+                  (format #t (l10n "  It is stopped.~%")))))
          ('starting
           (format #t (l10n "  It is starting.~%")))
          ('stopping
@@ -130,7 +133,14 @@ of pairs."
           (format #t (l10n "  Last respawned on ~a.~%")
                   (date->string
                    (time-utc->date (make-time time-utc 0 time)))))
-         (_ #t))))))
+         (_ #t))
+       (when (or (eq? status 'stopped) (not running))
+         (match startup-failures
+           ((time _ ...)
+            (format #t (l10n "  Failed to start at ~a.~%")
+                    (date->string
+                     (time-utc->date (make-time time-utc 0 time)))))
+           (_ #t)))))))
 
 (define root-service?
   ;; XXX: This procedure is written in a surprising way to work around a
diff --git a/tests/startup-failure.sh b/tests/startup-failure.sh
new file mode 100644
index 0000000..9351ca8
--- /dev/null
+++ b/tests/startup-failure.sh
@@ -0,0 +1,61 @@
+# GNU Shepherd --- Check service startup failure reporting
+# Copyright © 2023 Ludovic Courtès <ludo@gnu.org>
+#
+# This file is part of the GNU Shepherd.
+#
+# The GNU Shepherd is free software; you can redistribute it and/or modify it
+# under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or (at
+# your option) any later version.
+#
+# The GNU Shepherd is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with the GNU Shepherd.  If not, see <http://www.gnu.org/licenses/>.
+
+shepherd --version
+herd --version
+
+socket="t-socket-$$"
+conf="t-conf-$$"
+log="t-log-$$"
+pid="t-pid-$$"
+stamp="t-stamp-$$"
+
+herd="herd -s $socket"
+
+trap "rm -f $socket $conf $log $stamp;
+      test -f $pid && kill \`cat $pid\` || true; rm -f $pid" EXIT
+
+cat > "$conf" <<EOF
+(register-services
+ (service
+   '(may-fail)
+   #:start (lambda _
+             (file-exists? "$PWD/$stamp"))
+   #:respawn? #f))
+EOF
+
+rm -f "$pid" "$stamp" "$socket"
+shepherd -I -s "$socket" -c "$conf" --pid="$pid" --log="$log" &
+
+while ! test -f "$pid"; do sleep 0.5 ; done
+
+# When the service fails to start, 'herd status' should display that.
+if $herd start may-fail; then false; else true; fi
+$herd status may-fail | grep stopped
+$herd status may-fail | grep "Failed to start"
+
+touch "$stamp"
+$herd start may-fail
+$herd status may-fail | grep started
+$herd status may-fail | grep -v "Failed to start"
+
+# Once the service has been stopped gracefully, the "Failed to start" message
+# should not appear any longer.
+$herd stop may-fail
+$herd status may-fail | grep stopped
+$herd status may-fail | grep -v "Failed to start"