when i run dropwatch it doesn't show Enabling monitoring and anything.

[changliwei]

I run dropwatch in a increasing drops server, but when dropwatch not shows anything. Very bad luck, I’ve never succeeded when use this tools .please solve my puzzle .

`[root@localhost src]# ./dropwatch -l kas Initializing kallsyms db dropwatch> stats Getting statistics Software statistics: Tail dropped: 0 Hardware statistics: Tail dropped: 0 dropwatch> start dropwatch> << nothing updates here

[root@localhost ~]# ip -s -s link show dev enp125s0f0 2: enp125s0f0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 10:a4:da:cd:3b:27 brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped missed mcast
341502650 1433235 0 16390 0 72030 RX errors: length crc frame fifo overrun 0 0 0 0 0 TX: bytes packets errors dropped carrier collsns
7380034124 703307 0 0 0 0 TX errors: aborted fifo window heartbt transns 0 0 0 0 1 [root@localhost ~]# ip -s -s link show dev enp125s0f0 2: enp125s0f0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 10:a4:da:cd:3b:27 brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped missed mcast
341504231 1433250 0 16392 0 72038 RX errors: length crc frame fifo overrun 0 0 0 0 0 TX: bytes packets errors dropped carrier collsns
7380035718 703316 0 0 0 0 TX errors: aborted fifo window heartbt transns 0 0 0 0 1

and perf tools works [root@localhost ~]# perf record -e skb:kfree_skb
^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.608 MB perf.data (17 samples) ]

[root@localhost ~]# ip -s link show dev enp125s0f0 2: enp125s0f0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 10:a4:da:cd:3b:27 brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped missed mcast
341775464 1435679 0 16549 0 73801 TX: bytes packets errors dropped carrier collsns
7380065232 703555 0 0 0 0 [root@localhost ~]# perf script swapper 0 [011] 31202.866993: skb:kfree_skb: skbaddr=0xffff0020888e4200 rx_sk=(nil) protocol=4 location=__netif_receive_skb_core.constprop.0+0x690 reason: UNHANDLED_PROTO swapper 0 [011] 31203.692659: skb:kfree_skb: skbaddr=0xffff0020888e4400 rx_sk=(nil) protocol=39320 location=__netif_receive_skb_core.constprop.0+0x690 reason: UNHANDLED_PROTO swapper 0 [011] 31203.844043: skb:kfree_skb: skbaddr=0xffff0020888e4700 rx_sk=(nil) protocol=39320 location=__netif_receive_skb_core.constprop.0+0x690 reason: UNHANDLED_PROTO swapper 0 [011] 31204.846331: skb:kfree_skb: skbaddr=0xffff0020888e4900 rx_sk=(nil) protocol=4 location=__netif_receive_skb_core.constprop.0+0x690 reason: UNHANDLED_PROTO ....

[root@localhost ~]# uname -r 5.14.0-514.el9.aarch64 [root@localhost ~]# grep CONFIG_NET_DROP_MONITOR /boot/config-5.14.0-514.el9.aarch64 CONFIG_NET_DROP_MONITOR=y `

[nhorman]

Sorry to be so late on this, I completely missed it.

I'm honestly not sure whats going on here. The fact that you issue a start command and immediately get a dropwatch> prompt back suggests to me that something has gone wrong with the issuance of the start command over the netlink packet.

If you run the dropwatch command under strace, does it report any errors?

[nhorman]

what you're seeing in the output of the ip command are packets that were dropped in the Network cards hardware, i.e. before they've reached the kernel.

You need to enable hardware drop detection in dropwatch to see these. issue the command:

set hw true

in the dropwatch console to record these drops.