Package: release.debian.org Severity: normal Tags: jessie User: release.debian.org@packages.debian.org Usertags: pu Hello, I just uploaded systemd 215-17+deb8u1 to stable, with essentially the same contents/fixes as in 215-18 to unstable [1]. Annotated changelog: | [ Michael Biebl ] | * manager: Pass correct errno to strerror(), have_ask_password contains | negative error values which have to be negated when being passed to | strerror(). This is a leftover from a pre-jessie release team review, during that this little error got spotted. It just affects correct error message reporting. I can search the reference to that where it was reviewed/ack'ed if you care, but I suppose it's straightforward enough. | [ Martin Pitt ] | * Revert upstream commit 743970d which immediately SIGKILLs units during | shutdown. This leads to problems like bash not being able to write its | history, mosh not saving its state, and similar failed cleanup actions. | (Closes: #784720, LP: #1448259) Requested by Henrique and Niels, and quite a severe data loss condition. | * write_net_rules: Escape '{' and '}' characters as well, to make this work | with busybox grep. Thanks Faidon Liambotis! (Closes: #765577) RC bug with minimally invasive fix. (Aside from the jessie-ignore RC bug #780650 which will be fixed with 220-1, this is the only current RC bug) | * debian/gbp.conf: Point to jessie branch. git-buildpackage bureaucracy, no impact on builds or the binaries. debdiff is attached. It should be quite obvious which hunk applies to which change, but please don't hesitate to ask. Thanks, Martin [1] https://tracker.debian.org/news/685999 -- Martin Pitt | http://www.piware.de Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
diff -Nru systemd-215/debian/changelog systemd-215/debian/changelog --- systemd-215/debian/changelog 2015-04-16 17:26:58.000000000 +0200 +++ systemd-215/debian/changelog 2015-05-26 07:56:09.000000000 +0200 @@ -1,3 +1,21 @@ +systemd (215-17+deb8u1) stable; urgency=medium + + [ Michael Biebl ] + * manager: Pass correct errno to strerror(), have_ask_password contains + negative error values which have to be negated when being passed to + strerror(). + + [ Martin Pitt ] + * Revert upstream commit 743970d which immediately SIGKILLs units during + shutdown. This leads to problems like bash not being able to write its + history, mosh not saving its state, and similar failed cleanup actions. + (Closes: #784720, LP: #1448259) + * write_net_rules: Escape '{' and '}' characters as well, to make this work + with busybox grep. Thanks Faidon Liambotis! (Closes: #765577) + * debian/gbp.conf: Point to jessie branch. + + -- Martin Pitt <mpitt@debian.org> Tue, 26 May 2015 07:55:44 +0200 + systemd (215-17) unstable; urgency=high * cryptsetup: Implement offset and skip options. (Closes: #751707, diff -Nru systemd-215/debian/extra/write_net_rules systemd-215/debian/extra/write_net_rules --- systemd-215/debian/extra/write_net_rules 2015-04-16 17:26:58.000000000 +0200 +++ systemd-215/debian/extra/write_net_rules 2015-05-26 07:56:09.000000000 +0200 @@ -118,7 +118,7 @@ match="$match, KERNEL==\"$basename*\"" # build a regular expression that matches the new rule that we want to write -new_rule_pattern=$(echo "^SUBSYSTEM==\"net\", ACTION==\"add\"$match" | sed -re 's/([\?\*])/\\\1/g') +new_rule_pattern=$(echo "^SUBSYSTEM==\"net\", ACTION==\"add\"$match" | sed -re 's/([\?\*\{\}])/\\\1/g') # Double check if the new rule has already been written. This happens if # multiple add events are generated before the script returns and udevd diff -Nru systemd-215/debian/gbp.conf systemd-215/debian/gbp.conf --- systemd-215/debian/gbp.conf 2015-04-16 17:26:58.000000000 +0200 +++ systemd-215/debian/gbp.conf 2015-05-26 07:56:09.000000000 +0200 @@ -1,4 +1,4 @@ [DEFAULT] pristine-tar = True patch-numbers = False -debian-branch = master +debian-branch = jessie diff -Nru systemd-215/debian/patches/manager-pass-correct-errno-to-strerror.patch systemd-215/debian/patches/manager-pass-correct-errno-to-strerror.patch --- systemd-215/debian/patches/manager-pass-correct-errno-to-strerror.patch 1970-01-01 01:00:00.000000000 +0100 +++ systemd-215/debian/patches/manager-pass-correct-errno-to-strerror.patch 2015-05-26 07:56:09.000000000 +0200 @@ -0,0 +1,27 @@ +From 6d025bafdb9a3317fac014bdf19acf09256a1531 Mon Sep 17 00:00:00 2001 +From: Michael Biebl <biebl@debian.org> +Date: Thu, 16 Apr 2015 13:56:28 +0200 +Subject: [PATCH] manager: pass correct errno to strerror() + +have_ask_password contains negative error values which have to be +negated when being passed to strerror(). +--- + src/core/manager.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/core/manager.c b/src/core/manager.c +index 47e23ba..06ef376 100644 +--- a/src/core/manager.c ++++ b/src/core/manager.c +@@ -241,7 +241,7 @@ static int manager_dispatch_ask_password_fd(sd_event_source *source, + if (m->have_ask_password < 0) + /* Log error but continue. Negative have_ask_password + * is treated as unknown status. */ +- log_error("Failed to list /run/systemd/ask-password: %s", strerror(m->have_ask_password)); ++ log_error("Failed to list /run/systemd/ask-password: %s", strerror(-m->have_ask_password)); + + return 0; + } +-- +2.1.4 + diff -Nru systemd-215/debian/patches/Revert-core-one-step-back-again-for-nspawn-we-actual.patch systemd-215/debian/patches/Revert-core-one-step-back-again-for-nspawn-we-actual.patch --- systemd-215/debian/patches/Revert-core-one-step-back-again-for-nspawn-we-actual.patch 1970-01-01 01:00:00.000000000 +0100 +++ systemd-215/debian/patches/Revert-core-one-step-back-again-for-nspawn-we-actual.patch 2015-05-26 07:56:09.000000000 +0200 @@ -0,0 +1,36 @@ +From: Martin Pitt <martin.pitt@ubuntu.com> +Date: Mon, 27 Apr 2015 15:29:13 +0200 +Subject: Revert "core: one step back again, + for nspawn we actually can't wait for cgroups running empty since + systemd will get exactly zero notifications about it" + +This reverts commit 743970d2ea6d08aa7c7bff8220f6b7702f2b1db7. + +Bug-Debian: https://bugs.debian.org/784720 +Bug-Ubuntu: https://launchpad.net/bugs/1448259 +Bug-Fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1141137 +--- + src/core/unit.c | 10 +--------- + 1 file changed, 1 insertion(+), 9 deletions(-) + +diff --git a/src/core/unit.c b/src/core/unit.c +index 536fd5e..d82d5d1 100644 +--- a/src/core/unit.c ++++ b/src/core/unit.c +@@ -3218,15 +3218,7 @@ int unit_kill_context( + log_warning_unit(u->id, "Failed to kill control group: %s", strerror(-r)); + } else if (r > 0) { + +- /* FIXME: For now, we will not wait for the +- * cgroup members to die, simply because +- * cgroup notification is unreliable. It +- * doesn't work at all in containers, and +- * outside of containers it can be confused +- * easily by leaving directories in the +- * cgroup. */ +- +- /* wait_for_exit = true; */ ++ wait_for_exit = true; + + if (c->send_sighup && !sigkill) { + set_free(pid_set); diff -Nru systemd-215/debian/patches/series systemd-215/debian/patches/series --- systemd-215/debian/patches/series 2015-04-16 17:26:58.000000000 +0200 +++ systemd-215/debian/patches/series 2015-05-26 07:56:09.000000000 +0200 @@ -149,6 +149,7 @@ logind-handle-runtime-dir-without-CAP_SYS_ADMIN.patch sd-bus-create-clean-error-when-a-property-Set-call-w.patch manager-do-not-print-anything-while-passwords-are-be.patch +manager-pass-correct-errno-to-strerror.patch ## Debian specific patches: Add-back-support-for-Debian-specific-config-files.patch @@ -205,3 +206,4 @@ syslog-Increase-max_dgram_qlen-by-pulling-in-systemd.patch Skip-filesystem-check-if-already-done-by-the-initram.patch cryptsetup-Implement-offset-and-skip-options.patch +Revert-core-one-step-back-again-for-nspawn-we-actual.patch
Attachment:
signature.asc
Description: Digital signature