@robbiett said in Netgate 6100 - Significant Interface Interrupt Rates:

Are you trying to help me, as it has become a little unclear?

less and less when i read things like above.

anyhow, what i tried to convey is that the # of interrupts seems to be in line with what i find "normal" on my sites with a NG6100 device:

300mbit wan bandwidth + unknown amount of inter-vlan traffic:
1e2b7781-b3c2-4d2c-b226-2400cc3a267c-image.png

1.2gbit wan bandwidth + unknown amount of inter-vlan traffic:
1baee5c6-fa34-49f0-aa63-bb93a43a0cca-image.png

so, for me 70k interrupts does not cause any issues. it could be different for you, but it could also be that you have different issues that are not related to interrupts at all.

@furom said in [Solved] Please help with switch/vlan (802.1q) setup on Netgate 2100:

it is blazing fast

Nice. 😁

@stephenw10 Thanks, I've updated my support ticket. I don't think I'm still under warranty, but maybe tac can give me advice on whether anything can be done.

@stephenw10 Thank you very much.. This will help me get things going.
Have a great day!

learning more: I see node_exporter for armv7:
https://www.freshports.org/sysutils/node_exporter/

Does anyone have experience with this? How to get it installed, would it work on the SG3100?

VLAN1 untagged on every port would be fine. What's definietly not fine is having more than one VLAN untagged on any port such as VLAN20 is there.

@stephenw10 I was just in the process of filling out the form on the website when I got the email you had replied to my post. I'll have the ticket submitted shortly.

Yeah I think trying to diagnose this with the other other issue still present is going to be very difficult. There's s good chance they are related and it might already be fixed in 23.05.

The service watchdog is a trouble shooting tool and should be seen as such.

There are some situations where you might want to enable it on a service in a more permanent way but even then it's usually to address some underlying bug.
For example if you have an OpenVPN client and want it to be always up you might use the watchdog. But only because if the server side rejects the connection as unauthorized the client will exit and not retry. Most clients would never see that or if they did retrying would not help.

Steve

@nevolex The 1100 has not even be slated as "end of sale" so it's successor is still the 2100, whose successor is the 4100, whose successor is the 6100, whose successor is the 8200, whose successor is the 1537, whose successor is the 1541.

Also when there's an announcement to be made look for it to appear in Netgate Announcements

Yeah, you should be fine. Just reinstall 23.01 clean when you can.

Thanks @stephenw10 !

@captainron My previous offer still stands.

@stephenw10 said in IMIX benchmarks:

There has been no significant change there. I haven't seen any measurable change in local testing.

Thanks Stephen. Have a nice weekend!

@stephenw10 said in SG 1100 unable to check for updates:

@gdarends said in SG 1100 unable to check for updates:

Segmentation fault (core dumped)

The first thing to try here is a full power cycle to reset the crypto chip. That means actually pulling the power plug, a reboot does not reset it.

That was it! Thanks so much.
Gonna start prepping to do the upgrade now.

You might be able to get a new-old-stock SG5100 which works great for 1gb PPPoE, but with 1gb ports instead of the 2.5/10 on the 6100…

@iversa None of the Netgate devices have Wireless built in and there is a wireless function in pfSense but it is not well supported by FreeBSD and you should look at any other dedicated Wireless AP option instead.

As a company we do not have a specific recommendation of any particular product.

I would expect it to install and run fine but you would be missing some of the hardware support. So at least the LEDs and reset buttons would not function. That would also mean the firmware updater for BlinkBoot would not be available should there be an update.
It's not something I've tried though because it's not supported. I can do if it's something you need.

Steve

@stephenw10 Thank you. I'll message you my NDI shortly

It does sound like the eMMC has failed. There are some of us on 23.01 who have seen 1100 storage timeouts but they behave a bit differently than that (and I haven't seen them at all on 23.05).

While it is possible in theory to run from USB, in practice it's not that viable. Running live off a thumb drive will kill the thumb drive fast, they are not made to take constant writes. Last time I tried this it didn't even make it a month. Maybe an actual SSD in a USB enclosure might have a chance but it's debatable, given the 1100's pickiness with what it will/won't boot.

If you can find one that will boot from, You could go into the Marvell>> prompt and change the default bootcmd. Out of the box, it's set to bootcmd=run emmcboot to boot from eMMC. You'd need to run setenv bootcmd 'run usbboot' and then saveenv to make it boot from USB each time. But again, it's anyone's guess whether or not that would boot reliably.

To get that far you'd also need to manage to get the live/installed image onto a USB drive, and not the recovery installer.