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

Re: to those who want to support Debian longer...



On 08/29/2013 07:18 PM, Holger Levsen wrote:
> Hi,
> 
> please start with helping supporting the current stable release better:
> 
> http://udd.debian.org/bugs.cgi?release=wheezy&rc=1 shows 255 RC bugs in 
> wheezy, just four months after this counter was basically at zero.

Hi,

As much as I would like to have this number of problem go down, this has
nothing to do with *security* support of the very-old-stable distro as
we discussed. Besides this, the plan is to support *some* of the
very-old-stable for security, and to only track problems for the
packages which nobody cares. At least *I* do not intend to support
absolutely all of the packages.

For example, taking examples from your list, if we were talking about
Squeeze EOL, I wouldn't care about python-pip and python-virtualenv,
though I could care more about other python modules which I would use in
production (no example here, sorry).

That being said, I also would like to highlight the fact that it's quite
difficult to convince the release team that all of these bugs needs to
be fixed. For example (on one of my packages, so please don't use a
stick to beat me on it, and stick to the current discussion), see
#710507. I have made a fix for it, as well as many others in the same
source package, but no reply from the release team for more than 20 days
(see #719632 for the p-u discussion). I don't think this is an isolated
case. Though that's annoying that it is stalled, because there's a CVE
included in the list of fixes.

I'm in no way complaining about the release team. I know they are busy,
that they try to minimize problems introduced by changes, and that peer
review takes a lot of time. Though the fact is, because of it, some
people (not including myself) have completely (and understandably) given
up updating stable, and the list of bug you see there could be partly
due to it.

I'm not sure what is the way forward. On one side, relaxing the rules to
get fixes in could be dangerous, and on the other hand, blocking fixes
isn't nice. I'd vote for a bit more relaxed policy, but I would
understand anyone thinking otherwise.

Cheers,

Thomas

P.S: Also (and with all due respect I have for someone like you, Holger)
please avoid pointing fingers at a group of people that have the will to
do things. This is counter productive.


Reply to: