linux-kernel - Re: [PATCH] coredump_filter: enable to change the default filter

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <2f11576a0812260346j2df02362q52c8f813981a7773@mail.gmail.com>
Date:	Fri, 26 Dec 2008 20:46:03 +0900
From:	"KOSAKI Motohiro" <kosaki.motohiro@...fujitsu.com>
To:	"Hidehiro Kawai" <hidehiro.kawai.ez@...achi.com>
Cc:	"Roland McGrath" <roland@...hat.com>,
	"Andrew Morton" <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, yumiko.sugita.yf@...achi.com,
	satoshi.oshima.fk@...achi.com
Subject: Re: [PATCH] coredump_filter: enable to change the default filter

>>>>>However, we would need a special care for PID namespace if we use
>>>>>init's coredump_filter as default.  For a process with new PID
>>>>>namespace and new /proc, writing to /proc/1/coredump_filter doesn't
>>>>>mean changing default, although it's not a usual operation.
>>>>
>>>>I'd figured that private PID namespaces with their own init would want
>>>>their own private settings for this default too.
>>>
>>>Probably it's true, but I'm not sure if having coredump_filter have
>>>two means is acceptable.  init in a private PID namespace can die
>>>and dump a core file.  So its coredump_filter has two means; default
>>>setting for the namespace and setting for itself.
>>
>> FWIW, I think private namespace should have private settings framework.
>> I don't like to make special private settings code in this patch.
>>
>> Yeah, defenitely we need to private settings for various reason, I agreed.
>> but we also need consistent private settings rule and framework.
>> otherwise, we will make various non consistent crap. maybe..
>
> Anyway, as Andrew pointed out, current coredump filter doesn't
> care for user mode helpers.  So we need some sort of changes in
> coredump filter so that users can change user mode helpers'
> settings; use child_reapers setting as default (Roland's
> suggestion) or use a boot parameter.

hm, ok.
I agreed your patch is nice.

      Acked-by: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/