[AusNOG] Speed issues on ASR1k

Sat Mar 18 20:20:43 EST 2017

For example:

lns01.syd02.nsw#ping 203.219.113.35 source lo100 size 1540 df

Type escape sequence to abort.
Sending 5, 1540-byte ICMP Echos to 203.219.113.35, timeout is 2 seconds:
Packet sent with a source address of X.X.X.acket sent with the DF bit set
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 8/9/12 ms
lns01.syd02.nsw#

Kindest Regards,

Nathan Brookfield (VK2NAB)

Chief Executive Officer

Simtronic Technologies Pty Ltd

Local: (02) 4749 4949 | Fax: (02) 4749 4950 | Direct: (02) 4749 4951

Web: http://www.simtronic.com.au<http://www.simtronic.com.au/> | E-mail: nathan.brookfield at simtronic.com.au<mailto:nathan.brookfield at simtronic.com.au>

________________________________
From: AusNOG <ausnog-bounces at lists.ausnog.net> on behalf of Nathan Brookfield <Nathan.Brookfield at simtronic.com.au>
Sent: Saturday, 18 March 2017 8:14 PM
To: dynamic.mineral at gmail.com
Cc: ausnog at lists.ausnog.net
Subject: Re: [AusNOG] Speed issues on ASR1k

Do you have the physical interfaces on the ASR and your switch set to 9000 or 9216 for Jumbo's?

L2TP is going to take up 40 bytes of your packet so you should be able to ping the LAC at 1540 bytes without fragmentation for the best performance.

In saying that, the issue may also be virtual-reassembly on your LNS interfaces as well.

Nathan Brookfield
Chief Executive Officer

Simtronic Technologies Pty Ltd
http://www.simtronic.com.au

On 18 Mar 2017, at 20:02, Pouya Madani <dynamic.mineral at gmail.com<mailto:dynamic.mineral at gmail.com>> wrote:

ping to both sample LAC and 4.2.2.4 is at max MTU 1500 with DF bit set

#ping vrf INTERNET
Protocol [ip]:
Target IP address: 203.219.113.35
Repeat count [5]: 1
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]: y
Ingress ping [n]:
Source address or interface:
Type of service [0]:
Set DF bit in IP header? [no]: yes
Validate reply data? [no]:
Data pattern [0x0000ABCD]:
Loose, Strict, Record, Timestamp, Verbose[none]: V
Loose, Strict, Record, Timestamp, Verbose[V]:
Sweep range of sizes [n]: y
Sweep min size [36]: 1450
Sweep max size [18024]: 1550
Sweep interval [1]:
Type escape sequence to abort.
Sending 101, [1450..1550]-byte ICMP Echos to 203.219.113.35, timeout is 2 seconds:
Packet sent with the DF bit set
Reply to request 0 (2 ms) (size 1450)
Reply to request 1 (1 ms) (size 1451)
Reply to request 2 (2 ms) (size 1452)
Reply to request 3 (2 ms) (size 1453)
Reply to request 4 (1 ms) (size 1454)
Reply to request 5 (1 ms) (size 1455)
Reply to request 6 (2 ms) (size 1456)
Reply to request 7 (1 ms) (size 1457)
Reply to request 8 (1 ms) (size 1458)
Reply to request 9 (2 ms) (size 1459)
Reply to request 10 (1 ms) (size 1460)
Reply to request 11 (1 ms) (size 1461)
Reply to request 12 (2 ms) (size 1462)
Reply to request 13 (1 ms) (size 1463)
Reply to request 14 (1 ms) (size 1464)
Reply to request 15 (2 ms) (size 1465)
Reply to request 16 (1 ms) (size 1466)
Reply to request 17 (1 ms) (size 1467)
Reply to request 18 (2 ms) (size 1468)
Reply to request 19 (2 ms) (size 1469)
Reply to request 20 (1 ms) (size 1470)
Reply to request 21 (1 ms) (size 1471)
Reply to request 22 (1 ms) (size 1472)
Reply to request 23 (1 ms) (size 1473)
Reply to request 24 (2 ms) (size 1474)
Reply to request 25 (1 ms) (size 1475)
Reply to request 26 (1 ms) (size 1476)
Reply to request 27 (2 ms) (size 1477)
Reply to request 28 (1 ms) (size 1478)
Reply to request 29 (1 ms) (size 1479)
Reply to request 30 (2 ms) (size 1480)
Reply to request 31 (1 ms) (size 1481)
Reply to request 32 (1 ms) (size 1482)
Reply to request 33 (2 ms) (size 1483)
Reply to request 34 (1 ms) (size 1484)
Reply to request 35 (1 ms) (size 1485)
Reply to request 36 (1 ms) (size 1486)
Reply to request 37 (2 ms) (size 1487)
Reply to request 38 (2 ms) (size 1488)
Reply to request 39 (1 ms) (size 1489)
Reply to request 40 (1 ms) (size 1490)
Reply to request 41 (2 ms) (size 1491)
Reply to request 42 (1 ms) (size 1492)
Reply to request 43 (1 ms) (size 1493)
Reply to request 44 (2 ms) (size 1494)
Reply to request 45 (2 ms) (size 1495)
Reply to request 46 (1 ms) (size 1496)
Reply to request 47 (2 ms) (size 1497)
Reply to request 48 (1 ms) (size 1498)
Reply to request 49 (2 ms) (size 1499)
Reply to request 50 (1 ms) (size 1500)
Request 51 timed out (size 1501)
Request 52 timed out (size 1502)
Request 53 timed out (size 1503)
Request 54 timed out (size 1504)
Request 55 timed out (size 1505)
^C
Success rate is 91 percent (51/56), round-trip min/avg/max = 1/1/2 ms

ping vrf INTERNET
Protocol [ip]:
Target IP address: 4.2.2.4
Repeat count [5]: 1
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]: y
Ingress ping [n]:
Source address or interface:
Type of service [0]:
Set DF bit in IP header? [no]: yes
Validate reply data? [no]:
Data pattern [0x0000ABCD]:
Loose, Strict, Record, Timestamp, Verbose[none]: V
Loose, Strict, Record, Timestamp, Verbose[V]:
Sweep range of sizes [n]: y
Sweep min size [36]: 1450
Sweep max size [18024]: 1550
Sweep interval [1]:
Type escape sequence to abort.
Sending 101, [1450..1550]-byte ICMP Echos to 4.2.2.4, timeout is 2 seconds:
Packet sent with the DF bit set
Reply to request 0 (151 ms) (size 1450)
Reply to request 1 (151 ms) (size 1451)
Reply to request 2 (151 ms) (size 1452)
Reply to request 3 (152 ms) (size 1453)
Reply to request 4 (152 ms) (size 1454)
Reply to request 5 (151 ms) (size 1455)
Reply to request 6 (151 ms) (size 1456)
Reply to request 7 (152 ms) (size 1457)
Reply to request 8 (151 ms) (size 1458)
Reply to request 9 (151 ms) (size 1459)
Reply to request 10 (152 ms) (size 1460)
Reply to request 11 (152 ms) (size 1461)
Reply to request 12 (152 ms) (size 1462)
Reply to request 13 (151 ms) (size 1463)
Reply to request 14 (151 ms) (size 1464)
Reply to request 15 (152 ms) (size 1465)
Reply to request 16 (151 ms) (size 1466)
Reply to request 17 (151 ms) (size 1467)
Reply to request 18 (151 ms) (size 1468)
Reply to request 19 (151 ms) (size 1469)
Reply to request 20 (152 ms) (size 1470)
Reply to request 21 (152 ms) (size 1471)
Reply to request 22 (151 ms) (size 1472)
Reply to request 23 (151 ms) (size 1473)
Reply to request 24 (152 ms) (size 1474)
Reply to request 25 (151 ms) (size 1475)
Reply to request 26 (151 ms) (size 1476)
Reply to request 27 (152 ms) (size 1477)
Reply to request 28 (151 ms) (size 1478)
Reply to request 29 (151 ms) (size 1479)
Reply to request 30 (151 ms) (size 1480)
Reply to request 31 (151 ms) (size 1481)
Reply to request 32 (151 ms) (size 1482)
Reply to request 33 (152 ms) (size 1483)
Reply to request 34 (151 ms) (size 1484)
Reply to request 35 (152 ms) (size 1485)
Reply to request 36 (151 ms) (size 1486)
Reply to request 37 (151 ms) (size 1487)
Reply to request 38 (152 ms) (size 1488)
Reply to request 39 (151 ms) (size 1489)
Reply to request 40 (151 ms) (size 1490)
Reply to request 41 (152 ms) (size 1491)
Reply to request 42 (151 ms) (size 1492)
Reply to request 43 (151 ms) (size 1493)
Reply to request 44 (152 ms) (size 1494)
Reply to request 45 (151 ms) (size 1495)
Reply to request 46 (151 ms) (size 1496)
Reply to request 47 (155 ms) (size 1497)
Reply to request 48 (151 ms) (size 1498)
Reply to request 49 (151 ms) (size 1499)
Reply to request 50 (151 ms) (size 1500)
Request 51 timed out (size 1501)
Request 52 timed out (size 1502)
Request 53 timed out (size 1503)
Request 54 timed out (size 1504)
^C
Success rate is 92 percent (51/55), round-trip min/avg/max = 151/151/155 ms

On Sat, Mar 18, 2017 at 7:31 PM, Nathan Brookfield <Nathan.Brookfield at simtronic.com.au<mailto:Nathan.Brookfield at simtronic.com.au>> wrote:
Towards you upstream providers that are switching the L2TP sessions what is the maximum MTU you can ping with the DF bit sets

Nathan Brookfield
Chief Executive Officer

Simtronic Technologies Pty Ltd
http://www.simtronic.com.au

On 18 Mar 2017, at 19:16, Pouya Madani <dynamic.mineral at gmail.com<mailto:dynamic.mineral at gmail.com>> wrote:

Hi everyone,

Need thoughts on a speed issue and would appreciate if anyone might be able to help as I am running out of ideas.

I have a new ASR1002 in production to operate as LNS but services authenticated are experiencing speed issues and web pages not loading properly. My initial thought was the MTU size but this doesnt look to be the root cause as authenticated services can ping (bbc.com<http://bbc.com> for instance) by MTU size 1464 (28bit ICMP + 8bit PPP) ok, but (bbc again for example) web pages fail to load completely when browsed to. The sample service I am testing with is a 20MB EFM and speed test sits around 16MB up/down. No input/output errors on any of the interfaces and speed is 1000 duplex full on all.

I would appreciate if any one can suggest anything else I might be missing out.

Thanks in advance.
_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net<mailto:AusNOG at lists.ausnog.net>
http://lists.ausnog.net/mailman/listinfo/ausnog

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20170318/b88519e2/attachment.html>