Re: How to avoid stealth installation of systemd?
1. Could some competent person tell me the right way to tell apt that it should fail an upgrade rather than installing systemd? I guess I could make a dummy package that conflicts with systemd, but I'm sure there's a better way. I think you can just put Package: systemd Pin: origin "" Pin-Priority: -1 in your /etc/apt/preferences... -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/bb6c30a51d68aa2e2b773dc4f26d8...@yourcmc.ru
Re: Reverting to GNOME for jessie's default desktop
Hi, I'm not a Debian developer, just a Debian user, and I want to say that I was happy to see XFCE being the default DE. Just because it's small, classic and neutral DE - which GNOME 3 definitely isn't. I think XFCE is a better default... because I think it's not that uncommon for people to really dislike GNOME 3. But, generally speaking, I think that the whole debian-installer's "tasksel" is implemented inconveniently. You don't even know what you'll get after installing these very broadly defined "desktop environment", "web server", "database server", "mail server" tasks... It installs exim for mail server, but I never use exim, I use postfix; it installs postgresql, but I need mysql more often; same for gnome3 - I never use gnome3, I use either XFCE or KDE... So I have to either purge it after installation or check nothing in the installer, and install everything by hand from the console. Not a big deal of course, but for me it makes tasksel useless. Why debian-installer couldn't provide you with some choice? Of course if you're just using a CD/DVD image and have limited or no internet access you can't choose the DE that's not on your CD/DVD; but now there is no choice even when you use the netinstall image... -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/f434d21e55c9c7cea5c4ae2f3db0c...@yourcmc.ru
Re: Reverting to GNOME for jessie's default desktop
Can you qualify that statement? Especially the "neutral" part, it is not obvious to me how a DE can be, or not be "neutral". By "neutral" I've meant a DE without "dubious" solutions discussed in gnome3 flame wars all over the web. I don't really think we should also discuss it here, just because we don't really want to start another flame war :) I think XFCE is a better default... because I think it's not that uncommon for people to really dislike GNOME 3. Can you quantify that statement? What is "not that uncommon" in numbers (as in, when does something start to be "that" uncommon)? Doesn't have to be some sharp boundary, just a ballpark, as in > 1/10/20% of users…. Of course I have no real numbers! What I've said is just my personal opinion, based on that I've seen a lot of flame wars about GNOME 3 on the web, and a non-zero count of forks/derivatives of GNOME 3 all aiming at providing a "more classic" environment (MATE, Cinnamon, Consort, Budgie). I don't want to discuss the specific usability aspects of GNOME 3, but note there are NO such flame wars about XFCE :) And as I've already said, I would really prefer to see a choice in debian-installer than discuss what should be the default... -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/3edc26a8e7c1a3a3ed918936a8916...@yourcmc.ru
Re: Subversion 1.8.4
Yes, someone needs to review it, but before that the package tests need to be fixed so they a) run and b) pass, reliably and repeatably. After that, either get the package maintainers to discuss the new version, or find someone to take over the package if they no longer want to or can maintain it. Doing an NMU for a package to upgrade it to new upstream release should be done very carefully, and it should avoid to override the maintainers unless it's critical. I couldn't send a email to pe...@p12n.org - his mail server says "552 Rejected due to spam content (#5.7.1)". Are there any other ways of communicating to him? -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/fd676f1d7524e9f4c008dd273e8f5...@yourcmc.ru
Re: Subversion 1.8.4
Also I did serf update as Bug#716793, please dget http://www.mithril-linux.org/~henrich/debian/package/temp/serf_1.3.2-0.1.dsc and check it, too. I've actually already seen it... But it was after I've built it by myself... :) It seems your package has cleaner debian/rules so I would prefer it to be uploaded to Debian :) just 1 note: you have a typo - "overrdie_dh_auto_clean" :) -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/384c28dc11e5ab3251e674fc6011e...@yourcmc.ru
Nginx CVE-2013-4547
Hi! There is a nginx security advisory here: http://mailman.nginx.org/pipermail/nginx-announce/2013/000125.html Is a Debian security update expected to come out for it? -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/05e3cb67c40a8296b4691324d8ffe...@yourcmc.ru
Re: Nginx CVE-2013-4547
Is a Debian security update expected to come out for it? Yes. Nginx team has already submitted updated package to security team. Thanks for the information! I've actually found the bug stating this: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=730012 And what's the Debian process for such updates? Is it described somewhere? I.e. what are the next steps that are done with such updates? -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/6b7b9462839bae4ca2e03a4839be9...@yourcmc.ru
Re: Subversion 1.8.4
oh ;) thanks for your checking! 1 more note: scons doc states that if you want to install package in a "sandbox" for packaging, you should better supply real PREFIX=/usr, and then supply --install-sandbox=./debian/tmp option to 'scons install'. Don't know if it actually affects something in case of serf (I didn't test your build), but it seems that it's a more correct way... -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/f2067be80dda08498cf67ff550f87...@yourcmc.ru
Re: Bug#727708: Fsck SystemD and its developers and its users. GR to override this please.
- Debian has sent patches upstream
- Mageia is *much* smaller distribution, that packager has attended
*various* systemd hackfests
- Mageia package maintainer sent various patches upstream
- Patches are *not* accepted based on how many people you represent or
which company you work for (e.g. some Red Hat dude got a "no" during
hackfest before FOSDEM)
Glad to hear that! :)
QR codes is optional
Built-in HTTP server is optional
Binary logging - yeah, it logs stuff. Calling logging functionality in
a
program which is meant to log things is a bit much.
It's not just "calling logging functionality", it's "running an own
logging daemon" that supports a lot of other things. "calling logging
functionality" is syslog(3) or maybe even printf(3) :))) and it should
be enough for software that just "logs stuff". :)
You cannot separate it, if you look at status output of a service it'll
use the journal to give you useful output. Splitting things has
sideeffects. If the gain is uncertain of vague ("I don't like it" /
"UNIX philosophy"), then IMO time to trust the package maintainer.
Of course it has side effects! That's why people call it monolithic and
that's why it needs some decoupling. Do you agree that a modular program
is generally better than a non-modular one?
And yes, I can't say how much I don't like Windows-like ideas of binary
logs / binary configs / unified registry and etc. Yeah, yeah, I know
there are no binary configs in systemd, that's just an example. :)
And it seems I'm not the only one who doesn't like it! And I'm sure that
at least 50% of swear words addressed to systemd could be stopped at
once if the journal was made ALSO optional. So why not just do it?...
I'm not familiar with systemd code, but since it seems Debian faces
systemd becoming the default, I even want to try to make journal
optional by myself... :) the downside is that I don't know if systemd
developers will like it at all. :(
--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/7d88a7a010afdbb7ce8132d639bda...@yourcmc.ru
Re: Bug#727708: Fsck SystemD and its developers and its users. GR to override this please.
Because it's work, for no apparent gain. I mean, the systemd people didn't just code up all that journal stuff for no good reason, but because they perceived a need to have it. And let's face it, the ability to just see the stderr output from $FAILED_JOB with "systemctl status" is a whole damn lot better than to restart the thing in the foreground and hope to be able to reproduce the problem that caused it to die. You can split off systemd-journal and its supporting files into a separate binary package. That'd probably be quite simple. The question is, why would you even want to ..? Because I want logs to be plaintext in my system, not binary. And I don't see why a binary log format is needed to implement the stderr capture. -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/e2da4a65eec4579296f20f4cb4037...@yourcmc.ru
Re: Call to fork
Our users shouldn't care what init system we use. It's an implementation -- and purely technical -- detail of the OS. Sorry to interfere with your discussion, but it really sounds like some kind of proprietary software idea :) I'm sure a big percent of GNU/Linux (and especially Debian GNU/Linux) users like to know and understand what's going under the hood. Debian users are certainly not "average PC users". :) -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/14fceb839262a279e8d3b6cfa9d46...@yourcmc.ru
Re: Bug#727708: Fsck SystemD and its developers and its users. GR to override this please.
It is of course well-known that systemd developers like to make their life more complicated and love to implement binary formats instead of writing simple text parsers, just for the sake of having fun programming them, and absolutely not because they need things like indexing. The same goes for relational databases developers, for example. How silly of them all. 1) If you really need a binary index, it could be initially put in a separate file. 2) Binary index isn't needed at all if you just want to print output of a service - you can just put output of each unit to its own log file and just tail it. 3) If you don't want to print only last X lines, but want to print full output of a service since last start - you can remember the previous log position in the service state structure. 4) At a first glance I don't see any _real_ index (i.e. btree) implementation in systemd journal, so I assume it still does fullscans to print logs for a service - am I correct? 5) After all, I don't see why writing 1 regexp is a hard task. And it won't be really slower because of (4). DBMS is an incorrect example because DBMS is originally _meant_ to store and query structured data in different formats. -- With best regards, Vitaliy Filippov -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/6e0bd604d79e10a06ca5d08631300...@yourcmc.ru
Re: Bug#727708: Fsck SystemD and its developers and its users. GR to override this please.
Because I want logs to be plaintext in my system, not binary. Install syslog. Or maybe Debian will use both journal and syslog. I dislike the idea of binary logs so much that I want to really and totally disable journal. And I don't see why a binary log format is needed to implement the stderr capture. Try to find an efficient way to show the output of a particular daemon. Now of a cgroup. Now anything of a user. It's not about capturing, it is about doing something useful with it. You want to capture various properties with each message. No problem: one regexp, one more regexp, one more regexp. And if I _really_ needed a binary index, I would put it in a separate file. -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/e8c169d969d7e2f5b4d52aee88f20...@yourcmc.ru
