The 6100 PSU has a higher current rating (5A vs 3.3A) so it's impossible to recommend doing that. The supplied PSU is capable of delivering the maximum power the 6100 can draw which of course it usually does not however. Steve

https://docs.netgate.com/pfsense/en/latest/solutions/netgate-6100/wall-mount.html

@bmeeks Thanks, ticket created. Greg

@stephenw10 said in Netgate 6100 10 gb Routing Throttled to 1 gb?: @bpsantucci said in Netgate 6100 10 gb Routing Throttled to 1 gb?: I'm not sure what you mean by routing or NATing in this case... I mean if you are testing with a client on a 'LAN' and the server on a 'WAN' the auto outbound NAT rules will be NATing that out of the WAN interface. That requires additional processing limiting throughput. pfSense determines what is a WAN interface in that situation by whether or not it has a gateway defined on it. The throughput is limited by the CPU in the 6100 in that setup. As long as you have pf enabled and firewall rules in place ~5G is what I would expect to see in iperf3. Steve Got it...Never thought about it like that, but now I see how WAN/LAN NATing could create significant processing overhead. Thanks for the explanation!

Ordered mine on 11/24, just got an email this morning saying they will begin shipping the week of the 29th in the order received. Hopefully they got enough in to get mine shipped.... :)

@rb106098, your vent is founded IME, I had the exact same issues around the exact same times, both last year and this year. There was a post on reddit around the time of my first issue issue where the [user] (https://www.reddit.com/r/PFSENSE/comments/kv704j/never_buy_the_sg1100/) had said support confirmed there was a known issue (pretty sure I seen other post saying the same thing). In looking at reddit post today, I seen where a support person said they provided support even to EOL devices, with the statement that they provide "pride in our hardware", a joke, as they don't last 12 months. I am going to assume they take this thread/post down soon, but I am considering proposing a class-action suit, so please watch reddit in the coming days.

Ok, but importantly it was on lagg0.4091? The LAN is assigned as lagg0.4091? If you captured them on lagg0 for example they may have been incorrectly tagged. If you are seeing the DHCP requests on LAN however the DHCP server should respond unless the requests are somehow invalid. I would expect to see errors in the dhcp log though if that were the case. Do you see a state opened on the LAN for those requests? Perhaps they are being blocked by the firewall somehow? Steve

If whatever you ran is still running you can kill the process manually at the command line. More likely though it just had a very large output that caused PHP to fall over and has now finished. Steve

Mmm, yeah I would expect it to have logged battery low if it did reach that. Maybe it restarted for some other reason? I think you will need to do a test and see what it actually logs and what the UPS stats are reported as when it's discharging. I would probably run the 3100 directly and not through the UPS during that test so it doesn't actually lose power if/when the batteries are drained. Steve

@b_hayward I just updated a 3100 (also arm) from 2.4.4p3 to 21.05.2, yeah the built in update was being a bit flaky.. Got the image from netgate, did a clean install and restored my config from backup.. bing bang zoom, bob's your uncle!

Yes but there are some caveats, Because the 3100 shares the same NIC(s) as the 1100 the VLAN interfaces defined on the 1100 would simply be created on the 3100 and that's not how you want it configured there. We can help you convert that config to allow it to import directly if you open a ticket: https://go.netgate.com/ Steve

The system was marked clean, but booting from it isn't possible. @metalgravy said in SG1100 repeatedly auto-reboots: Is there anything in particular that I should be looking for? It's video time. And put a post it on the box ;: don"'t remove the power from this box - you the GUI shut down command" ;) Add an UPS.

@stephenw10 said in flashing blue lights after update failed: Yes, if you just got the device and don't know it's history or have a complex config on it already then I would just reflash it with 21.05.2 clean. Open a ticket with us to get the recovery image: https://go.netgate.com/ Steve i just opened the ticket. thanks for the help. cheers

Just as an update for whoever might run into this, we haven't experienced it since the "available version" zabbix items have been disabled and rebooting the gateways afterwards. That has been over 2 months ago. You can check more info in my previous comment. There is still the open question on why do the processes get stuck during this check, but since this isn't that important, I'm going to leave it like this.

@stephenw10 OK, I'll give it a try tonight when I'm done with my work, in case I manage to screw the whole thing up... :) I'll post the results (I have an old Linksys router as a backup...) Thank you again for the help. -N

Nice news, the SG-3100 is a very good home device. And yes, the 22.01 Dev runs very solid and nice on it. Looks like a very nice Release Version is incoming...

@1amt0ny can confirm the WD SN520 NVMe SSD - 256 GB works perfectly fine in my box. Set it up with ZFS as explained in https://forum.netgate.com/topic/168038/proper-steps-for-zfs-pave-re-install-on-6100

@jonathanlee I see subnetting like creating cubicles. As long as you have them your good, however yes if I just had LAN everything that is bad. Thanks for pointing that out. No Any Any rules.

Ah, OK. That's on the 5100 so not some issue specific to the 6100 eMMC. Certainly it's possible it may fail at some point, as any component might. There is no known problem with the 5100 eMMC either though. The write life of the eMMC, which is what usually concerns people, should be easily sufficient for the expected life if the device in normal use. That will be reduced if the device ends up swapping continually or you have installed packages that create a lot a logs. It can also be extended by installing without SWAP and enabling RAM disks but that is not necessary with eMMC. I only do that on the few test devices I have still running from CF or if I ever run anything from USB directly. Both of which have far far more limited write life. However, as you say, if for some reason the eMMC does fail it's possible to install an NVMe drive. Steve