Tails

#1 Updated by BitingBird 2015-02-23 05:34:42

• related to Bug #8124: Scary failure messages when shutting down the system while the Unsafe Browser is still running added

#2 Updated by hybridwipe 2015-08-09 23:26:41

• File disable-running-cups-ekeyd-i2p-pcscd-rsyslog-speech.patch added

This patch stops several services from stopping on shutdown: cups, ekeyd, i2p, pcscd, rsyslog, speech-dispatcher and tor

#3 Updated by intrigeri 2015-08-18 05:41:26

• Assignee set to anonym
• QA Check set to Ready for QA

#4 Updated by BitingBird 2015-08-25 11:44:03

• Status changed from Confirmed to In Progress
• Target version set to Tails_1.6
• % Done changed from 0 to 30

Ready for QA -> next milestone

#5 Updated by intrigeri 2015-08-25 13:10:25

• Target version changed from Tails_1.6 to Tails_1.7

> Ready for QA -> next milestone

I don’t think that it’s suitable for a stabilization release.

#7 Updated by anonym 2015-10-07 08:24:46

• % Done changed from 30 to 50
• Feature Branch set to feature/6463-faster-shutdown

#8 Updated by anonym 2015-10-08 03:24:37

• Assignee changed from anonym to hybridwipe
• QA Check changed from Ready for QA to Info Needed

First of all, we already have two patches, config/chroot_local-patches/do_not_start_{tor,i2p}_on_boot.diff, for /etc/init.d/{tor,i2p} and due how you created your patches, and the order they are applied (lexical order) the patching fails => build failure. I fixed it by merging the relevant parts of your patches into ours.

Second, I’ve looked closely at many shutdowns with and without these patches applied, and I cannot notice any difference. I wonder what the point of these patches are if they do nothing from the users’ perspective (even if a laboratory testing would show that there are ~200 ms of improvement). I feel tempted to just drop this and reject this ticket — the patches will be irrelevant once we’re based on Debian Jessie due to systemd.

What do you think?

#9 Updated by hybridwipe 2015-10-13 08:10:37

• Assignee changed from hybridwipe to intrigeri

anonym wrote:
> First of all, we already have two patches, config/chroot_local-patches/do_not_start_{tor,i2p}_on_boot.diff, for /etc/init.d/{tor,i2p} and due how you created your patches, and the order they are applied (lexical order) the patching fails => build failure. I fixed it by merging the relevant parts of your patches into ours.
>
> Second, I’ve looked closely at many shutdowns with and without these patches applied, and I cannot notice any difference. I wonder what the point of these patches are if they do nothing from the users’ perspective (even if a laboratory testing would show that there are ~200 ms of improvement). I feel tempted to just drop this and reject this ticket — the patches will be irrelevant once we’re based on Debian Jessie due to systemd.
>
> What do you think?

Sure, closing is fine by me. I didn’t notice much difference either when testing myself. Might want to get @intrigeri’s opinion, as they originally reported it.

I only started working on this because there was a bug and it was easy enough to fix ;).

#10 Updated by intrigeri 2015-10-14 00:54:05

• Status changed from In Progress to Rejected

OK, let’s not do it then. Sorry for the noise and for wasting your time.

#11 Updated by anonym 2015-10-15 07:30:26

• Assignee deleted (intrigeri)
• % Done changed from 50 to 0
• QA Check deleted (Info Needed)
• Feature Branch deleted (feature/6463-faster-shutdown)

hybridwipe wrote:
> anonym wrote:
> > First of all, we already have two patches, config/chroot_local-patches/do_not_start_{tor,i2p}_on_boot.diff, for /etc/init.d/{tor,i2p} and due how you created your patches, and the order they are applied (lexical order) the patching fails => build failure. I fixed it by merging the relevant parts of your patches into ours.
> >
> > Second, I’ve looked closely at many shutdowns with and without these patches applied, and I cannot notice any difference. I wonder what the point of these patches are if they do nothing from the users’ perspective (even if a laboratory testing would show that there are ~200 ms of improvement). I feel tempted to just drop this and reject this ticket — the patches will be irrelevant once we’re based on Debian Jessie due to systemd.
> >
> > What do you think?
>
> Sure, closing is fine by me. I didn’t notice much difference either when testing myself. Might want to get @intrigeri’s opinion, as they originally reported it.
>
> I only started working on this because there was a bug and it was easy enough to fix ;).

Understood. I hope this won’t discourage you from further contributions! It was still worthwhile to investigate this, which is valuable in itself.

intrigeri wrote:
> OK, let’s not do it then.

I’ve removed the feature/6463-faster-shutdown branch, and reverted its commits from experimental.

> Sorry for the noise and for wasting your time.

Hey, like I said above it was worthwhile to investigate! :)