lighttpd 1.4.59 is part of Ubuntu 21.04 (Hirsute Hippo)
no reason to keep this open for years if the Ubuntu lighttpd package
maintainers are never going to get around to backporting the changes
** Changed in: lighttpd (Ubuntu)
Status: Fix Committed => Fix Released
--
You received this bu
See Bug #1721635
https://bugs.launchpad.net/ubuntu/+source/lighttpd/+bug/1721635
** Changed in: lighttpd (Ubuntu)
Status: In Progress => Fix Committed
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/b
Well, the issue is still not resolved in Ubuntu 16.04 and probably 18.04
as well. And 16.04 is still supported. Let's wait until 16.04 support
ends and close this instead of fixing or backporting new version.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
lighttpd 1.4.46 was released in Oct 2017, almost 3 years ago.
Is there any reason this is still open?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd
** Tags removed: bitesize
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
To manage notifications about this bug go to:
https://bugs.launc
** Tags removed: server-next server-next-drop
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
To manage notifications about this bug go to
** Tags added: server-next-drop
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
To manage notifications about this bug go to:
https://bugs
Hmm, I was wrong. Actually, sysv script in Debian Wheezy was using TERM
for stop (and thus for restart as well), and INT for reload and force-
reload. So yes, you were right, there was a way to kill it ungracefully.
But documentation never suggested anything but force-reload to apply new
settings,
Actually this wasn't the case with sysv script. If you remember, sysv
script was the first thing I attached patch for, because it was
obviously bugged: both --signal and change --retry options given, so a
signal given with --signal was simply ignored. If you check sysv script
in Debian, at least in
I would like the opinion of more people, yes. I fear we are changing the
behavior of the restart action. Before it would be immediate (tearing
down live connections), whereas with the fix it is transformed into a
graceful-restart. Normally services have these two options separate so
that the admin
So, what is going on with this bug? Do more people need to test the
package before it goes into repository?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lig
Test packages are building here for xenial, zesty and artful:
https://launchpad.net/~ahasenack/+archive/ubuntu/lighttpd-
reload-1707312-1721635
What they do:
a) remove the "reload" action from the SysV initscript
b) apply the systemd service file changes suggested here
Expected outcome:
- slight
Filed https://redmine.lighttpd.net/issues/2825 upstream
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
To manage notifications about this
>From systemd documentation:
"If set to control-group, all remaining processes in the control group
of this unit will be killed on unit stop (for services: after the stop
command is executed, as configured with ExecStop=)."
We don't want that. We want lighttpd to care about spawned fcgi
processes
Can you comment on why KillMode needs to be set to "mixed"? The default
is control-group. The sysv script also only seems to kill the main
process (which could be a bug, granted). Is it because of CGI scripts?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
** Changed in: lighttpd (Ubuntu)
Status: Triaged => In Progress
** Changed in: lighttpd (Ubuntu)
Importance: Undecided => Medium
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
Yeah, CanReload will only be true if the service has a ExecReload config. I was
toying with this for a bit:
[Service]
...
ExecReload=/bin/kill -HUP $MAINPID
It works, the reload from the sysv script is ignored. But this isn't a
proper reload, it's just about reopening the logfiles, so I guess we
Nice! "KillMode=mixed" is important here, though.
Regarding "reload", in /lib/lsb/init-functions.d/40-systemd included
from /lib/lsb/init-functions included from /etc/init.d/lighttpd there is
this code:
if [ "$(systemctl -p CanReload show $service 2>/dev/null)" = "CanReload=no" ]
&& [ "${1:-}" =
The moment we have a reload action in systemd (via ExecReload), then
"service lighttpd reload" will call that instead of the misnamed and
broken one from the sysv script. We are changing behavior of the reload
action, but I think we have to: we are changing it from a broken restart
to an actually w
Haha, I had almost the same thing :)
[Service]
ExecStartPre=/usr/sbin/lighttpd -t -f /etc/lighttpd/lighttpd.conf
ExecStart=/usr/sbin/lighttpd -D -f /etc/lighttpd/lighttpd.conf
KillSignal=SIGINT
TimeoutStopSec=10
I was using 10s for testing :)
That essentially makes any action that involves a "st
Ok, I'm really sorry for spamming, but here's a proper fix for systemd's
lighttpd.service. After these changes command `systemctl force-reload
lighttpd.service` does a proper graceful reastart. It will also kill all
children after 60 seconds in case something goes wrong (for example, if
lighttpd fa
I see. Can you just fix systemctl's force-reload action to send SIGINT
instead of SIGTERM? I am not familiar with systemd and couldn't find
where it's defined.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/
https://bugs.launchpad.net/ubuntu/+source/lighttpd/+bug/1721635
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
To manage notifications ab
Ok, sorry for spamming on this bug, but I'm not happy with that reload
action, fixed or not. Everything breaks after it's issued, it's not just
status: stop, start and restart also don't work anymore. I'm having a
hard time overlooking that. I'll file a separate bug.
--
You received this bug noti
Yes, if all is done via systemd, then status after reload works. But
that change is too much for an SRU I think.
That being said, status after reload is already broken, so we are not
making things worse by fixing that reload action.
--
You received this bug notification because you are a member
I believe it should all be done via systemd instead. Less work in the
future when sysv will be deprecated. But as a quick solution an obvious
bug should be fixed right now.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.l
Ok, so I just need to sort out the fact that "status" doesn't work after
that "reload" (because there is a new process, and the reload was done
via the sysv script while the status is done via systemd). One could
argue this is a different bug, though.
--
You received this bug notification because
Ah, nice. So "force-reload" is the right name, but for some reason
"reload" is aliased to it.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd graceful
"reload" and "force-reload" are defined in
https://www.debian.org/doc/debian-policy/#writing-the-scripts
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lightt
The only way to reload configuration is actually through a graceful
restart, so that's why they called it "force-reload", I guess. I don't
know when "force-reload" and "reload" became aliases, though. Renaming
might be a good idea, but then /usr/sbin/lighty-{enable,disable}-mod
scripts should be up
xenial is a confusing case, at least it was to me. The package has both
a sysv initscript and a systemd service file.
They don't have the same action parameters, though. The systemd service
has just start/stop/restart (which is stop+start). The sysv initscript
has a lot more.
Depending on what ac
"reload" the way it is implemented should be called something else:
"graceful-restart" probably, because that's what it is (even in the
source code).
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/17073
The analysis is correct and easy to reproduce. Just keep a connection
open to the server and issue the reload. Without the fix, the connection
is terminated immediately. With the fix, the timeouts are respected, and
if the connection is closed before the timeout is reached, the reload
action contin
** Changed in: lighttpd (Ubuntu)
Assignee: (unassigned) => Andreas Hasenack (ahasenack)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefull
Andreas, could you validate this and convert it into a suitable form for
upload, please?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefully
T
Since Ubuntu Review Team is ignoring this ticket, I subscribed Robie
Basak and Artur Rona. Hope you guys can help.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut d
Hello? Anyone here? If this package is not supported, please remove it
from repository then.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload does not shut down lighttpd gracefull
Can someone please review and accept patch? This is pretty-much critical
issue for those using lighttpd as a primary web server.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1707312
Title:
reload d
The attachment "lighttpd.diff" seems to be a patch. If it isn't, please
remove the "patch" flag from the attachment, remove the "patch" tag, and
if you are a member of the ~ubuntu-reviewers, unsubscribe the team.
[This is an automated message performed by a Launchpad user owned by
~brian-murray,
After reading man start-stop-daemon I found the problem.
"... schedule is a list of at least two items separated by slashes (/);
each item may be -signal-number or [-]signal-name, which means to send
that signal, or timeout, which means to wait that many seconds for
processes to exit ..."
"If a s
** Description changed:
I noticed that /etc/init.d/lighttpd reload kills lighttpd almost
instantly. I can manually shut down lighttpd gracefully by sending it
SIGINT, so this is an issue somewhere within init.d script.
Here's log message when I call init.d reload:
- 2017-07-28 22:54:
41 matches
Mail list logo
