[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#839814: jessie-pu: package sympa/6.1.23~dfsg-2+deb8u1



Package: release.debian.org
Severity: normal
Tags: jessie
User: release.debian.org@packages.debian.org
Usertags: pu

Hi Stable release managers

Sympa is affected by a nasty bug when used under systemd, that after
logrotation with the original logrotate setup in jessie, sympa is in a
confused state. This is tracked as #804066 and was fixed in unstable
already with 6.2.16~dfsg-1.

I would like to propose that this is fixed as well for stable,
attached is the proposed debidff, where I needed to use
reload-or-try-reload (try-reload-or-restart in unstable).

Does this look acceptable for the upcoming jessie point release?

Regards,
Salvatore
diff -Nru sympa-6.1.23~dfsg/debian/changelog sympa-6.1.23~dfsg/debian/changelog
--- sympa-6.1.23~dfsg/debian/changelog	2015-01-16 02:48:24.000000000 +0100
+++ sympa-6.1.23~dfsg/debian/changelog	2016-10-05 12:22:20.000000000 +0200
@@ -1,3 +1,11 @@
+sympa (6.1.23~dfsg-2+deb8u1) jessie; urgency=medium
+
+  * Non-maintainer upload.
+  * Fix logrotate configuration so that sympa is not left in a confused state
+    when systemd is used (Closes: #804066)
+
+ -- Salvatore Bonaccorso <carnil@debian.org>  Wed, 05 Oct 2016 12:22:20 +0200
+
 sympa (6.1.23~dfsg-2) unstable; urgency=medium
 
   * Fix a bug in /etc/sympa/facility conffile handling (Closes: #774877)
diff -Nru sympa-6.1.23~dfsg/debian/sympa.logrotate sympa-6.1.23~dfsg/debian/sympa.logrotate
--- sympa-6.1.23~dfsg/debian/sympa.logrotate	2014-03-22 09:18:36.000000000 +0100
+++ sympa-6.1.23~dfsg/debian/sympa.logrotate	2016-10-05 12:22:20.000000000 +0200
@@ -7,7 +7,11 @@
 	delaycompress
 	create 640 sympa adm
 	postrotate
-		invoke-rc.d --quiet sympa reload > /dev/null
+		if [ -e /proc/1/exe ] && readlink -f /proc/1/exe | grep -q 'systemd' ; then
+			systemctl reload-or-try-restart sympa
+		else
+			invoke-rc.d --quiet sympa reload > /dev/null
+		fi
 		invoke-rc.d --quiet rsyslog rotate > /dev/null || true
 	endscript
 }

Reply to: