Final Conclusion Re: Question Regarding HDLC [7:60337]
Hi Priscilla/All, Thanx for the reply.all are absolutely right when keepalives are exchanged they are not compressed.i would like to mention here only keepalives.I debug the o/p again ,even the cdp frames are compressed. Like to share the Debug O/P with u all for dummy setup client--server(DCE) 1.1.1.1 1.1.1.2 cdp enabled cdp enabled Stac enabled no compression In this scenario the keepalives are exchanged properly and the link status also remains up. Debug all client side# 00:15:55: Serial0: HDLC myseq 27, mineseen 27*, yourseen 28, line up 00:15:55: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14 00:15:57: CDP-PA: Packet received from server on interface Serial0 00:16:05: Serial0: HDLC myseq 28, mineseen 28*, yourseen 29, line up 00:16:05: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14 00:16:15: Serial0: HDLC myseq 29, mineseen 29*, yourseen 30, line up 00:16:15: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14PQU 00:16:25: Serial0: HDLC myseq 30, mineseen 30*, yourseen 31, line up 00:16:25: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14PQU debug all server# *Mar 1 00:15:35.791: Serial0: HDLC myseq 27, mineseen 27*, yourseen 27, line up * *Mar 1 00:15:45.791: Serial0: HDLC myseq 28, mineseen 28*, yourseen 28, line up *Mar 1 00:15:55.791: Serial0: HDLC myseq 29, mineseen 29*, yourseen 29, line up *Mar 1 00:16:05.791: Serial0: HDLC myseq 30, mineseen 30*, yourseen 30, line up *Mar 1 00:16:08.815: crm_send_periodic_update *Mar 1 00:16:09.763: IP-ST: if_list try 0 *Mar 1 00:16:09.763: IP-ST: gw_list total 0, try 0, completed list TRUE *Mar 1 00:16:09.763: IP-Static: all_list, time elapsed 0 msPQUICC_FEC(0/0): PHY *Mar 1 00:16:15.791: Serial0: HDLC myseq 31, mineseen 31*, yourseen 31, line up *Mar 1 00:16:25.211: CDP-EV: Bad checksum in header *Mar 1 00:16:25.791: Serial0: HDLC myseq 32, mineseen 32*, yourseen 32, line up See the Cdp Bad checksum error in header. client side# sh cdp nei Device IDLocal Intrfce HoldtmeCapability Platform Port ID server Ser 0 171 R1721 Ser 0 This Indicates that still I can accept Uncompressed packets although compression is enabled.But when it will send packets it will only send compressed packets. server side# sh cdp nei nothing is displayed Client side# ping 1.1.1.2(server) Server side#debug all *Mar 1 00:18:50.663: IP: s=120.7.3.160 (Serial0), d=32.64.1.230, len 6, dispose ip.formaterr *Mar 1 00:18:52.663: IP: s=112.7.3.160 (Serial0), d=32.64.1.214, len 6, dispose ip.formaterr *Mar 1 00:18:54.663: IP: s=104.7.3.160 (Serial0), d=32.64.1.200, len 6, dispose ip.formaterr *Mar 1 00:18:55.791: Serial0: HDLC myseq 47, mineseen 47*, yourseen 47, line up *Mar 1 00:18:56.663: IP: s=96.7.3.160 (Serial0), d=44.120.220.46, len 6, dispos e ip.formaterr * ial0: HDLC myseq 55, mineseen 55*, yourseen 55, line up Again the errors as it doesnot understand compressed data.But what are these ips ,ok the client before sending the packets on the interface have compressed the data thats why it shows strange ips after conversion.It is unable to decompress the data and displaying compressed data. anyway the question was whether keepalives are compressed are not so I used show compress which gave me the indication that no counters get incremented for compressed stats when keepalives are exchanged,but uncompressed sats were increasing that too with 20 byte increment that what i suspect is Compression stats messages which are exchanged although here on ine side command worked as compression was not enabled.this might not be a keepalive increment as keepalives as nothing to do with compression. Now for tcp header compression if enabled on one side I was unable to send tcp traffic from any side but we can ping as it doesnot use layer 4.. Thanx for all answers Bye Priscilla Oppenheimer wrote: HDLC sequences numbers aren't in data frames. They are in separate keepalive frames. They aren't like TCP sequence numbers, which sequence the data. They aren't in the header of the data frame. They are in separate frames in the control plane. Which, to make a long and winding story short, probably supports our theory that the Cisco HDLC sequence numbers are not compressed. But it's not possible to tell from Cisco documentation and sniffing is difficult because most of us can't afford a WAN sniffer. But I think that what the original poster is seeing may prove the point also. Priscilla The Long and Winding Road wrote: Priscilla Oppenheimer wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... WAN compression usually compresses the data payload. The HDLC sequence numbers are not in data packets; they are in keepalive packets. They are in the control plane, not the user plane. I can't say for sure, but my guess
Re: Final Conclusion Re: Question Regarding HDLC [7:60337]
nice job of examination and observation. thanks. may I suggest that CDP packets, as with ftp, tftp, or any other data packets, are payload to the HDLC frame. -- TANSTAAFL there ain't no such thing as a free lunch Simmi Singla wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... Hi Priscilla/All, Thanx for the reply.all are absolutely right when keepalives are exchanged they are not compressed.i would like to mention here only keepalives.I debug the o/p again ,even the cdp frames are compressed. Like to share the Debug O/P with u all for dummy setup client--server(DCE) 1.1.1.1 1.1.1.2 cdp enabled cdp enabled Stac enabled no compression In this scenario the keepalives are exchanged properly and the link status also remains up. Debug all client side# 00:15:55: Serial0: HDLC myseq 27, mineseen 27*, yourseen 28, line up 00:15:55: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14 00:15:57: CDP-PA: Packet received from server on interface Serial0 00:16:05: Serial0: HDLC myseq 28, mineseen 28*, yourseen 29, line up 00:16:05: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14 00:16:15: Serial0: HDLC myseq 29, mineseen 29*, yourseen 30, line up 00:16:15: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14PQU 00:16:25: Serial0: HDLC myseq 30, mineseen 30*, yourseen 31, line up 00:16:25: Serial0: COMPRESS: (expansion) status: 6, size in: 20, size out: 14PQU debug all server# *Mar 1 00:15:35.791: Serial0: HDLC myseq 27, mineseen 27*, yourseen 27, line up * *Mar 1 00:15:45.791: Serial0: HDLC myseq 28, mineseen 28*, yourseen 28, line up *Mar 1 00:15:55.791: Serial0: HDLC myseq 29, mineseen 29*, yourseen 29, line up *Mar 1 00:16:05.791: Serial0: HDLC myseq 30, mineseen 30*, yourseen 30, line up *Mar 1 00:16:08.815: crm_send_periodic_update *Mar 1 00:16:09.763: IP-ST: if_list try 0 *Mar 1 00:16:09.763: IP-ST: gw_list total 0, try 0, completed list TRUE *Mar 1 00:16:09.763: IP-Static: all_list, time elapsed 0 msPQUICC_FEC(0/0): PHY *Mar 1 00:16:15.791: Serial0: HDLC myseq 31, mineseen 31*, yourseen 31, line up *Mar 1 00:16:25.211: CDP-EV: Bad checksum in header *Mar 1 00:16:25.791: Serial0: HDLC myseq 32, mineseen 32*, yourseen 32, line up See the Cdp Bad checksum error in header. client side# sh cdp nei Device IDLocal Intrfce HoldtmeCapability Platform Port ID server Ser 0 171 R1721 Ser 0 This Indicates that still I can accept Uncompressed packets although compression is enabled.But when it will send packets it will only send compressed packets. server side# sh cdp nei nothing is displayed Client side# ping 1.1.1.2(server) Server side#debug all *Mar 1 00:18:50.663: IP: s=120.7.3.160 (Serial0), d=32.64.1.230, len 6, dispose ip.formaterr *Mar 1 00:18:52.663: IP: s=112.7.3.160 (Serial0), d=32.64.1.214, len 6, dispose ip.formaterr *Mar 1 00:18:54.663: IP: s=104.7.3.160 (Serial0), d=32.64.1.200, len 6, dispose ip.formaterr *Mar 1 00:18:55.791: Serial0: HDLC myseq 47, mineseen 47*, yourseen 47, line up *Mar 1 00:18:56.663: IP: s=96.7.3.160 (Serial0), d=44.120.220.46, len 6, dispos e ip.formaterr * ial0: HDLC myseq 55, mineseen 55*, yourseen 55, line up Again the errors as it doesnot understand compressed data.But what are these ips ,ok the client before sending the packets on the interface have compressed the data thats why it shows strange ips after conversion.It is unable to decompress the data and displaying compressed data. anyway the question was whether keepalives are compressed are not so I used show compress which gave me the indication that no counters get incremented for compressed stats when keepalives are exchanged,but uncompressed sats were increasing that too with 20 byte increment that what i suspect is Compression stats messages which are exchanged although here on ine side command worked as compression was not enabled.this might not be a keepalive increment as keepalives as nothing to do with compression. Now for tcp header compression if enabled on one side I was unable to send tcp traffic from any side but we can ping as it doesnot use layer 4.. Thanx for all answers Bye Priscilla Oppenheimer wrote: HDLC sequences numbers aren't in data frames. They are in separate keepalive frames. They aren't like TCP sequence numbers, which sequence the data. They aren't in the header of the data frame. They are in separate frames in the control plane. Which, to make a long and winding story short, probably supports our theory that the Cisco HDLC sequence numbers are not compressed. But it's not possible to tell from Cisco documentation and sniffing is difficult because most of us can't afford a WAN sniffer. But I think that what the original poster is seeing may prove the point
Question Regarding HDLC [7:60337]
Hi All, I have question regarding HDLC,a silly question but still a doubt.See I have HDLC connection back to back.on one interface I configure compression and other interface on other router no compression.Now when I debug the my seq and mine seen no.s are in sync I mean same they inncrease ,that different no layer 3 communication I am able to do.If such case occurs how from debug commands I will come to know. I enabled debug serial interface but it only shows me that the no get increment that too in proper order but I have read that in Karl Solie book Vol1 that it will stop increment the keepalives. So can anybody guide me. Message Posted at: http://www.groupstudy.com/form/read.php?f=7i=60337t=60337 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Question Regarding HDLC [7:60337]
WAN compression usually compresses the data payload. The HDLC sequence numbers are not in data packets; they are in keepalive packets. They are in the control plane, not the user plane. I can't say for sure, but my guess is that they are not compressed. If they were, the interfaces wouldn't have a way to monitor the status of the link. What does Cisco documentation say? Sorry I'm in a rush and can't look it up. Good question! Thanks for returning the forum to something meaningfull. Priscilla Simmi Singla wrote: Hi All, I have question regarding HDLC,a silly question but still a doubt.See I have HDLC connection back to back.on one interface I configure compression and other interface on other router no compression.Now when I debug the my seq and mine seen no.s are in sync I mean same they inncrease ,that different no layer 3 communication I am able to do.If such case occurs how from debug commands I will come to know. I enabled debug serial interface but it only shows me that the no get increment that too in proper order but I have read that in Karl Solie book Vol1 that it will stop increment the keepalives. So can anybody guide me. Message Posted at: http://www.groupstudy.com/form/read.php?f=7i=60356t=60337 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Question Regarding HDLC [7:60337]
Priscilla Oppenheimer wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... WAN compression usually compresses the data payload. The HDLC sequence numbers are not in data packets; they are in keepalive packets. They are in the control plane, not the user plane. I can't say for sure, but my guess is that they are not compressed. If they were, the interfaces wouldn't have a way to monitor the status of the link. What does Cisco documentation say? Sorry I'm in a rush and can't look it up. I checked the 12.1 docs. no help there. both the config guide and the command reference talk about frame compression I believe the implication is that the data itself is compressed, but the frame header is not. Separately, there are sections on RTP header compression, but that is a different issue. Let me step out on a limb here and postulate that in terms of process, the router first compresses the data, then attaches the HDLC frame. Much the same way that when compression is uese on the LAN, the data is compressed, but the L2 header is not. Anyone snifed this and have a scientific answer? Good question! Thanks for returning the forum to something meaningfull. Priscilla Simmi Singla wrote: Hi All, I have question regarding HDLC,a silly question but still a doubt.See I have HDLC connection back to back.on one interface I configure compression and other interface on other router no compression.Now when I debug the my seq and mine seen no.s are in sync I mean same they inncrease ,that different no layer 3 communication I am able to do.If such case occurs how from debug commands I will come to know. I enabled debug serial interface but it only shows me that the no get increment that too in proper order but I have read that in Karl Solie book Vol1 that it will stop increment the keepalives. So can anybody guide me. Message Posted at: http://www.groupstudy.com/form/read.php?f=7i=60358t=60337 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Question Regarding HDLC [7:60337]
HDLC sequences numbers aren't in data frames. They are in separate keepalive frames. They aren't like TCP sequence numbers, which sequence the data. They aren't in the header of the data frame. They are in separate frames in the control plane. Which, to make a long and winding story short, probably supports our theory that the Cisco HDLC sequence numbers are not compressed. But it's not possible to tell from Cisco documentation and sniffing is difficult because most of us can't afford a WAN sniffer. But I think that what the original poster is seeing may prove the point also. Priscilla The Long and Winding Road wrote: Priscilla Oppenheimer wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... WAN compression usually compresses the data payload. The HDLC sequence numbers are not in data packets; they are in keepalive packets. They are in the control plane, not the user plane. I can't say for sure, but my guess is that they are not compressed. If they were, the interfaces wouldn't have a way to monitor the status of the link. What does Cisco documentation say? Sorry I'm in a rush and can't look it up. I checked the 12.1 docs. no help there. both the config guide and the command reference talk about frame compression I believe the implication is that the data itself is compressed, but the frame header is not. Separately, there are sections on RTP header compression, but that is a different issue. Let me step out on a limb here and postulate that in terms of process, the router first compresses the data, then attaches the HDLC frame. Much the same way that when compression is uese on the LAN, the data is compressed, but the L2 header is not. Anyone snifed this and have a scientific answer? Good question! Thanks for returning the forum to something meaningfull. Priscilla Simmi Singla wrote: Hi All, I have question regarding HDLC,a silly question but still a doubt.See I have HDLC connection back to back.on one interface I configure compression and other interface on other router no compression.Now when I debug the my seq and mine seen no.s are in sync I mean same they inncrease ,that different no layer 3 communication I am able to do.If such case occurs how from debug commands I will come to know. I enabled debug serial interface but it only shows me that the no get increment that too in proper order but I have read that in Karl Solie book Vol1 that it will stop increment the keepalives. So can anybody guide me. Message Posted at: http://www.groupstudy.com/form/read.php?f=7i=60364t=60337 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
