Bug#929527: [pkg-netfilter-team] Bug#929527: Bug#929527: Bug#914694

2019-06-26 Thread Arturo Borrero Gonzalez
Control: severity -1 important

On 6/26/19 2:28 PM, Thomas Lamprecht wrote:
> 
> Hmm, but that's a grave issue which may just render the firewall void
> for _any_ intermediate chain and produces segmentation faults errors.
> 

The issue you found is not a general-case issue.
The segfault is only produced apparently if you:

* define a custom chain
* flush all rules of that custom chain (not required, because the chain was just
created)
* add a rule to that custom chain

all in the same batch.

I may understand that this is important for some scripts or robots making use of
the iptables interface in that particular way, but is not the general case of
how people define and add rules to custom chain/ruleset.
Because of this, I think we should lower the severity of this bug.

I understand is annoying in your use case, and I'm sorry for that.
Thankfully, we already have an iptables version fixing the issue, but
unfortunately it won't make it to Debian Buster in the first round as I already
explained in my previous email.

> How about a minimal patch which places higher update-alternative priority
> to the the -legacy parts of iptables so that the alternative currently
> working in Buster is used by default. Once the fixed nft based is rolled
> out the priorities could then be switched again (or if that cannot be done
> for a stable release, in Bullseye).
> 

No, sorry, we won't do this at this point.



Processed: Re: [pkg-netfilter-team] Bug#929527: Bug#929527: Bug#914694

2019-06-26 Thread Debian Bug Tracking System
Processing control commands:

> severity -1 important
Bug #929527 [iptables] /usr/sbin/xtables-nft-multi: restoring IP Tables with an 
self-defined chain segfaults in libnftnl.so
Severity set to 'important' from 'grave'

-- 
929527: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=929527
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems



Bug#929527: [pkg-netfilter-team] Bug#929527: Bug#914694

2019-06-26 Thread Thomas Lamprecht
On 6/26/19 2:14 PM, Arturo Borrero Gonzalez wrote:
> On 6/25/19 10:25 AM, Thomas Lamprecht wrote:
>> Don't want to nag to much but is there any news regarding this?
>> Buster is planned to release pretty soon (<2 weeks) and iptables
>> is quite a important package, IMO. Maybe it went under my radar
>> but I saw no unblock request on d.o release list.
>>
>> For now I just used update-alternative to use the legacy variants,
>> which work fine here, but if my understanding is correct then this
>> package (version?) could be thrown out of Buster if it still has RC
>> bug so close to the planned release, I mean iptables may be an
>> exception as it's quite relevant and still used by a lot but still.
>>
> 
> The last upstream release of iptables won't make it into Debian Buster at this
> point.
> 
> Once buster is released I will:
> 
> * provide uptodate package backports of newer upstream releases in
> buster-backports (for both iptables and nftables)
> * for important bugs, I would try backporting concrete patches to the version 
> in
> buster-stable.
> 
> 

Hmm, but that's a grave issue which may just render the firewall void
for _any_ intermediate chain and produces segmentation faults errors.

How about a minimal patch which places higher update-alternative priority
to the the -legacy parts of iptables so that the alternative currently
working in Buster is used by default. Once the fixed nft based is rolled
out the priorities could then be switched again (or if that cannot be done
for a stable release, in Bullseye).



Bug#929527: [pkg-netfilter-team] Bug#929527: Bug#914694

2019-06-26 Thread Arturo Borrero Gonzalez
On 6/25/19 10:25 AM, Thomas Lamprecht wrote:
> Don't want to nag to much but is there any news regarding this?
> Buster is planned to release pretty soon (<2 weeks) and iptables
> is quite a important package, IMO. Maybe it went under my radar
> but I saw no unblock request on d.o release list.
> 
> For now I just used update-alternative to use the legacy variants,
> which work fine here, but if my understanding is correct then this
> package (version?) could be thrown out of Buster if it still has RC
> bug so close to the planned release, I mean iptables may be an
> exception as it's quite relevant and still used by a lot but still.
> 

The last upstream release of iptables won't make it into Debian Buster at this
point.

Once buster is released I will:

* provide uptodate package backports of newer upstream releases in
buster-backports (for both iptables and nftables)
* for important bugs, I would try backporting concrete patches to the version in
buster-stable.



Bug#929527: Bug#914694

2019-06-25 Thread Thomas Lamprecht
Don't want to nag to much but is there any news regarding this?
Buster is planned to release pretty soon (<2 weeks) and iptables
is quite a important package, IMO. Maybe it went under my radar
but I saw no unblock request on d.o release list.

For now I just used update-alternative to use the legacy variants,
which work fine here, but if my understanding is correct then this
package (version?) could be thrown out of Buster if it still has RC
bug so close to the planned release, I mean iptables may be an
exception as it's quite relevant and still used by a lot but still.



Bug#929527: Bug#914694

2019-06-01 Thread Arturo Borrero Gonzalez
iptables 1.8.3-1~exp1 is already uploaded, currently waiting in the NEW queue.
The upload is for experimental, since the build depends on a newer release of
libnftnl (already in experimental), and we are at a point of the Buster freeze
that I would like to make extra sure I'm allowed to push such a big diff into
unstable.

Hopefully all this will be untangled in the upcoming weeks.