Re: Query on querier version in IGMPv3
Indranil Bhattacharya <[email protected]> Mon, 19 Jul 2010 14:51:07 +0530
| Newsgroups | gmane.ietf.magma |
|---|---|
| Message-ID | <[email protected]> |
--===============1422185637==
Content-Type: multipart/alternative; boundary=00c09f88d160b1c0a0048bba16cd
--00c09f88d160b1c0a0048bba16cd
Content-Type: text/plain; charset=ISO-8859-1
Hi Bharat,
The first part is applicable to a router which has v3 but has
been configured as v2.
The 'reselection' part was an answer for when the router
should revert back to the configured higher query version. This will
be done when transmitting general query. After that if it receives v2/v1
query then it will lower the version again. Even if you take the new timer
approach, problem is that how do you control sequence of queries from
routers? Say, after timer expiry version is changed and v3 GQ is sent. V2
non-querier does not understand v3 query so it sends v2 GQ.
Section 7.3.1 says that it is an administrative
responsibility. Excerpt is given below:
"If any older versions of IGMP are present on routers, the querier
MUST use the lowest version of IGMP present on the network. This
must be administratively assured; routers that desire to be
compatible with IGMPv1 and IGMPv2 MUST have a configuration option
to act in IGMPv1 or IGMPv2 compatibility modes."
Thanks,
Indranil
On Mon, Jul 19, 2010 at 1:45 PM, Bharat Joshi <[email protected]>wrote:
> Indranil,
>
> Ok. But RFC 3376 clearly says that a router should lower its version to
> the lowest version of query heard on a network.
>
> One thing I did not get in your reply is 'Let the v2 updates's it
> querier timer from v3 query...'. While second part of this statement is
> clear, I am not sure about the first part. A version 2 querier may/may not
> support V3 so things won't work any way and there will be multiple number of
> queries on network which may not be desirable.
>
> Another thing I did not understand is the last sentence. What do you
> mean by 'reselection can happen after every general query'? I think RFC 3376
> suggested this to avoid multiple queries on a network. If a router do what
> is suggested in RFC 3376, at least there will be only one query in the
> network.
>
> I checked couple of open-source implementation and it seems none of
> them store this information and always use the highest configured version to
> send out an IGMP query. Not sure if there was any specific reason for not
> doing this.
>
> Regards,
> Bharat
> ________________________________________
> From: Indranil Bhattacharya [[email protected]]
> Sent: Monday, July 19, 2010 1:43 PM
> To: Bharat Joshi
> Cc: [email protected]
> Subject: Re: [magma] Query on querier version in IGMPv3
>
> Hi Bharat,
>
> Do not lower the querier version. Let the v2 updates's it
> querier timer from v3 query or let it use it's configured query interval.
> Otherwise, the reselection can happen after every General query.
>
> Thanks,
> Indranil
>
> On Mon, Jul 19, 2010 at 10:39 AM, Bharat Joshi <[email protected]
> <mailto:[email protected]>> wrote:
> Hi All,
>
> I was re-reading RFC 3376 for handling compatibility for lower version
> of querier.
>
> In section 7.3.1, it is mentioned that a querier configured to run in
> V3 mode
> should lower its version when it sees a lower version query on the same
> network. From that time onwards, a router should send the lower version
> queries.
>
> Now the question is, when this router should revert back to its
> configured higher
> version? The text in section 7.3.1 or elsewhere does not talk about
> this.
>
> Should a timer 'lower_version_querier' be started and updated as and
> when lower
> version queries are removed? Please note this is not same as
> other-querier-present
> timer.
>
> Can someone, who has implemented IGMPv3, let me know how this
> particular case
> has been handled in their implementation?
>
> Regards,
> Bharat
> **************** CAUTION - Disclaimer *****************
> This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended
> solely
> for the use of the addressee(s). If you are not the intended recipient,
> please
> notify the sender by e-mail and delete the original message. Further, you
> are not
> to copy, disclose, or distribute this e-mail or its contents to any other
> person and
> any such actions are unlawful. This e-mail may contain viruses. Infosys has
> taken
> every reasonable precaution to minimize this risk, but is not liable for
> any damage
> you may sustain as a result of any virus in this e-mail. You should carry
> out your
> own virus checks before opening the e-mail or attachment. Infosys reserves
> the
> right to monitor and review the content of all messages sent to or from
> this e-mail
> address. Messages sent to or from this e-mail address may be stored on the
> Infosys e-mail system.
> ***INFOSYS******** End of Disclaimer ********INFOSYS***
> _______________________________________________
> magma mailing list
> [email protected]<mailto:[email protected]>
> https://www.ietf.org/mailman/listinfo/magma
>
>
--00c09f88d160b1c0a0048bba16cd
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<div>Hi Bharat,</div>
<div>=A0</div>
<div>=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 The first part is applicable t=
o a router which has v3 but has been configured as v2. </div>
<div>=A0</div>
<div>=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0The 'reselection'=
part was an answer for=A0when the router should revert back to the=A0confi=
gured higher query version. This will be=A0done=A0when=A0transmitting=A0gen=
eral query. After that if it receives v2/v1 query then it will lower the ve=
rsion again. Even if you take the new timer approach, problem is that how d=
o you control sequence of queries from routers? Say, after timer expiry ver=
sion is changed and v3=A0GQ is sent. V2 non-querier does not understand v3 =
query so it sends v2 GQ. </div>
<div>=A0</div>
<div>=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 Section 7.3.1 says that =
it is an administrative responsibility. Excerpt is given below:</div>
<div>"If any older versions of IGMP are present on routers, the querie=
r<br>=A0=A0=A0=A0 MUST use the lowest version of IGMP present on the networ=
k.=A0 This<br>=A0=A0=A0=A0 must be administratively assured; routers that d=
esire to be<br>
=A0=A0=A0=A0 compatible with IGMPv1 and IGMPv2 MUST have a configuration op=
tion<br>=A0=A0=A0=A0 to act in IGMPv1 or IGMPv2 compatibility modes."<=
/div>
<div>=A0</div>
<div>Thanks,</div>
<div>Indranil<br><br></div>
<div class=3D"gmail_quote">On Mon, Jul 19, 2010 at 1:45 PM, Bharat Joshi <s=
pan dir=3D"ltr"><<a href=3D"mailto:[email protected]">bharat_josh=
[email protected]</a>></span> wrote:<br>
<blockquote style=3D"BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex=
; PADDING-LEFT: 1ex" class=3D"gmail_quote">Indranil,<br><br>=A0 =A0 Ok. But=
RFC 3376 clearly says that a router should lower its version to the lowest=
version of query heard on a network.<br>
<br>=A0 =A0 One thing I did not get in your reply is 'Let the v2 update=
s's it querier timer from v3 query...'. While second part of this s=
tatement is clear, I am not sure about the first part. A version 2 querier =
may/may not support V3 so things won't work any way and there will be m=
ultiple number of queries on network which may not be desirable.<br>
<br>=A0 =A0 Another thing I did not understand is the last sentence. What d=
o you mean by 'reselection can happen after every general query'? I=
think RFC 3376 suggested this to avoid multiple queries on a network. If a=
router do what is suggested in RFC 3376, at least there will be only one q=
uery in the network.<br>
<br>=A0 =A0 I checked couple of open-source implementation and it seems non=
e of them store this information and always use the highest configured vers=
ion to send out an IGMP query. Not sure if there was any specific reason fo=
r not doing this.<br>
<br>Regards,<br>Bharat<br>________________________________________<br>From:=
Indranil Bhattacharya [<a href=3D"mailto:[email protected]">myselfi=
[email protected]</a>]<br>Sent: Monday, July 19, 2010 1:43 PM<br>To: Bharat=
Joshi<br>
Cc: <a href=3D"mailto:[email protected]">[email protected]</a><br>Subject: Re: [m=
agma] Query on querier version in IGMPv3<br>
<div class=3D"im"><br>Hi Bharat,<br><br>=A0 =A0 =A0 =A0 =A0 =A0 =A0Do not l=
ower the querier version. Let the v2 updates's it querier timer from v3=
query or let it use it's configured query interval. Otherwise, the res=
election can happen after every General query.<br>
<br>Thanks,<br>Indranil<br><br></div>
<div>
<div></div>
<div class=3D"h5">On Mon, Jul 19, 2010 at 10:39 AM, Bharat Joshi <<a hre=
f=3D"mailto:[email protected]">[email protected]</a><mailt=
o:<a href=3D"mailto:[email protected]">[email protected]</a>&=
gt;> wrote:<br>
Hi All,<br><br>=A0 =A0 I was re-reading RFC 3376 for handling compatibility=
for lower version of querier.<br><br>=A0 =A0 In section 7.3.1, it is menti=
oned that a querier configured to run in V3 mode<br>=A0 =A0 should lower it=
s version when it sees a lower version query on the same<br>
=A0 =A0 network. From that time onwards, a router should send the lower ver=
sion<br>=A0 =A0 queries.<br><br>=A0 =A0 Now the question is, when this rout=
er should revert back to its configured higher<br>=A0 =A0 version? The text=
in section 7.3.1 or elsewhere does not talk about this.<br>
<br>=A0 =A0 Should a timer 'lower_version_querier' be started and u=
pdated as and when lower<br>=A0 =A0 version queries are removed? Please not=
e this is not same as other-querier-present<br>=A0 =A0 timer.<br><br>=A0 =
=A0 Can someone, who has implemented IGMPv3, let me know how this particula=
r case<br>
=A0 =A0 has been handled in their implementation?<br><br>Regards,<br>Bharat=
<br>**************** CAUTION - Disclaimer *****************<br>This e-mail =
contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended solely<br>for the=
use of the addressee(s). If you are not the intended recipient, please<br>
notify the sender by e-mail and delete the original message. Further, you a=
re not<br>to copy, disclose, or distribute this e-mail or its contents to a=
ny other person and<br>any such actions are unlawful. This e-mail may conta=
in viruses. Infosys has taken<br>
every reasonable precaution to minimize this risk, but is not liable for an=
y damage<br>you may sustain as a result of any virus in this e-mail. You sh=
ould carry out your<br>own virus checks before opening the e-mail or attach=
ment. Infosys reserves the<br>
right to monitor and review the content of all messages sent to or from thi=
s e-mail<br>address. Messages sent to or from this e-mail address may be st=
ored on the<br>Infosys e-mail system.<br>***INFOSYS******** End of Disclaim=
er ********INFOSYS***<br>
_______________________________________________<br>magma mailing list<br></=
div></div><a href=3D"mailto:[email protected]">[email protected]</a><mailto:<a=
href=3D"mailto:[email protected]">[email protected]</a>><br>
<div>
<div></div>
<div class=3D"h5"><a href=3D"https://www.ietf.org/mailman/listinfo/magma" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/magma</a><br><br></d=
iv></div></blockquote></div><br>
--00c09f88d160b1c0a0048bba16cd--
--===============1422185637==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
magma mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/magma
--===============1422185637==--