Re: [f-nsp] XMR Assistance

--===============7896999066101628014==
Content-Type: multipart/alternative; boundary="00000000000071d8da058c44146c"

--00000000000071d8da058c44146c
Content-Type: text/plain; charset="UTF-8"

Are you using the default CAM partition profile?  It's probably not
suitable for a full table.

Do all of your cards have the same FIB size, and is this large enough for
the number of installed routes?

Any IPv6 routes?  IIRC, the advertised max FIB sizes are with all CAM
allocated to IPv4.

-- 
Eldon

On Wed, Jun 26, 2019, 13:13 Derek Maxwell <[email protected]>
wrote:

> Hello Foundry List,
>
> We're currently launching a new site and using a Brocade/Foundry XMR.
> While I've built a number of simple to moderate configs in RX over the last
> few years, the XMR is new territory. When I tried to build the same config
> that I would use in a RX, I encountered odd, intermittent, and unexplained
> behavior. I didn't realize that IronWare would be significantly different
> between the two, as the syntax looks the same, so I was caught a bit off
> guard.
>
> I'm looking for a consultant to do a one-off project to build a solid
> config, likely addressing things that I would miss even if I didn't have
> the odd issues listed above. I'd prefer to pay to do things properly the
> first time.
>
> This unit is for a colocation site with 2 full BGP tables coming in, a
> Layer 2 circuit which provides IP to our office in the same metro, a
> handful of routed virtual interfaces, and a small number of VLANs. The
> single XMR comprises the entire colo network at this point, as it is just a
> single cabinet deployment. We may add Arista 25G equipment for
> hyperconverged storage purposes in the future, but that would likely not
> extend to the Brocade side.
>
> Does anyone have interest in building this out? We are currently running
> the Layer 2 and a single BGP (default only) table on a FCX and it's stable,
> so there isn't incredible urgency until I return from vacation at the end
> of July.
>
> Thanks,
>
> Derek Maxwell
> _______________________________________________
> foundry-nsp mailing list
> [email protected]
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>

--00000000000071d8da058c44146c
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div>Are you using the default CAM partition profile?=C2=
=A0 It&#39;s probably not suitable for a full table.<div dir=3D"auto"><br><=
/div><div dir=3D"auto">Do all of your cards have the same FIB size, and is =
this large enough for the number of installed routes?</div><div dir=3D"auto=
"><br></div><div dir=3D"auto">Any IPv6 routes?=C2=A0 IIRC, the advertised m=
ax FIB sizes are with all CAM allocated to IPv4.</div><div dir=3D"auto"><br=
></div><div dir=3D"auto">--=C2=A0</div><div dir=3D"auto">Eldon</div><br><di=
v class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Jun 2=
6, 2019, 13:13 Derek Maxwell &lt;<a href=3D"mailto:derek.maxwell@trueinnova=
tion.net">[email protected]</a>&gt; wrote:<br></div><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc s=
olid;padding-left:1ex"><div dir=3D"ltr">Hello Foundry List,<div><br></div><=
div>We&#39;re currently launching a new site and using a Brocade/Foundry XM=
R. While I&#39;ve built a number of simple to moderate configs in RX over t=
he last few years, the XMR is new territory. When I tried to build the same=
 config that I would use in a RX, I encountered odd, intermittent, and unex=
plained behavior. I didn&#39;t realize that IronWare would be significantly=
 different between the two, as the syntax looks the same, so I was caught a=
 bit off guard.=C2=A0</div><div><br></div><div>I&#39;m looking for a consul=
tant to do a one-off project to build a solid config, likely addressing thi=
ngs that I would miss even if I didn&#39;t have the odd issues listed above=
. I&#39;d prefer to pay to do things properly the first time.</div><div><br=
></div><div>This unit is for a colocation site with 2 full BGP tables comin=
g in, a Layer 2 circuit which provides IP to our office in the same metro, =
a handful of routed virtual interfaces, and a small number of VLANs. The si=
ngle XMR comprises the entire colo network at this point, as it is just a s=
ingle cabinet deployment. We may add Arista 25G equipment for hyperconverge=
d storage purposes in the future, but that would likely not extend to the B=
rocade side.</div><div><br></div><div>Does anyone have interest in building=
 this out? We are currently running the Layer 2 and a single BGP (default o=
nly) table on a FCX and it&#39;s stable, so there isn&#39;t incredible urge=
ncy until I return from vacation at the end of July.</div><div><br></div><d=
iv>Thanks,</div><div><br></div><div><div><div dir=3D"ltr" class=3D"m_-31956=
69392968968294gmail_signature" data-smartmail=3D"gmail_signature"><div dir=
=3D"ltr"><div>Derek Maxwell</div></div></div></div></div></div>
_______________________________________________<br>
foundry-nsp mailing list<br>
<a href=3D"mailto:[email protected]" target=3D"_blank" rel=3D"nor=
eferrer">[email protected]</a><br>
<a href=3D"http://puck.nether.net/mailman/listinfo/foundry-nsp" rel=3D"nore=
ferrer noreferrer" target=3D"_blank">http://puck.nether.net/mailman/listinf=
o/foundry-nsp</a><br>
</blockquote></div></div></div>

--00000000000071d8da058c44146c--

--===============7896999066101628014==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
foundry-nsp mailing list
[email protected]
http://puck.nether.net/mailman/listinfo/foundry-nsp

--===============7896999066101628014==--