Re: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
On Fri, 25 Jul 2025 14:39:49 +0400 (+04) Ivan Malov wrote: > Hi Ernesto, > > On Fri, 25 Jul 2025, Ernesto Ruffini wrote: > > > Hi, > > Did you try using usertools/dpdk-telemetry.py? > > It attaches to a running DPDK process (so your sending process should stay > > running and not close immediately), and can show many counters, including > > packets output by a device. > > At the command prompt, you can type "/ethdev/stats,0" and it will show all > > input and output packet counters for each configured queue in device 0. > > > > https://doc.dpdk.org/guides/howto/telemetry.html > > Thanks, that may indeed be helpful. But even without a secondary process, just > inserting one invocation of statistics API and a printout to the port stop > sequence of the target application would suffice to show the drops (if any). > > Thank you. The DPDK dumpcap tool also adds statistics at end of capture into the pcapng format file.
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi Ernesto, On Fri, 25 Jul 2025, Ernesto Ruffini wrote: Hi, Did you try using usertools/dpdk-telemetry.py? It attaches to a running DPDK process (so your sending process should stay running and not close immediately), and can show many counters, including packets output by a device. At the command prompt, you can type "/ethdev/stats,0" and it will show all input and output packet counters for each configured queue in device 0. https://doc.dpdk.org/guides/howto/telemetry.html Thanks, that may indeed be helpful. But even without a secondary process, just inserting one invocation of statistics API and a printout to the port stop sequence of the target application would suffice to show the drops (if any). Thank you. Regards, Ernesto -Original Message- From: Gokul K R (MS/ETA7-ETAS) Sent: Friday, July 25, 2025 08:33 To: Ivan Malov Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi Ivan, Yes, I rechecked. The PCAP file has the correct source and destination MAC addresses, and promiscuous mode is enabled on receiver NIC. Question 1: Is there a recommended way to verify whether packets are actually being transmitted from the DPDK-bound NIC? Mit freundlichen Grüßen / Best regards K R Gokul -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 10:06 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi, In the meantime, I was able to successfully replay my PCAP file using the dpdk-burst-replay tool from your repository (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from the tool: sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. -> Needed MBUF size: 174 -> Needed number of MBUFS: 100 -> Needed Memory = 16.992 Mo -> Needed Hugepages of 1 Go = 1 -> Needed CPUs: 1 -> Create mempool of 100 mbufs of 174 octs. -> Will cache 100 pkts on 1 caches. file read at 100.00% -> NIC port 0 ready. RESULTS : [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts dropped) TOTAL: 1.703 Gbit/s. 3874767.514 pps. Total dropped: 0/100 packets (0.00%) Cannot close started device (port 0) I have a question: #1: The tool shows that packets are transmitted, but they are not received on the other end. The same setup works fine with the testpmd application. Since the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or tcpdump to verify transmission. Is there any recommended way to confirm whether packets are actually being sent from the DPDK-bound NIC? Once again, are you sure destination MAC addresses in the replayed packets match that of the receiver NIC? Have you enabled promiscuous mode on the receiver? Thank you. In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to replay my PCAP file. Mit freundlichen Grüßen / Best regards K R Gokul Compact Devices, ESxx, Product and System Acceptance Tests (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | [email protected] Registered Office: Stuttgart, Registration Court: Amtsgericht Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 3:55 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan, I hope you're doing well. Point #1: Since the DPDK-Burst Replay tool is not an official part of the DPDK project and is community-developed, could you please let us know whom we should contact for support or further information regarding this tool? For the very same reason (not an in-house tool), I have no way of knowing. Perhaps take a look at repository owner's profile on GitHub or something. Point #2: We have identified an issue while using dpdk-burst replay tool. During EAL mempool initialization, “--pci-whitelist” parameter is given as an argumen
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi, Did you try using usertools/dpdk-telemetry.py? It attaches to a running DPDK process (so your sending process should stay running and not close immediately), and can show many counters, including packets output by a device. At the command prompt, you can type "/ethdev/stats,0" and it will show all input and output packet counters for each configured queue in device 0. https://doc.dpdk.org/guides/howto/telemetry.html Regards, Ernesto -Original Message- From: Gokul K R (MS/ETA7-ETAS) Sent: Friday, July 25, 2025 08:33 To: Ivan Malov Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi Ivan, Yes, I rechecked. The PCAP file has the correct source and destination MAC addresses, and promiscuous mode is enabled on receiver NIC. Question 1: Is there a recommended way to verify whether packets are actually being transmitted from the DPDK-bound NIC? Mit freundlichen Grüßen / Best regards K R Gokul -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 10:06 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > Hi, > > In the meantime, I was able to successfully replay my PCAP file using the > dpdk-burst-replay tool from your repository > (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from > the tool: > > sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading > wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read > 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. > -> Needed MBUF size: 174 > -> Needed number of MBUFS: 100 > -> Needed Memory = 16.992 Mo > -> Needed Hugepages of 1 Go = 1 > -> Needed CPUs: 1 > -> Create mempool of 100 mbufs of 174 octs. > -> Will cache 100 pkts on 1 caches. > file read at 100.00% > -> NIC port 0 ready. > > RESULTS : > [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts > dropped) > TOTAL: 1.703 Gbit/s. 3874767.514 pps. > Total dropped: 0/100 packets (0.00%) Cannot close started device > (port 0) I have a question: > > #1: The tool shows that packets are transmitted, but they are not received on > the other end. The same setup works fine with the testpmd application. Since > the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or > tcpdump to verify transmission. Is there any recommended way to confirm > whether packets are actually being sent from the DPDK-bound NIC? Once again, are you sure destination MAC addresses in the replayed packets match that of the receiver NIC? Have you enabled promiscuous mode on the receiver? Thank you. > > In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to > replay my PCAP file. > > > Mit freundlichen Grüßen / Best regards > > K R Gokul > > Compact Devices, ESxx, Product and System Acceptance Tests > (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart > | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | > [email protected] > > Registered Office: Stuttgart, Registration Court: Amtsgericht > Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. > Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. > Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus > Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert > > -Original Message- > From: Ivan Malov > Sent: Thursday, July 24, 2025 3:55 PM > To: Gokul K R (MS/ETA7-ETAS) > Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram > (ETAS-ICA/XPC-Fe6) ; Aloysius > Nishanth Britto (MS/ETA7-ETAS) > Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission > Observed Despite Successful Replay > > Hi, > > On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > >> Hi Ivan, >> >> I hope you're doing well. >> >> Point #1: >> Since the DPDK-Burst Replay tool is not an official part of the DPDK project >> and is community-developed, could you please let us know whom we should >> contact for support or further information regarding this tool? > > For the very same reason (not an in-house tool), I have no way of knowing. > Perhaps take a look at repository owner's profile on GitHub or something. > >> >> Point #2: >> We have identified an is
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi Ivan, Yes, I rechecked. The PCAP file has the correct source and destination MAC addresses, and promiscuous mode is enabled on receiver NIC. Question 1: Is there a recommended way to verify whether packets are actually being transmitted from the DPDK-bound NIC? Mit freundlichen Grüßen / Best regards K R Gokul -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 10:06 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > Hi, > > In the meantime, I was able to successfully replay my PCAP file using the > dpdk-burst-replay tool from your repository > (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from > the tool: > > sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading > wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read > 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. > -> Needed MBUF size: 174 > -> Needed number of MBUFS: 100 > -> Needed Memory = 16.992 Mo > -> Needed Hugepages of 1 Go = 1 > -> Needed CPUs: 1 > -> Create mempool of 100 mbufs of 174 octs. > -> Will cache 100 pkts on 1 caches. > file read at 100.00% > -> NIC port 0 ready. > > RESULTS : > [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts > dropped) > TOTAL: 1.703 Gbit/s. 3874767.514 pps. > Total dropped: 0/100 packets (0.00%) Cannot close started device > (port 0) I have a question: > > #1: The tool shows that packets are transmitted, but they are not received on > the other end. The same setup works fine with the testpmd application. Since > the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or > tcpdump to verify transmission. Is there any recommended way to confirm > whether packets are actually being sent from the DPDK-bound NIC? Once again, are you sure destination MAC addresses in the replayed packets match that of the receiver NIC? Have you enabled promiscuous mode on the receiver? Thank you. > > In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to > replay my PCAP file. > > > Mit freundlichen Grüßen / Best regards > > K R Gokul > > Compact Devices, ESxx, Product and System Acceptance Tests > (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart > | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | > [email protected] > > Registered Office: Stuttgart, Registration Court: Amtsgericht > Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. > Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. > Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus > Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert > > -Original Message- > From: Ivan Malov > Sent: Thursday, July 24, 2025 3:55 PM > To: Gokul K R (MS/ETA7-ETAS) > Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram > (ETAS-ICA/XPC-Fe6) ; Aloysius > Nishanth Britto (MS/ETA7-ETAS) > Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission > Observed Despite Successful Replay > > Hi, > > On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > >> Hi Ivan, >> >> I hope you're doing well. >> >> Point #1: >> Since the DPDK-Burst Replay tool is not an official part of the DPDK project >> and is community-developed, could you please let us know whom we should >> contact for support or further information regarding this tool? > > For the very same reason (not an in-house tool), I have no way of knowing. > Perhaps take a look at repository owner's profile on GitHub or something. > >> >> Point #2: >> We have identified an issue while using dpdk-burst replay tool. >> During EAL mempool initialization, “--pci-whitelist” parameter is given as >> an argument for “rte_eal_init” function. However, when rte_eal_init is >> executed with these arguments, it fails with the following error: >> >> ./dpdk-replay: unrecognized option ‘--pci-whitelist’ >> Invalid command line argument >> >> We couldn’t find any official documentation on the --pci-whitelist >> parameter. Could you help clarify its purpose and whether this argument is >> essential for the tool to operate correctly? > > 1) In 2020, commit db27370b5720 ("eal: replace blacklist/whitelist options") >renamed this argument. Now it is '-a' ('--allow') and '-b' ('
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi, In the meantime, I was able to successfully replay my PCAP file using the dpdk-burst-replay tool from your repository (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from the tool: sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. -> Needed MBUF size: 174 -> Needed number of MBUFS: 100 -> Needed Memory = 16.992 Mo -> Needed Hugepages of 1 Go = 1 -> Needed CPUs: 1 -> Create mempool of 100 mbufs of 174 octs. -> Will cache 100 pkts on 1 caches. file read at 100.00% -> NIC port 0 ready. RESULTS : [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts dropped) TOTAL: 1.703 Gbit/s. 3874767.514 pps. Total dropped: 0/100 packets (0.00%) Cannot close started device (port 0) I have a question: #1: The tool shows that packets are transmitted, but they are not received on the other end. The same setup works fine with the testpmd application. Since the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or tcpdump to verify transmission. Is there any recommended way to confirm whether packets are actually being sent from the DPDK-bound NIC? In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to replay my PCAP file. Mit freundlichen Grüßen / Best regards K R Gokul Compact Devices, ESxx, Product and System Acceptance Tests (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | [email protected] Registered Office: Stuttgart, Registration Court: Amtsgericht Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 3:55 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > Hi Ivan, > > I hope you're doing well. > > Point #1: > Since the DPDK-Burst Replay tool is not an official part of the DPDK project > and is community-developed, could you please let us know whom we should > contact for support or further information regarding this tool? For the very same reason (not an in-house tool), I have no way of knowing. Perhaps take a look at repository owner's profile on GitHub or something. > > Point #2: > We have identified an issue while using dpdk-burst replay tool. > During EAL mempool initialization, “--pci-whitelist” parameter is given as an > argument for “rte_eal_init” function. However, when rte_eal_init is executed > with these arguments, it fails with the following error: > > ./dpdk-replay: unrecognized option ‘--pci-whitelist’ > Invalid command line argument > > We couldn’t find any official documentation on the --pci-whitelist parameter. > Could you help clarify its purpose and whether this argument is essential for > the tool to operate correctly? 1) In 2020, commit db27370b5720 ("eal: replace blacklist/whitelist options") renamed this argument. Now it is '-a' ('--allow') and '-b' ('--block') [1]. 2) The argument is application-agnostic and is designed to give a finer control on which devices get picked by DPDK during EAL initialisation. 3) Whether the argument is needed for the use of this particular application is not apparent to me. One should refer to the application's documentation. > > Point #3: > Could you also suggest any official DPDK tool that supports replaying PCAP > files? This would help us explore alternatives that are actively maintained > by the DPDK community. Please take a look at DPDK pcap PMD [2]. It may meet your needs. [1] https://doc.dpdk.org/guides-25.07/linux_gsg/linux_eal_parameters.html#device-related-options [2] https://doc.dpdk.org/guides-25.07/nics/pcap_ring.html Thank you. > > Looking forward to your inputs. > > Best regards, > Gokul K R > -Original Message- > From: Ivan Malov > Sent: Wednesday, July 23, 2025 11:43 AM > To: Gokul K R (MS/ETA7-ETAS) > Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram > (ETAS-ICA/XPC-Fe6) > Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission > Observed Despite Successful Replay > > On Wed, 23 Jul 2025, Ivan Malov wrote: > >> Hi, >> >> On Wed, 23 Jul 2025, Gokul K R (MS/ETA
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi, On Fri, 25 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan, Yes, I rechecked. The PCAP file has the correct source and destination MAC addresses, and promiscuous mode is enabled on receiver NIC. Question 1: Is there a recommended way to verify whether packets are actually being transmitted from the DPDK-bound NIC? 1) One can take notes on what 'rte_eth_tx_burst' returns (in the sender code). If non-zero, then it likely submits the packets for transmission to the HW. 2) STATISTICS -- both on sender (APIs [1], [2]. [3]) and at the receiver. If the receiver is a Linux interface (not DPDK), then 'sudo ethtool -S '. Provided that 'rte_eth_tx_burst' does not return zero, are the drops reflected in ethdev statistics? Are the drops reflected in statistics at the receiver? [1] https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#adec226574c53ae413252c9b15f6f4bab [2] https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#a418ad970673eb171673185e36044fd79 [3] https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#a300d75b583c1f5acfe5b162a5d8c0ac1 Thank you. Mit freundlichen Grüßen / Best regards K R Gokul -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 10:06 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi, In the meantime, I was able to successfully replay my PCAP file using the dpdk-burst-replay tool from your repository (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from the tool: sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. -> Needed MBUF size: 174 -> Needed number of MBUFS: 100 -> Needed Memory = 16.992 Mo -> Needed Hugepages of 1 Go = 1 -> Needed CPUs: 1 -> Create mempool of 100 mbufs of 174 octs. -> Will cache 100 pkts on 1 caches. file read at 100.00% -> NIC port 0 ready. RESULTS : [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts dropped) TOTAL: 1.703 Gbit/s. 3874767.514 pps. Total dropped: 0/100 packets (0.00%) Cannot close started device (port 0) I have a question: #1: The tool shows that packets are transmitted, but they are not received on the other end. The same setup works fine with the testpmd application. Since the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or tcpdump to verify transmission. Is there any recommended way to confirm whether packets are actually being sent from the DPDK-bound NIC? Once again, are you sure destination MAC addresses in the replayed packets match that of the receiver NIC? Have you enabled promiscuous mode on the receiver? Thank you. In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to replay my PCAP file. Mit freundlichen Grüßen / Best regards K R Gokul Compact Devices, ESxx, Product and System Acceptance Tests (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | [email protected] Registered Office: Stuttgart, Registration Court: Amtsgericht Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 3:55 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan, I hope you're doing well. Point #1: Since the DPDK-Burst Replay tool is not an official part of the DPDK project and is community-developed, could you please let us know whom we should contact for support or further information regarding this tool? For the very same reason (not an in-house tool), I have no way of knowing. Perhaps take a look at repository owner's profile on GitHub or something. Point #2: We have identified an issue while using dpdk-burst replay tool. During EAL mempool initialization, “--pci-whitelist” parameter is given as an argument for “rte_eal_init” function. However, when rte_eal_init is executed with these arguments, it fails with the following error: ./dpdk-replay: unrecognized option ‘--pci-whitelist’ Invalid command line argument We couldn’t find any official documentation on the --pci-whit
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi, In the meantime, I was able to successfully replay my PCAP file using the dpdk-burst-replay tool from your repository (https://github.com/FraudBuster/dpdk-burst-replay). Below is the output from the tool: sudo ./src/dpdk-replay wendutput.pcap :01:00.0 preloading wendutput.pcap file (of size: 5900 bytes) file read at 100.00% read 100 pkts (for a total of 5900 bytes). max packet length = 43 bytes. -> Needed MBUF size: 174 -> Needed number of MBUFS: 100 -> Needed Memory = 16.992 Mo -> Needed Hugepages of 1 Go = 1 -> Needed CPUs: 1 -> Create mempool of 100 mbufs of 174 octs. -> Will cache 100 pkts on 1 caches. file read at 100.00% -> NIC port 0 ready. RESULTS : [thread 00]: 1.703287 Gbit/s, 3874767.513949 pps on 0.26 sec (0 pkts dropped) TOTAL: 1.703 Gbit/s. 3874767.514 pps. Total dropped: 0/100 packets (0.00%) Cannot close started device (port 0) I have a question: #1: The tool shows that packets are transmitted, but they are not received on the other end. The same setup works fine with the testpmd application. Since the transmit-side NIC is bound to DPDK, I cannot use tools like Wireshark or tcpdump to verify transmission. Is there any recommended way to confirm whether packets are actually being sent from the DPDK-bound NIC? Once again, are you sure destination MAC addresses in the replayed packets match that of the receiver NIC? Have you enabled promiscuous mode on the receiver? Thank you. In parallel, as per your suggestion, I’m also exploring the DPDK PCAP PMD to replay my PCAP file. Mit freundlichen Grüßen / Best regards K R Gokul Compact Devices, ESxx, Product and System Acceptance Tests (MS/ETA7-ETAS) Robert Bosch GmbH | Postfach 10 60 50 | 70049 Stuttgart | GERMANY | http://www.bosch.com/ Mobile +91-9003755978 | [email protected] Registered Office: Stuttgart, Registration Court: Amtsgericht Stuttgart, HRB 14000; Chairman of the Supervisory Board: Prof. Dr. Stefan Asenkerschbaumer; Managing Directors: Dr. Stefan Hartung, Dr. Christian Fischer, Dr. Markus Forschner, Stefan Grosch, Dr. Markus Heyn, Dr. Frank Meyer, Katja von Raven, Dr. Tanja Rückert -Original Message- From: Ivan Malov Sent: Thursday, July 24, 2025 3:55 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) ; Aloysius Nishanth Britto (MS/ETA7-ETAS) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan, I hope you're doing well. Point #1: Since the DPDK-Burst Replay tool is not an official part of the DPDK project and is community-developed, could you please let us know whom we should contact for support or further information regarding this tool? For the very same reason (not an in-house tool), I have no way of knowing. Perhaps take a look at repository owner's profile on GitHub or something. Point #2: We have identified an issue while using dpdk-burst replay tool. During EAL mempool initialization, “--pci-whitelist” parameter is given as an argument for “rte_eal_init” function. However, when rte_eal_init is executed with these arguments, it fails with the following error: ./dpdk-replay: unrecognized option ‘--pci-whitelist’ Invalid command line argument We couldn’t find any official documentation on the --pci-whitelist parameter. Could you help clarify its purpose and whether this argument is essential for the tool to operate correctly? 1) In 2020, commit db27370b5720 ("eal: replace blacklist/whitelist options") renamed this argument. Now it is '-a' ('--allow') and '-b' ('--block') [1]. 2) The argument is application-agnostic and is designed to give a finer control on which devices get picked by DPDK during EAL initialisation. 3) Whether the argument is needed for the use of this particular application is not apparent to me. One should refer to the application's documentation. Point #3: Could you also suggest any official DPDK tool that supports replaying PCAP files? This would help us explore alternatives that are actively maintained by the DPDK community. Please take a look at DPDK pcap PMD [2]. It may meet your needs. [1] https://doc.dpdk.org/guides-25.07/linux_gsg/linux_eal_parameters.html#device-related-options [2] https://doc.dpdk.org/guides-25.07/nics/pcap_ring.html Thank you. Looking forward to your inputs. Best regards, Gokul K R -Original Message- From: Ivan Malov Sent: Wednesday, July 23, 2025 11:43 AM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Wed, 23 Jul 2025, Ivan Malov wrote: Hi, On Wed, 23 Jul 2025, Gokul K
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi Ivan, I hope you're doing well. Point #1: Since the DPDK-Burst Replay tool is not an official part of the DPDK project and is community-developed, could you please let us know whom we should contact for support or further information regarding this tool? Point #2: We have identified an issue while using dpdk-burst replay tool. During EAL mempool initialization, “--pci-whitelist” parameter is given as an argument for “rte_eal_init” function. However, when rte_eal_init is executed with these arguments, it fails with the following error: ./dpdk-replay: unrecognized option ‘--pci-whitelist’ Invalid command line argument We couldn’t find any official documentation on the --pci-whitelist parameter. Could you help clarify its purpose and whether this argument is essential for the tool to operate correctly? Point #3: Could you also suggest any official DPDK tool that supports replaying PCAP files? This would help us explore alternatives that are actively maintained by the DPDK community. Looking forward to your inputs. Best regards, Gokul K R -Original Message- From: Ivan Malov Sent: Wednesday, July 23, 2025 11:43 AM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6) Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay On Wed, 23 Jul 2025, Ivan Malov wrote: > Hi, > > On Wed, 23 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > >> Hi Ivan Malov 😊 >> >> Use Case: >> I'm currently using the DPDK-Burst-Replay tool to replay captured >> PCAP files at specific data rates (e.g., 150–200 Mbps). >> >> Response to your feedback: >> Point 1: "Port 0 is not on the good NUMA ID (-1)" >> I’m aware that this message is printed due to the NUMA ID being >> returned as -1. >> I've just started diving into the source code and found that the call >> to >> rte_eth_dev_socket_id() returns -1, which typically indicates an error. > > No, -1 typically translates to 'SOCKET_ID_ANY'. In order to rule out 'EINVAL' > in 'rte_errno', one should attempt to invoke 'rte_eth_dev_socket_id' > within the loop of 'RTE_ETH_FOREACH_DEV' (see examples in DPDK) and > print the socket ID. > >> However, the current implementation does not output the rte_errno, >> which could help identify the root cause. I'm working on modifying >> the code to print the error code for better debugging. >> >> Point 2: "NIC Link is UP" >> Yes, on the NIC side, the link is up. I'm also able to transmit >> packets successfully using the testpmd application. >> >> Question: >> Could you please confirm if the current version of the >> DPDK-Burst-Replay tool supports replaying Ethernet frames larger than >> 64 bytes (e.g., up to >> 1500 bytes)? Or has the tool been enhanced to support this use case? > > The tool seems like an external application. Try to look for any > mentions of API 'rte_eth_dev_set_mtu' or just the term 'mtu' in that > source code. If there are no such mentions, then default MTU applies, > which depends on the driver. > > Have you tried querying statistics to find the cause of the drops? Also, given the fact that the application replays some pcap traffic, have you made sure the receiver (where you seek to watch the replayed traffic arrive) has got 'promiscuous' mode [1] enabled? IIRC, 'test-pmd' enables it by default. [1] https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#a5dd1dedaa45f05c72bcc35495e441e91 Thank you. > > Thank you. > >> >> Thanks for your time and support! >> >> Best regards, >> Gokul K.R >> >> >> >> >> >> -Original Message- >> From: Ivan Malov >> Sent: Friday, July 18, 2025 5:29 PM >> To: Gokul K R (MS/ETA7-ETAS) >> Cc: [email protected]; [email protected] >> Subject: Re: Issue with DPDK-Burst Replay – No Frame Transmission >> Observed Despite Successful Replay >> >> Hi, >> >> (please see below) >> >> On Fri, 18 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: >> >>> >>> Hi Team, >>> >>> I’m currently working with the dpdk-burst-replay tool and >>> encountered an issue during execution. Below are the details: >>> >>> >>> >>> __ >>> >>> __ __ >>> >>> >>> Observation: >>> During replay, we received the follow
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi,
On Thu, 24 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote:
Hi Ivan,
I hope you're doing well.
Point #1:
Since the DPDK-Burst Replay tool is not an official part of the DPDK project
and is community-developed, could you please let us know whom we should contact
for support or further information regarding this tool?
For the very same reason (not an in-house tool), I have no way of knowing.
Perhaps take a look at repository owner's profile on GitHub or something.
Point #2:
We have identified an issue while using dpdk-burst replay tool.
During EAL mempool initialization, “--pci-whitelist” parameter is given as an
argument for “rte_eal_init” function. However, when rte_eal_init is executed
with these arguments, it fails with the following error:
./dpdk-replay: unrecognized option ‘--pci-whitelist’
Invalid command line argument
We couldn’t find any official documentation on the --pci-whitelist parameter.
Could you help clarify its purpose and whether this argument is essential for
the tool to operate correctly?
1) In 2020, commit db27370b5720 ("eal: replace blacklist/whitelist options")
renamed this argument. Now it is '-a' ('--allow') and '-b' ('--block') [1].
2) The argument is application-agnostic and is designed to give a finer control
on which devices get picked by DPDK during EAL initialisation.
3) Whether the argument is needed for the use of this particular application
is not apparent to me. One should refer to the application's documentation.
Point #3:
Could you also suggest any official DPDK tool that supports replaying PCAP
files? This would help us explore alternatives that are actively maintained by
the DPDK community.
Please take a look at DPDK pcap PMD [2]. It may meet your needs.
[1]
https://doc.dpdk.org/guides-25.07/linux_gsg/linux_eal_parameters.html#device-related-options
[2] https://doc.dpdk.org/guides-25.07/nics/pcap_ring.html
Thank you.
Looking forward to your inputs.
Best regards,
Gokul K R
-Original Message-
From: Ivan Malov
Sent: Wednesday, July 23, 2025 11:43 AM
To: Gokul K R (MS/ETA7-ETAS)
Cc: [email protected]; [email protected]; Nivethitha N Shanmugasundaram (ETAS-ICA/XPC-Fe6)
Subject: RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed
Despite Successful Replay
On Wed, 23 Jul 2025, Ivan Malov wrote:
Hi,
On Wed, 23 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote:
Hi Ivan Malov 😊
Use Case:
I'm currently using the DPDK-Burst-Replay tool to replay captured
PCAP files at specific data rates (e.g., 150–200 Mbps).
Response to your feedback:
Point 1: "Port 0 is not on the good NUMA ID (-1)"
I’m aware that this message is printed due to the NUMA ID being
returned as -1.
I've just started diving into the source code and found that the call
to
rte_eth_dev_socket_id() returns -1, which typically indicates an error.
No, -1 typically translates to 'SOCKET_ID_ANY'. In order to rule out 'EINVAL'
in 'rte_errno', one should attempt to invoke 'rte_eth_dev_socket_id'
within the loop of 'RTE_ETH_FOREACH_DEV' (see examples in DPDK) and
print the socket ID.
However, the current implementation does not output the rte_errno,
which could help identify the root cause. I'm working on modifying
the code to print the error code for better debugging.
Point 2: "NIC Link is UP"
Yes, on the NIC side, the link is up. I'm also able to transmit
packets successfully using the testpmd application.
Question:
Could you please confirm if the current version of the
DPDK-Burst-Replay tool supports replaying Ethernet frames larger than
64 bytes (e.g., up to
1500 bytes)? Or has the tool been enhanced to support this use case?
The tool seems like an external application. Try to look for any
mentions of API 'rte_eth_dev_set_mtu' or just the term 'mtu' in that
source code. If there are no such mentions, then default MTU applies,
which depends on the driver.
Have you tried querying statistics to find the cause of the drops?
Also, given the fact that the application replays some pcap traffic, have you
made sure the receiver (where you seek to watch the replayed traffic arrive)
has got 'promiscuous' mode [1] enabled? IIRC, 'test-pmd' enables it by default.
[1]
https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#a5dd1dedaa45f05c72bcc35495e441e91
Thank you.
Thank you.
Thanks for your time and support!
Best regards,
Gokul K.R
-----Original Message-----
From: Ivan Malov
Sent: Friday, July 18, 2025 5:29 PM
To: Gokul K R (MS/ETA7-ETAS)
Cc: [email protected]; [email protected]
Subject: Re: Issue with DPDK-Burst Replay – No Frame Transmission
Observed Despite Successful Replay
Hi,
(please s
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi Ivan Malov 😊 Use Case: I'm currently using the DPDK-Burst-Replay tool to replay captured PCAP files at specific data rates (e.g., 150–200 Mbps). Response to your feedback: Point 1: "Port 0 is not on the good NUMA ID (-1)" I’m aware that this message is printed due to the NUMA ID being returned as -1. I've just started diving into the source code and found that the call to rte_eth_dev_socket_id() returns -1, which typically indicates an error. However, the current implementation does not output the rte_errno, which could help identify the root cause. I'm working on modifying the code to print the error code for better debugging. Point 2: "NIC Link is UP" Yes, on the NIC side, the link is up. I'm also able to transmit packets successfully using the testpmd application. Question: Could you please confirm if the current version of the DPDK-Burst-Replay tool supports replaying Ethernet frames larger than 64 bytes (e.g., up to 1500 bytes)? Or has the tool been enhanced to support this use case? Thanks for your time and support! Best regards, Gokul K.R -Original Message- From: Ivan Malov Sent: Friday, July 18, 2025 5:29 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected] Subject: Re: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, (please see below) On Fri, 18 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: > > Hi Team, > > I’m currently working with the dpdk-burst-replay tool and encountered an > issue during execution. Below are the details: > > > __ > __ > __ > > > Observation: > During replay, we received the following informational message: > > port 0 is not on the good numa id (-1) Which API was used to check this? Was API [1] used? If not, what does it show in the absence of 'numactl' command? [1] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#ad032e25f712e6ffeb0c19eab1ec1fd2e > > As per the DPDK mailing list discussions, this warning is typically > benign—often seen on NICs like Intel I225/I210, which do not report NUMA > affinity. Hence, we proceeded with execution. > > > __ > __ > __ > > > Command Used: > > sudo numactl --cpunodebind=0 --membind=0 ./src/dpdk-replay > Original_MAC.pcap :01:00.1 > > Execution Output: > > preloading Original_MAC.pcap file (of size: 143959 bytes) > > file read at 100.00% > > read 675 pkts (for a total of 143959 bytes). max packet length = 1518 bytes. > > -> Needed MBUF size: 1648 > > -> Needed number of MBUFs: 675 > > -> Needed Memory = 1.061 MB > > -> Needed Hugepages of 1 GB = 1 > > -> Needed CPUs: 1 > > -> Create mempool of 675 mbufs of 1648 octets. > > -> Will cache 675 pkts on 1 caches. What does this 'cache' stand for? Does it refer to the mempool per-lcore cache? If so, please note that, according to API [2] documentation, cache size "must be lower or equal to RTE_MEMPOOL_CACHE_MAX_SIZE and n / 1.5", where 'n' stands for the number of mbufs. Also, documentation says it is advised to choose cache_size to have "n modulo cache_size == 0". Does your code meet these requirements? By the looks of it, it doesn't (cache_size = n = 675). Consider to double-check. [2] https://doc.dpdk.org/api-25.03/rte__mempool_8h.html#a0b64d611bc140a4d2a0c94911580efd5 > > > __ > __ > __ > > > Issue: > Despite successful parsing of the pcap file and proper initialization, no > frames were transmitted or received on either the sender or receiver sides. Is this observation based solely on watching APIs [3] and [4] return 0 all the time? If yes, one can consider to introduce invocations of APIs [5], [6] and [7] in order to periodically poll and print statistics (may be with 1-second delay), which may, for example, shed light on mbuf allocation errors (extended stats). Do statistics display any interesting figures to be discussed? [3] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#a3e7d76a451b46348686ea97d6367f102 [4] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
On Wed, 23 Jul 2025, Ivan Malov wrote: Hi, On Wed, 23 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan Malov 😊 Use Case: I'm currently using the DPDK-Burst-Replay tool to replay captured PCAP files at specific data rates (e.g., 150–200 Mbps). Response to your feedback: Point 1: "Port 0 is not on the good NUMA ID (-1)" I’m aware that this message is printed due to the NUMA ID being returned as -1. I've just started diving into the source code and found that the call to rte_eth_dev_socket_id() returns -1, which typically indicates an error. No, -1 typically translates to 'SOCKET_ID_ANY'. In order to rule out 'EINVAL' in 'rte_errno', one should attempt to invoke 'rte_eth_dev_socket_id' within the loop of 'RTE_ETH_FOREACH_DEV' (see examples in DPDK) and print the socket ID. However, the current implementation does not output the rte_errno, which could help identify the root cause. I'm working on modifying the code to print the error code for better debugging. Point 2: "NIC Link is UP" Yes, on the NIC side, the link is up. I'm also able to transmit packets successfully using the testpmd application. Question: Could you please confirm if the current version of the DPDK-Burst-Replay tool supports replaying Ethernet frames larger than 64 bytes (e.g., up to 1500 bytes)? Or has the tool been enhanced to support this use case? The tool seems like an external application. Try to look for any mentions of API 'rte_eth_dev_set_mtu' or just the term 'mtu' in that source code. If there are no such mentions, then default MTU applies, which depends on the driver. Have you tried querying statistics to find the cause of the drops? Also, given the fact that the application replays some pcap traffic, have you made sure the receiver (where you seek to watch the replayed traffic arrive) has got 'promiscuous' mode [1] enabled? IIRC, 'test-pmd' enables it by default. [1] https://doc.dpdk.org/api-25.07/rte__ethdev_8h.html#a5dd1dedaa45f05c72bcc35495e441e91 Thank you. Thank you. Thanks for your time and support! Best regards, Gokul K.R -Original Message----- From: Ivan Malov Sent: Friday, July 18, 2025 5:29 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected] Subject: Re: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, (please see below) On Fri, 18 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Team, I’m currently working with the dpdk-burst-replay tool and encountered an issue during execution. Below are the details: __ __ __ Observation: During replay, we received the following informational message: port 0 is not on the good numa id (-1) Which API was used to check this? Was API [1] used? If not, what does it show in the absence of 'numactl' command? [1] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#ad032e25f712e6ffeb0c19eab1ec1fd2e As per the DPDK mailing list discussions, this warning is typically benign—often seen on NICs like Intel I225/I210, which do not report NUMA affinity. Hence, we proceeded with execution. __ __ __ Command Used: sudo numactl --cpunodebind=0 --membind=0 ./src/dpdk-replay Original_MAC.pcap :01:00.1 Execution Output: preloading Original_MAC.pcap file (of size: 143959 bytes) file read at 100.00% read 675 pkts (for a total of 143959 bytes). max packet length = 1518 bytes. -> Needed MBUF size: 1648 -> Needed number of MBUFs: 675 -> Needed Memory = 1.061 MB -> Needed Hugepages of 1 GB = 1 -> Needed CPUs: 1 -> Create mempool of 675 mbufs of 1648 octets. -> Will cache 675 pkts on 1 caches. What does this 'cache' stand for? Does it refer to the mempool per-lcore cache? If so, please note that, according to API [2] documentation, cache size "must be lower or equal to RTE_MEMPOOL_CACHE_MAX_SIZE and n / 1.5", where 'n' stands for the number of mbufs. Also, documentation says it is advised to choose cache_size to have "n modulo cache_size == 0". Does your code meet these requirements? By the looks of it, it doesn't (cache_size = n = 675). Consider to double-check. [2] https://doc.dpdk.org/api-25.03/rte__mempool_8h.html#a0b64d611bc140a4d2a0c94911580efd5 __ __ __ Issue: Despite successful parsing of the pc
RE: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi, On Wed, 23 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Ivan Malov 😊 Use Case: I'm currently using the DPDK-Burst-Replay tool to replay captured PCAP files at specific data rates (e.g., 150–200 Mbps). Response to your feedback: Point 1: "Port 0 is not on the good NUMA ID (-1)" I’m aware that this message is printed due to the NUMA ID being returned as -1. I've just started diving into the source code and found that the call to rte_eth_dev_socket_id() returns -1, which typically indicates an error. No, -1 typically translates to 'SOCKET_ID_ANY'. In order to rule out 'EINVAL' in 'rte_errno', one should attempt to invoke 'rte_eth_dev_socket_id' within the loop of 'RTE_ETH_FOREACH_DEV' (see examples in DPDK) and print the socket ID. However, the current implementation does not output the rte_errno, which could help identify the root cause. I'm working on modifying the code to print the error code for better debugging. Point 2: "NIC Link is UP" Yes, on the NIC side, the link is up. I'm also able to transmit packets successfully using the testpmd application. Question: Could you please confirm if the current version of the DPDK-Burst-Replay tool supports replaying Ethernet frames larger than 64 bytes (e.g., up to 1500 bytes)? Or has the tool been enhanced to support this use case? The tool seems like an external application. Try to look for any mentions of API 'rte_eth_dev_set_mtu' or just the term 'mtu' in that source code. If there are no such mentions, then default MTU applies, which depends on the driver. Have you tried querying statistics to find the cause of the drops? Thank you. Thanks for your time and support! Best regards, Gokul K.R -Original Message- From: Ivan Malov Sent: Friday, July 18, 2025 5:29 PM To: Gokul K R (MS/ETA7-ETAS) Cc: [email protected]; [email protected] Subject: Re: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay Hi, (please see below) On Fri, 18 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Team, I’m currently working with the dpdk-burst-replay tool and encountered an issue during execution. Below are the details: __ __ __ Observation: During replay, we received the following informational message: port 0 is not on the good numa id (-1) Which API was used to check this? Was API [1] used? If not, what does it show in the absence of 'numactl' command? [1] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#ad032e25f712e6ffeb0c19eab1ec1fd2e As per the DPDK mailing list discussions, this warning is typically benign—often seen on NICs like Intel I225/I210, which do not report NUMA affinity. Hence, we proceeded with execution. __ __ __ Command Used: sudo numactl --cpunodebind=0 --membind=0 ./src/dpdk-replay Original_MAC.pcap :01:00.1 Execution Output: preloading Original_MAC.pcap file (of size: 143959 bytes) file read at 100.00% read 675 pkts (for a total of 143959 bytes). max packet length = 1518 bytes. -> Needed MBUF size: 1648 -> Needed number of MBUFs: 675 -> Needed Memory = 1.061 MB -> Needed Hugepages of 1 GB = 1 -> Needed CPUs: 1 -> Create mempool of 675 mbufs of 1648 octets. -> Will cache 675 pkts on 1 caches. What does this 'cache' stand for? Does it refer to the mempool per-lcore cache? If so, please note that, according to API [2] documentation, cache size "must be lower or equal to RTE_MEMPOOL_CACHE_MAX_SIZE and n / 1.5", where 'n' stands for the number of mbufs. Also, documentation says it is advised to choose cache_size to have "n modulo cache_size == 0". Does your code meet these requirements? By the looks of it, it doesn't (cache_size = n = 675). Consider to double-check. [2] https://doc.dpdk.org/api-25.03/rte__mempool_8h.html#a0b64d611bc140a4d2a0c94911580efd5 __ __ __ Issue: Despite successful parsing of the pcap file and proper initialization, no frames were transmitted or received on either the sender or receiver sides. Is this observation based solely on watching APIs [3] and [4] return 0 all the time? If yes, one can consider to introduce invocations of APIs [5], [6] and [7] in order to periodically poll and print statistics (may be with 1-second delay), which may, for example, shed light on mbuf allocation error
Re: Issue with DPDK-Burst Replay – No Frame Transmission Observed Despite Successful Replay
Hi, (please see below) On Fri, 18 Jul 2025, Gokul K R (MS/ETA7-ETAS) wrote: Hi Team, I’m currently working with the dpdk-burst-replay tool and encountered an issue during execution. Below are the details: __ Observation: During replay, we received the following informational message: port 0 is not on the good numa id (-1) Which API was used to check this? Was API [1] used? If not, what does it show in the absence of 'numactl' command? [1] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#ad032e25f712e6ffeb0c19eab1ec1fd2e As per the DPDK mailing list discussions, this warning is typically benign—often seen on NICs like Intel I225/I210, which do not report NUMA affinity. Hence, we proceeded with execution. __ Command Used: sudo numactl --cpunodebind=0 --membind=0 ./src/dpdk-replay Original_MAC.pcap :01:00.1 Execution Output: preloading Original_MAC.pcap file (of size: 143959 bytes) file read at 100.00% read 675 pkts (for a total of 143959 bytes). max packet length = 1518 bytes. -> Needed MBUF size: 1648 -> Needed number of MBUFs: 675 -> Needed Memory = 1.061 MB -> Needed Hugepages of 1 GB = 1 -> Needed CPUs: 1 -> Create mempool of 675 mbufs of 1648 octets. -> Will cache 675 pkts on 1 caches. What does this 'cache' stand for? Does it refer to the mempool per-lcore cache? If so, please note that, according to API [2] documentation, cache size "must be lower or equal to RTE_MEMPOOL_CACHE_MAX_SIZE and n / 1.5", where 'n' stands for the number of mbufs. Also, documentation says it is advised to choose cache_size to have "n modulo cache_size == 0". Does your code meet these requirements? By the looks of it, it doesn't (cache_size = n = 675). Consider to double-check. [2] https://doc.dpdk.org/api-25.03/rte__mempool_8h.html#a0b64d611bc140a4d2a0c94911580efd5 __ Issue: Despite successful parsing of the pcap file and proper initialization, no frames were transmitted or received on either the sender or receiver sides. Is this observation based solely on watching APIs [3] and [4] return 0 all the time? If yes, one can consider to introduce invocations of APIs [5], [6] and [7] in order to periodically poll and print statistics (may be with 1-second delay), which may, for example, shed light on mbuf allocation errors (extended stats). Do statistics display any interesting figures to be discussed? [3] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#a3e7d76a451b46348686ea97d6367f102 [4] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#a83e56cabbd31637efd648e3fc010392b [5] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#adec226574c53ae413252c9b15f6f4bab [6] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#a418ad970673eb171673185e36044fd79 [7] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#a300d75b583c1f5acfe5b162a5d8c0ac1 __ Environment Details: * NIC used: Intel I225/I210 * Hugepages configured: 1 GB * NUMA binding: --cpunodebind=0 --membind=0 * OS: [Your Linux distribution, e.g., Ubuntu 20.04] * DPDK version: [Mention if known] __ Could you please advise if any additional setup, configuration, or known limitations may be impacting the packet transmission? This may be a wild suggestion from my side, but it pays to check whether link status is "up" upon port start on both ends. One can use API [8] to do that. [8] https://doc.dpdk.org/api-25.03/rte__ethdev_8h.html#ac05878578e4fd9ef3551d2c1c175ebe7 Thank you. Thank you in advance for your support! Best regards, Gokul K R
