You may have old ZFS BEs using disk space.

But you should install an m.2 SATA drive in the 5100 if you need to log a lot to it.
It's especially important on the 5100 because the eMMC failing due to drive writes can result it in not booting at all.

Steve

@stephenw10

thanks all working - except the haproxy which seems to be related to the update of the package. hoping someone can help. necroed a very old thread with my old config.

Thanks for your very fast responses.

@stephenw10 Ahh OK I thought that a port-channel between 3 interfaces on the pfsense 2100 and a Cisco switch might result in greater bandwidth. But if the internal PFSense switch can only pass one 1Gbps then I can remove the complexity and just use a single port. Thanks for your insight.

@darnoldvcs said in Netgate 1537 cannot achieve 10G throughput:

I guess that processor has more grunt that the Xeon D 1537.

They should be hard enough for the 10 GBit/s also Supermicro is
selling devices with such SOC´s and soldering 4 * 10 GBit/s port
on it.

I can achieve 10G with 5 threads on the 1537, and I appreciate I would need more interfaces to achieve the marketing numbers of 18+GB.

The 18+GB would in my eyes the plain routing power! Please don´t mismatch
with firewall rules on top or vpn traffic.

😁

@stephenw10 Yes, of course you're right. I was thinking my Calix ONT was the GP110X with only 2.5GBaseT, but it's the GP1101X w/10GbE.

@Brad303 In view of other discussions on this topic, I'm curious about how hot it runs.

Another user recently recommended one: https://forum.netgate.com/topic/187986/netgate-7100-working-10gbaset-sfp
That specifically claims to be low power.

I've never tested it myself though.

@stephenw10 we had the fiber optic installer here yesterday, and WE were running cables to the ONT I forgot to unplug the cable yes me and it pulled the firewall over and that vibration caused the pcie mini card to move a tiny bit when the firewall fell over, it was rebooting after with use of the pcie card, weird reseated it and it fixed it but just to be sure I want to add that other stand off. The firewall is fine it just got bumped is all but it was enough vibration to move the card some how. That card has had tons of errors on it for years. I am going to replace it also when I fix the standoff

Hmm, OK. Yes the installer image is the same for both versions; it can install multiple available versions.

However that's the wrong image, the 1100 is not an amd64 device it's aarch64(arm64). The recover image you should use is: netgate-installer-aarch64.img.gz

You should see:
Screenshot from 2024-05-02 21-02-26.png

I would certainly expect it to be. There is no complex custom code required for that, for example, which would need significant developer time to maintain.

At some point it will be marked EoL and from then it would be maintained as best effort. If some new bug or security issue was discovered specific to that platform it might potentially be excluded but AFAIK that's never happened to any previous devices. Hard to predict the future though!

Urgh that sucks! But good to find a cause.

Hey, I was experiencing the same issue with the Crucial CT16G4SFD824A 16 GB (DDR4, 2400 MT/s, PC4-19200, Dual Rank x8, SODIMM, 260-Pin) Memory. No matter how long I waited, it would never boot. This was weird because I even contacted Netgate and that's what they said they put in the upgraded models of the 7100. I did end up returning the Crucial stick and buying a Kingston KCP424SD8/16 OEM 16GB DDR4 2400MHz SODIMM Laptop RAM Memory Module and it ended up booting and working perfectly. I figured I'd add to the thread of ram modules that worked for the 7100. I am running 24.03-RELEASE with coreboot 01.00.00.20.

Yeah the drive should show there on one of the SATA ports.

Try a full power cycle if you have not already. If an SSD fails like that it will some re-appear after a power cycle. Of course if it does it should be replaced as soon as possible because it will fail again. That just proves it's a bad drive.

Steve

The forum routing issue was not related to the pkg repo outage. It wasn't down for anything close to 52 hours. It was more like 6.

Anyway it's back up now and our monitoring has been changed appropriately.

Apologies for the inconvenience.

Steve

The NDI is on the label, the serial number (also on the label) will also identify the unit for TAC.

And, yes, there is no Net Installer for armv7. Yet.

Hmm, it would only boot to that if it nothing else to boot from for some reason. Obviously it should boot from eMMC there.

If you power cycle it and enter the BIOS setup (hit DEL) do you see the emmc uefi boot option as the first device listed? Or listed at all in the boot override?
Screenshot from 2024-04-25 23-06-25.png

If it were just a DHCP issue (unlikely IMO) then, yes, you would be able connect to the VPN still.

I would try to replicate the state then test what's actually broken from the console.

Ok, done. See Todo #15441, under pfSense Docs.

Mmm, it can take far longer to boot than expected if you change the RAM size as it has to retrain it. But not longer than, say, 10mins so it sounds like that stick really isn't compatible.