<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">On 2012-11-29 03:37, support support
wrote:<br>
</div>
<blockquote
cite="mid:CA+gmPydxSxBQsSiX+0XH=qV+bovy7ssRGBaV_yxoNk=2jp2y4g@mail.gmail.com"
type="cite">Well...
<div>We have updated bootrom to 2.0.2.66, after that tried to boot
to 11.1.2.3</div>
<div>Things got worse - we still had the 1Mpps problem, got 2 new
problems - OSPF couldn't load routing table, BGP failed on
updates with Aggregate AS 0.</div>
<div>We had no time to fall back to 6.2 release, so we asked not
to send us Fullview, configured static routing, left it untill
this morning.</div>
<div><br>
</div>
<div>This morning we falled back to 6.2 release and tried to move
the most loaded link to an 10ge-oc192-1-port card(we have
installed it yesturday.</div>
<div>When i tried to configure port - i got message "
<span style="color:rgb(0,0,0);font-family:Arial;font-size:13px">cannot
bind this circuit to the interface. It has an incompatible NAT
policy.</span> "</div>
<div>So we decided to upgrade to 11.1.2.6p1. Now we have this
release working. OSPF works as if it were 6.2 release, BGP must
work, we shall give it a try on one of our uplinks.</div>
<div><br>
</div>
<div>Now I have some questions:</div>
<div>1. what is the oversubscription of 10ge-4-port card? Does it
limit the sum of traffic on each pair of ports to 10G
full-duplex? If so, then we reached this limit.</div>
</blockquote>
<br>
All third generation cards (based on PPA3) have 20G FullDuplex
throughput.<br>
So, if you have 4 x 10G, you can have max 20G in + 20G out. The
oversubsription is 2:1.<br>
If you use just 2 of this ports you will never overload it. However,
additional ports are quite usefull in many configurations and... you
can't have flat 10G load on a port<br>
and keep proper quality so...<br>
<br>
<blockquote
cite="mid:CA+gmPydxSxBQsSiX+0XH=qV+bovy7ssRGBaV_yxoNk=2jp2y4g@mail.gmail.com"
type="cite">
<div>2. what should we do with our chassis, if it fails to
determine hardware type on warm start? As we do "release
upgrade" or "upgrade boot", it reloads and fails to find out
chassis type(sometimes it says 0x9 - unknown, sometimes 0x0) and
sets chassis type to default SE800. After boot finishes - we
unplug power cords, plug them in, watch boot up process -
everything is ok, chassis type is correct.</div>
</blockquote>
<br>
on ours:<br>
Slot : N/A Type :
backplane <br>
Serial No : G61BQ3711Q05NE Hardware Rev :
01B <br>
EEPROM id/ver : 0x5a/4 Mfg Date :
06-SEP-2011 <br>
MAC Address : 00:30:88:04:96:DC POD Status :
Success <br>
ODD Status : Not Available <br>
Chassis Type : SE600 <br>
<br>
Slot : N/A Type :
backplane <br>
Serial No : G61BQ4310Q01E1 Hardware Rev :
01B <br>
EEPROM id/ver : 0x5a/4 Mfg Date :
29-OCT-2010 <br>
MAC Address : 00:30:88:04:91:FE POD Status :
Success <br>
ODD Status : Not Available <br>
Chassis Type : SE600 <br>
<br>
Lot's of reloads since powered up, never had any problems with
chassis type.<br>
Maybe you should replace it.<br>
<br>
Regards,<br>
Marcin<br>
<br>
<br>
<br>
<blockquote
cite="mid:CA+gmPydxSxBQsSiX+0XH=qV+bovy7ssRGBaV_yxoNk=2jp2y4g@mail.gmail.com"
type="cite">
<div><br>
</div>
<div>
<div><br>
</div>
-- <br>
Regards,<br>
Michael<br>
<br>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
redback-nsp mailing list
<a class="moz-txt-link-abbreviated" href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a>
<a class="moz-txt-link-freetext" href="https://puck.nether.net/mailman/listinfo/redback-nsp">https://puck.nether.net/mailman/listinfo/redback-nsp</a>
</pre>
</blockquote>
<br>
</body>
</html>