All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dilip Daya <dilip.daya@hp.com>
To: linux-sctp@vger.kernel.org
Subject: Re: SCTP sysctl /proc/sys/net/sctp/rto_min/initial/max variables
Date: Thu, 29 Nov 2012 22:53:55 +0000	[thread overview]
Message-ID: <1354229635.6915.146.camel@pro6455b.example.com> (raw)
In-Reply-To: <1354221337.6915.107.camel@pro6455b.example.com>

Hi Vlad / Niel,

Please disregard this issue. I extracted sctp from kernel linux-3.2.34
and it does not seem to be a problem. Looks like the sctp I have in
3.2.32 is the problem.

Once again, I very sorry to engage you in this.

Thanking you in advance for all your time and effort.
-DilipD.


On Thu, 2012-11-29 at 16:33 -0500, Dilip Daya wrote:
> Hi Vlad,
> 
> On Thu, 2012-11-29 at 15:49 -0500, Vlad Yasevich wrote:
> > On 11/29/2012 03:35 PM, Dilip Daya wrote:
> > > Hi,
> > >
> > > Testing kernel 3.2.32 x86_64 environment show the sctp sysctl variables:
> > >
> > >   - net.sctp.rto_initial,
> > >   - net.sctp.rto_min,
> > >   - net.sctp.rto_max
> > >
> > > are not overriding the hard-coded recommended values of:
> > > - http://lxr.linux.no/linux+v3.2.34/include/net/sctp/constants.h#L276
> > >
> > > I've confirmed the following values are correct to what I need modified:
> > >   - /proc/sys/net/sctp/rto_initial
> > >   - /proc/sys/net/sctp/rto_min
> > >   - /proc/sys/net/sctp/rto_max
> > >
> > > If I change the hard-coded values and rebuild sctp they take effect. Is
> > > this an issue on my end or is this a sctp sysctl issue?
> > >
> > > Thanks,
> > > -DilipD.
> > >
> > 
> > Dilip
> > 
> > I am not sure what you mean...   Are you saying that issuing
> > 	sysctl -w net.sctp.rto_initial\x100
> > doesn't change the value of rto_initial shown in /proc/sys/net/sctp?
> 
> No, it does change the value in /proc/sys/net/sctp but not picked up by
> sctp associations. Only default recommended values are picked up, i.e.
> 
> # sysctl -a 2>/dev/null | grep ".rto_"
> net.ipv4.tcp_frto_response = 0
> net.sctp.rto_initial = 25
> net.sctp.rto_min = 20
> net.sctp.rto_max = 30
> net.sctp.rto_alpha_exp_divisor = 3
> net.sctp.rto_beta_exp_divisor = 2
> 
> # cat /proc/sys/net/sctp/rto_max
> 30
> 
> # cat /proc/sys/net/sctp/rto_min 
> 20
> 
> # cat /proc/sys/net/sctp/rto_initial 
> 25
> 
> Selective sctp debug:
> [  206.783286] sctp_init_sock: net_params->rto_min: 1000,
> net_params->rto_initial: 3000, net_params->rto_max: 60000
> [  206.783291] sctp_init_sock: sp->rtoinfo.srto_min: 1000,
> sp->rtoinfo.srto_initial: 3000, sp->rtoinfo.srto_max: 60000
> [  206.783403] sctp_association_init: sp->rtoinfo.srto_min: 1000,
> sp->rtoinfo.srto_initial: 3000, sp->rtoinfo.srto_max: 60000
> [  206.783407] sctp_association_init: asoc->rto_min: 250,
> asoc->rto_initial: 750, asoc->rto_max: 15000
> [  206.783426] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.783993] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784008] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784019] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784028] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784038] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784047] sctp_transport_init: net_params->rto_initial: 3000,
> peer->rto: 750
> [  206.784731] sctp_transport_update_rto: transport: ffff88011e1b0400,
> rtt: 0, srtt: 0 rttvar: 1, rto: 250, rto-msecs: 1000, rto_min: 250,
> rto_max: 15000
> ...
> 
> # cat /proc/net/sctp/remaddr 
> ADDR ASSOC_ID HB_ACT RTO MAX_PATH_RTX REM_ADDR_RTX  START
> 192.168.2.62  1 1 250 5 0 0
> 
> # cat /proc/net/sctp/assocs_info | grep "R:"
> R:192.168.2.62  1 1500 0 5 11832 0 21040 250 0 1 0
> 
> i.e.
> 
> Default hard-coded recommended values are:
> rto_min     = 1000  msec = 1000/4  =   250 jiffies
> rto_initial = 3000  msec = 3000/4  =   750 jiffies
> rto_max     = 60000 msec = 60000/4 = 15000 jiffies
> 
> 
> => Seems sctp sysctl rto_ variables do not effect.
> 
> -DilipD.
> 
> > 
> > -vlad
> > 
> > > --
> > > To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> > > the body of a message to majordomo@vger.kernel.org
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > >
> > 


      parent reply	other threads:[~2012-11-29 22:53 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-29 20:35 SCTP sysctl /proc/sys/net/sctp/rto_min/initial/max variables Dilip Daya
2012-11-29 20:49 ` Vlad Yasevich
2012-11-29 21:33 ` Dilip Daya
2012-11-29 22:53 ` Dilip Daya [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1354229635.6915.146.camel@pro6455b.example.com \
    --to=dilip.daya@hp.com \
    --cc=linux-sctp@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.