You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have downloaded solarcapture python codebase and have enterprise onload running on xtremescale 2522+ nic. I run using python solar_capture. I am not getting packets capture statistics like tcpdump such as packet capture, dropped by kernel, dropped by interface and the basics to determine my capture performance. I can get only n_rx_pkts which is same as the number of packets in my output pcap file. Is there any solution already available or do I need to make my own custom one? Please guide me through.
The text was updated successfully, but these errors were encountered:
Detailed statistics are available via "solar_capture_monitor dump"
But the more useful place to immediately look is the interface statistic "rx_port_nodesc_drop" (in ethtool -S). This is packets dropped at the interface due to the RX ring not having space for them - which is the only place SolarCapture can cause drops. (Either directly due to insufficient capture_cores, or indirectly due to the disk writer being too slow and then back-pressuring the buffers, leading to the RX ring getting starved.)
I have downloaded solarcapture python codebase and have enterprise onload running on xtremescale 2522+ nic. I run using python solar_capture. I am not getting packets capture statistics like tcpdump such as packet capture, dropped by kernel, dropped by interface and the basics to determine my capture performance. I can get only n_rx_pkts which is same as the number of packets in my output pcap file. Is there any solution already available or do I need to make my own custom one? Please guide me through.
The text was updated successfully, but these errors were encountered: