Re: 60+% ping packet loss on Pi3 under -current and stable-13
Date: Tue, 03 May 2022 02:23:10 UTC
On Mon, May 02, 2022 at 06:48:58PM +0200, Hans Petter Selasky wrote: > > You may also want to correlate: > > usbconfig dump_stats > On the -current Pi3 I see in part: ugen1.4: <JMicron SABRENT> at usbus1, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=ON (500mA) { UE_CONTROL_OK : 83 UE_ISOCHRONOUS_OK : 0 UE_BULK_OK : 3730195 UE_INTERRUPT_OK : 0 UE_CONTROL_FAIL : 0 UE_ISOCHRONOUS_FAIL : 0 UE_BULK_FAIL : 61 UE_INTERRUPT_FAIL : 0 } That's the USB3 hard disk boot drive. All other devices report zero bulk fails. > with netstat for ue0 . > On the same machine one gets: root@www:/home/bob # netstat -i -n ue0 Name Mtu Network Address Ipkts Ierrs Idrop Opkts Oerrs Coll lo0 16384 <Link#1> lo0 185 0 0 185 0 0 lo0 - ::1/128 ::1 0 - - 0 - - lo0 - fe80::%lo0/64 fe80::1%lo0 0 - - 0 - - lo0 - 127.0.0.0/8 127.0.0.1 185 - - 185 - - ue0 1500 <Link#2> b8:27:eb:71:46:4e 244278 0 0 213240 0 0 ue0 - 50.1.20.0/24 50.1.20.28 213957 - - 210970 - - root@www:/home/bob # Nothing obviously amiss....if I'm looking at the right data. > Especially the BULK column. If something is wrong on the USB level, it will > show up there. > On a whim, I checked a Pi4 running -current which also has a usb-sata boot drive. It reports UE_BULK_FAIL : 56 for the disk and performance seems good. An old Pi2 running -current with a relatively old USB2-sata boot disk reports zero bulk errors. No bulk errors seems desirable, but is some non-zero error rate normal? All three machines have been updated within a couple of days. Thanks for writing! bob prohaska