NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Gateway server experiencing degraded performance

On Mon, 16 May 2022, Andrew K Adams wrote:
I've looked at traffic with tcpdump, but there is so much in both
scenarios that I can't really say that one service, or a
packet-of-death is triggering the degraded state.  But thinking in
aggregate, if there was a way to monitor the average packets/sec the
gateway is forwarding, in both scenarios, I'd at least know for
certain that it is or is not my gateway (I guess it's conceivable that
by rebooting I'm tweaking state in Comcast's next-hop during DHCP
negotiation or something ...)

net/vnstat in pkgsrc might be useful for this.

Hi, I’ve noticed a problem with my NetBSD server that I really could
use some help with.  The server is acting as my home router (gateway)
in ‘pass-through’ mode with Comcast Xfinity as the service provider.
The only functions/services enabled on the server are: ip-forwarding,
dhcp-client, and NTP.   And on the internal-facing lan: NAT, DHCP, and
SSH.  The internal network supports 5+ wired hosts & 10+ wireless
devices.  I currently have the 1Gig option with Comcast and when
things are operating as expected, I see from an internal (wired) host
750mbs of throughput according to Comcast’s Xfinity speed test
website.  The problem is that performance doesn’t last.  In anywhere
from 3 weeks to 24 hours, my network performance (again, using the
same wired host measured by Xfinity’s speed test) will drop to ~20mbs.
And the change is clearly observable in all networking applications,
e.g., Netflix.  I’m querying here, though, because the fix I’ve found
is to reboot the server.  Upon reboot, I immediately go back to
750mbs.  Hence, I’d like to figure out how to troubleshoot what on my
server is (eventually) degrading my network performance.

The server is relatively new; I built it using an ASUS Mini ITX with
two onboard GigE ethernet ports.  Here’s some relevant dmesg output:

[     1.000000] NetBSD 9.1 (GENERIC) #0: Sun Oct 18 19:24:30 UTC 2020
[     1.000000]
[     1.000000] total memory = 3457 MB
[     1.000000] avail memory = 3380 MB

[     1.025286] wm0 at pci0 dev 25 function 0: I217 LM Ethernet
Connection (rev. 0x05)
[     1.025286] wm0: interrupting at msi2 vec 0
[     1.025286] wm0: PCI-Express bus
[     1.025286] wm0: 2048 words FLASH, version 0.13.4
[     1.025286] wm0: 0x6a4480<FLASH,IOH_VALID,PCIE,ASF_FIRM,AMT,WOL,EEE>

[     1.025286] re0 at pci3 dev 0 function 0: RealTek 8168/8111 PCIe
Gigabit Ethernet (rev. 0x0c)
[     1.025286] re0: interrupting at msix4 vec 0
[     1.025286] re0: using 256 tx descriptors

And from ifconfig (note, I’ve removed my IP & MAC address from the output):

iquitos# ifconfig -a
        media: Ethernet autoselect (1000baseT
        status: active
        inet X.X.X.X/20 broadcast flags 0x0
        Inet6 X%wm0/64 flags 0x0 scopeid 0x1
        media: Ethernet autoselect (1000baseT full-duplex)
        status: active
        inet broadcast flags 0x0
        inet6 fe80::728b:cdff:febc:831b%re0/64 flags 0x0 scopeid 0x2
lo0: flags=0x8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 33176
        inet flags 0x0
        inet6 ::1/128 flags 0x20<NODAD>
        inet6 fe80::1%lo0/64 flags 0x0 scopeid 0x3

I’m not sure what other information would be useful, just let me know,
and thank you in advance for help you can give me!

Just a question, not a requirement, but have you looked at Wireshark
traces comparing good performance vs. bad performance?

It's just another way to find possibly find a problem. I'm sure
someone else will have better ideas.


Home | Main Index | Thread Index | Old Index