Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob

From	严海双 <yanhaishuang@cmss.chinamobile.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob
Date	2017-09-09 12:30 +0200
Message-ID	<unLbr-47m-3@gated-at.bofh.it> (permalink)
References	<unCKS-6S9-3@gated-at.bofh.it> <unFIJ-mk-5@gated-at.bofh.it> <unGls-P5-11@gated-at.bofh.it> <unGls-P5-9@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw


> On 2017年9月9日, at 下午1:16, David Miller <davem@davemloft.net> wrote:
> 
> From: 严海双 <yanhaishuang@cmss.chinamobile.com>
> Date: Sat, 9 Sep 2017 13:09:57 +0800
> 
>> 
>> 
>>> On 2017年9月9日, at 下午12:35, Cong Wang <xiyou.wangcong@gmail.com> wrote:
>>> 
>>> On Fri, Sep 8, 2017 at 6:25 PM, 严海双 <yanhaishuang@cmss.chinamobile.com> wrote:
>>>> 
>>>> 
>>>>> On 2017年9月9日, at 上午6:13, Cong Wang <xiyou.wangcong@gmail.com> wrote:
>>>>> 
>>>>> On Wed, Sep 6, 2017 at 8:10 PM, Haishuang Yan
>>>>> <yanhaishuang@cmss.chinamobile.com> wrote:
>>>>>> Different namespace application might require different maximal number
>>>>>> of TCP sockets independently of the host.
>>>>> 
>>>>> So after your patch we could have N * net->ipv4.sysctl_tcp_max_orphans
>>>>> in a whole system, right? This just makes OOM easier to trigger.
>>>>> 
>>>> 
>>>> From my understanding, before the patch, we had N * net->ipv4.sysctl_tcp_max_orphans,
>>>> and after the patch, we could have ns1.sysctl_tcp_max_orphans + ns2.sysctl_tcp_max_orphans
>>>> + ns3.sysctl_tcp_max_orphans, is that right? Thanks for your reviewing.
>>> 
>>> Nope, by N I mean the number of containers. Before your patch, the limit
>>> is global, after your patch it is per container.
>>> 
>> 
>> Yeah, for example, if there is N containers, before the patch, I mean the limit is:
>> 
>> 	N * net->ipv4.sysctl_tcp_max_orphans
>> 
>> After the patch, the limit is:
>> 
>> 	ns1. net->ipv4.sysctl_tcp_max_orphans + ns2. net->ipv4.sysctl_tcp_max_orphans + …
> 
> Not true.
> 
> Please remove "N" from your equation of the current situation.
> 
> "sysctl_tcp_max_orphans" applies to entire system, it is a global limit,
> comparing one limit against all orphans in the system, there is no N.

Yes, it’s right. I browse the source code and found that it’s a global limit, 
sorry for my mistake.

Thanks David and Cong.

Thread

[PATCH] ipv4: Namespaceify tcp_max_orphans knob Haishuang Yan <yanhaishuang@cmss.chinamobile.com> - 2017-09-07 05:20 +0200
  Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob Cong Wang <xiyou.wangcong@gmail.com> - 2017-09-09 00:20 +0200
    Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob 严海双 <yanhaishuang@cmss.chinamobile.com> - 2017-09-09 03:30 +0200
      Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob Cong Wang <xiyou.wangcong@gmail.com> - 2017-09-09 06:40 +0200
        Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob David Miller <davem@davemloft.net> - 2017-09-09 07:20 +0200
          Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob 严海双 <yanhaishuang@cmss.chinamobile.com> - 2017-09-09 12:30 +0200
        Re: [PATCH] ipv4: Namespaceify tcp_max_orphans knob 严海双 <yanhaishuang@cmss.chinamobile.com> - 2017-09-09 07:20 +0200

csiph-web