cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Announcements
Join Customer Connection to register!
2456
Views
0
Helpful
10
Replies
lcaruso
Contributor

1600 byte mtu for DSL?

This is my first experience with an 887VA router. I was surprised to see the ATM mtu is 1600.

Is there anything I can do to speed this router's DSL performance? It is slower than an off the shelf ActionTec costing less than $100.

I've already updated the firmware to the latest release. Same with IOS.

dslrouter#sh controller vdsl 0

Controller VDSL 0 is UP

Daemon Status:           Up

                        XTU-R (DS)              XTU-C (US)

Chip Vendor ID:         'BDCM'                   'GSPN'

Chip Vendor Specific:   0x0000                   0x000D

Chip Vendor Country:    0xB500                   0xFFB5

Modem Vendor ID:        'CSCO'                   '    '

Modem Vendor Specific:  0x4602                   0x0000

Modem Vendor Country:   0xB500                   0x0000

Serial Number Near:    FTX151902CJ 887VA-SE 15.1(4)M 

Serial Number Far:    

Modem Version Near:    15.1(4)M

Modem Version Far:     0x000d

Modem Status:            TC Sync (Showtime!)

DSL Config Mode:         AUTO

Trained Mode:            G.992.5 (ADSL2+) Annex A

TC Mode:                 ATM

Selftest Result:         0x00

DELT configuration:      disabled

DELT state:              not running

Trellis:                 ON                       ON

Line Attenuation:         9.5 dB                  2.3 dB

Signal Attenuation:       9.5 dB                 43.8 dB

Noise Margin:            11.1 dB                 14.0 dB

Attainable Rate:        20728 kbits/s            28740 kbits/s

Actual Power:             dBm             7.6 dBm

Total FECS:             0                        0

Total ES:               0                        0

Total SES:              0                        0

Total LOSS:             0                        0

Total UAS:              0                        0

Total LPRS:             0                        0

Total LOFS:             0                        0

Total LOLS:             0                        0

Bit swap:               0                        18

Full inits:             1

Failed full inits:      0

Short inits:            0

Failed short inits:     0

Firmware        Source          File Name (version)

--------        ------          -------------------

VDSL            default         flash:vdsl.bin (10)

Modem FW  Version:      110201_1505-4.02L.03.A2pv6C033c.d23e

Modem PHY Version:      A2pv6C033c.d23e

                  DS Channel1     DS Channel0   US Channel1       US Channel0

Speed (kbps):             0            14975             0               766

Previous Speed:           0                0             0                 0

Total Cells:              0         37649525             0           1886643

User Cells:               0           510515             0             85265

Reed-Solomon EC:          0                0             0                 0

CRC Errors:               0                0             0                 0

Header Errors:            0                0             0                 0

Interleave (ms):       0.00             3.94          0.00              2.31

Actual INP:            0.00             0.51          0.00              0.57

Training Log :  Stopped

Training Log Filename : flash:vdsllog.bin

dslrouter#sh int

ATM0 is up, line protocol is up

  Hardware is MPC ATMSAR, address is a.b.c (bia 503d.e574.b444)

  MTU 1600 bytes, sub MTU 1600, BW 766 Kbit/sec, DLY 480 usec,

     reliability 255/255, txload 14/255, rxload 101/255

  Encapsulation ATM, loopback not set

  Keepalive not supported

  Encapsulation(s): AAL5

  4 maximum active VCs, 1024 VCs per VP, 1 current VCCs

  VC Auto Creation Disabled.

  VC idle disconnect time: 300 seconds

  Last input 00:12:29, output 00:00:01, output hang never

  Last clearing of "show interface" counters never

  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0

  Queueing strategy: Per VC Queueing

  5 minute input rate 304000 bits/sec, 21 packets/sec

  5 minute output rate 45000 bits/sec, 10 packets/sec

     18773 packets input, 23530615 bytes, 0 no buffer

     Received 0 broadcasts (0 IP multicasts)

     0 runts, 0 giants, 0 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort

     14193 packets output, 3635602 bytes, 0 underruns

     0 output errors, 0 collisions, 0 interface resets

     0 unknown protocol drops

     0 output buffer failures, 0 output buffers swapped out

ATM0.1 is up, line protocol is up

  Hardware is MPC ATMSAR, address is a.b.c (bia a.b.c)

  MTU 1600 bytes, BW 766 Kbit/sec, DLY 480 usec,

     reliability 255/255, txload 14/255, rxload 101/255

  Encapsulation ATM

  Keepalive not supported

     18773 packets input, 23718261 bytes

     14193 packets output, 3635602 bytes

     0 OAM cells input, 0 OAM cells output

  AAL5 CRC errors : 0

  AAL5 SAR Timeouts : 0

  AAL5 Oversized SDUs : 0

  Last clearing of "show interface" counters never

BVI1 is up, line protocol is up

  Hardware is BVI, address is a.b.c (bia 0000.0000.0000)

  Description: TDS

  Internet address is a.b.c.d/30

  MTU 1600 bytes, BW 4608 Kbit/sec, DLY 5000 usec,

     reliability 255/255, txload 2/255, rxload 17/255

  Encapsulation ARPA, loopback not set

  Keepalive set (10 sec)

  ARP type: ARPA, ARP Timeout 04:00:00

  Last input never, output never, output hang never

  Last clearing of "show interface" counters never

  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0

  Queueing strategy: fifo

  Output queue: 0/0 (size/max)

  5 minute input rate 310000 bits/sec, 21 packets/sec

  5 minute output rate 42000 bits/sec, 10 packets/sec

     18766 packets input, 23792529 bytes, 0 no buffer

     Received 0 broadcasts (0 IP multicasts)

     0 runts, 0 giants, 0 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort

     13821 packets output, 3481612 bytes, 0 underruns

     0 output errors, 0 collisions, 0 interface resets

     0 unknown protocol drops

     0 output buffer failures, 0 output buffers swapped out

5 ACCEPTED SOLUTIONS

Accepted Solutions
paolo bevilacqua
Hall of Fame Master

Possibly, you have firewall, inspection, cbac, and other unnecessary configurations.

If so, remove them and test again.

Also, you can try RBE (atm route-bridged ip) instead of BVI/IRB.

View solution in original post

RBE works fine for CPE, it's worth to try it.

Just remove ip virtual-reassembly.

Can you post a numeric comparison of speedtest.net, between Cisco and the other router ?

View solution in original post

There is a difference, albeit not large. (to easily copy speedtest.net results to clipbpard, click "my results" and select the line(s) of interest).

The problem maybe has something to do with interleave or other suble ADSL detail. Usually Cisco can fixes this kind of stuff, but it takes time.  If you have a support contract, you can insist that your problem is given  to engineering, as you can demonstrate how a competitor product performs better.

RBE is very simple, under subif remove bridge, place ip address, and "atm route-bridged ip". Also remove virtual reassembly.

Please remember to rate useful posts clicking on the stars below.

View solution in original post

Unfortunately a clean configuration doesn't always mean better performances, but at least you've tried what is in your power.

It would be interesting to know if the other router uses interleaving. Supposedly it should have to, as interleaving is controlled by the DSLAM. See: http://www.kitz.co.uk/adsl/interleaving.htm

Thanks for the nice rating and good luck!

View solution in original post

No, it will forward forward fragments. It is supposedly done to help consistency of ACL, NAT and so.

But all works better without it.

View solution in original post

10 REPLIES 10
paolo bevilacqua
Hall of Fame Master

Possibly, you have firewall, inspection, cbac, and other unnecessary configurations.

If so, remove them and test again.

Also, you can try RBE (atm route-bridged ip) instead of BVI/IRB.

View solution in original post

Hi,

I was told by TAC that RBE is not for CPE. Are they right?

I don't have much of a configuration that would slow things down, as I do know what you mean.

RBE works fine for CPE, it's worth to try it.

Just remove ip virtual-reassembly.

Can you post a numeric comparison of speedtest.net, between Cisco and the other router ?

View solution in original post

Thanks for the tip on RBE! I'll try it.

Do you have an example of RBE for RFC 1483  bridged?

I have speedtest.net results, but I find them somewhat erratic. My best metric is ping.

When I ping the next hop router across my DSL connection, the TDS-branded ActionTec724R performs a consistent 7ms faster. ActionTec 11ms consistent average. 887VA 18ms consistent average.

TDS uses an Entrisphere DSLAM, don't know which model. I can only assume they tweaked the BIOS on their branded modem to work a little better with their DSLAM.

I wonder if there are any tweaks I could make on the 887VA besides trying RBE?

There is a difference, albeit not large. (to easily copy speedtest.net results to clipbpard, click "my results" and select the line(s) of interest).

The problem maybe has something to do with interleave or other suble ADSL detail. Usually Cisco can fixes this kind of stuff, but it takes time.  If you have a support contract, you can insist that your problem is given  to engineering, as you can demonstrate how a competitor product performs better.

RBE is very simple, under subif remove bridge, place ip address, and "atm route-bridged ip". Also remove virtual reassembly.

Please remember to rate useful posts clicking on the stars below.

View solution in original post

RBE works, so you are correct--thank you!

It was surprising though, given how much cleaner RBE is, that my speeds didn't really increase appreciably.

Thanks for the tips about engineering and such. You are a great contributer to these forums.

Unfortunately a clean configuration doesn't always mean better performances, but at least you've tried what is in your power.

It would be interesting to know if the other router uses interleaving. Supposedly it should have to, as interleaving is controlled by the DSLAM. See: http://www.kitz.co.uk/adsl/interleaving.htm

Thanks for the nice rating and good luck!

View solution in original post

Thanks for the link. Looks like a good site.

Was removing ip virtual-reassembly just for the ATM0.1 subinterface and/or for all interfaces on the router? Does this mean the router will drop fragmented packets?

No, it will forward forward fragments. It is supposedly done to help consistency of ACL, NAT and so.

But all works better without it.

View solution in original post