Thank-you for documenting this - the setup worked for me too!!

Hello! In software engineering, some seemingly simple requests users make can be deceptively difficult to implement, whilst others that users perceive as incredibly complex are actually quite simple to crank out. The engineer can be seen as either a goat or a hero. It is not fair to presume where this request falls in that continuum, but you are free to roll up your sleeves, dive into the open source code, and make that determination for yourself. John

If you're getting that far, then that's a good sign at least. To me it looks like it's using the wrong console port past that point. The BIOS/initial boot is going to the right port but after that it may be switching to the wrong port. That's almost certainly the case for the pfSense USB image unless you used the correct one named "ADI". Linux bootable drives would have the same issue, you have to edit the boot parameters and tell it to use the right port. In each case you'll want to direct them to use com2 / 0x2F8 / console=ttyS1,115200n8 or similar depending on the OS.

If i reboot manually. After restore it does not automatically reboot.

[image: 1591150669945-9a557bff-ab84-4e71-9cd0-dc1d1a412508-image.png] Doesn't look to be at the moment. .27.1 is the latest on 2.4.5

Hello, Thanks all for the help. It looks good now. Looks like as @jimp mentioned there might be problem with package server. Thanks to all

Seems multi threading not working properly on pppoe:( #PPPOE interface not using all queues: [2.4.5-RELEASE][root@router.kilsol.com]/root: sysctl -a | grep '.ix.1..*x_pack' dev.ix.1.queue3.rx_packets: 0 dev.ix.1.queue3.tx_packets: 5535356 dev.ix.1.queue2.rx_packets: 0 dev.ix.1.queue2.tx_packets: 6279749 dev.ix.1.queue1.rx_packets: 0 dev.ix.1.queue1.tx_packets: 6031232 dev.ix.1.queue0.rx_packets: 51346277 dev.ix.1.queue0.tx_packets: 9096401 #Non PPPOE interface is using all queues: [2.4.5-RELEASE][root@router.kilsol.com]/root: sysctl -a | grep '.ix.0..*x_pack' dev.ix.0.queue3.rx_packets: 5396172 dev.ix.0.queue3.tx_packets: 81242 dev.ix.0.queue2.rx_packets: 6706346 dev.ix.0.queue2.tx_packets: 618190 dev.ix.0.queue1.rx_packets: 6020980 dev.ix.0.queue1.tx_packets: 174660 dev.ix.0.queue0.rx_packets: 7843425 dev.ix.0.queue0.tx_packets: 48490796 [2.4.5-RELEASE][root@router.kilsol.com]/root: #Tunables do indeed match # of cores [2.4.5-RELEASE][root@router.kilsol.com]/root: sysctl net.isr.maxthreads net.isr.maxthreads: 4 [2.4.5-RELEASE][root@router.kilsol.com]/root: sysctl net.isr.numthreads net.isr.numthreads: 4 #net.isr.dispatch was set correctly [2.4.5-RELEASE][root@router.kilsol.com]/root: cat /boot/loader.conf.local net.isr.dispatch=deferred

https://www.netgate.com/products/appliances/ , tech specs tab says "16 GB DDR4 RDlMM (Expandable to 32 GB ECC)"

Hi, found your question almost year after it was posted. So this is just for the record. I had the opposite situation and it worked fine. I had SG-2220 which broke and I recovered its backup using a spare, SG-1000. It worked fine, I just had to accept the proposed changes to exclude a NIC that was NOT present at SG-1000. It was all presented during the restoration process. Everything else worked fine, so all aliases and rules were preserved. I named it "hardware downgrade".

Yes, and make sure to push the issue within the warranty period. My friends box rebooted about once every one/two weeks right from the beginning. Some log entries during uptime suggested the MMC was a bit flacky, but ususally the reboots was at night, so he didn’t really bother to try and fix it. Two days ago it just rebooted as usual, bu this time not being able to boot as the MMC storage was finally really dead. Unfortunately that was 3 weeks past the 1 years waranty, and Netgate just washed their hands and said sorry.... Exceptionally bad customer service in my opinion, so my confidence in them just went bottoms up. So make sure to have it fixed before your warranty expires.

@TadN said in Can SG-3100 handle 800MB down / 20 up with CodelQ active to remove buffer bloat?: I should not be seeing any buffer bloat even without pfsense? Bufferbloat happens at the funnel. You plan on testing direct with the modem so test that also. Bufferbloat happens. DSLR seems to have started this knowledge of the subject without really explaining it. Its actually logical that some bufferbloat will happen. If you dump 20 gallons of water into a five gallon funnel its going to take some time to get through. Fixing all the upstream issues will help but sooner or later you will see latency increase as your network tries to push and pull more data to the head end than your plan includes. But devices in your path can create more than should occur. Modem first. Why? Its the first line between you and them. Modems should be (IMHO) just a bridge between you and them. ( I hate all in ones) and that bridge should work full speed doing what you pay them for. Problem then becomes whether or not your node is oversold or not. But I do believe in most areas that is becoming less of a problem these days. If your modem is driven by a Puma6 chipset though.. I would not trust it. Neither does Intel at this point.

Glad it worked for you and the SG-5100 is online again. :-) -Rico

@ebcdic my two devices are actually wifi AP's specifically Synology WiFi AP's linked in a Mesh, the vendor says i should turn off IPv6 if i dont want to see the messages...it does appear to be a bug

@trombone Understood. VLANs are kinda stupid easy to learn and use. I'm kicking myself that I waited so long to try, and then eventually implement and use. The hardest part I found was keeping the terminology between the manufacturers of networking gear straight - one of them calls it trunks, the other one calls it tagged or untagged, another one calls it access, another one calls it something else, etc. Once you grasp those terms and how they're used, it makes good sense. Programming your managed switch is the other variable that was new to learn, but not too difficult. This is a good basic explanation of the concepts: https://www.computernetworkingnotes.com/ccna-study-guide/vlan-basic-concepts-explained-with-examples.html Good luck! Jeff

@Mark_07 Nice! Jeff

open a ticket at https://go.netgate.com and ask for support they usually respond hastily and they can help you out directly

@Tantamount Thank you for this. My SG-2440 couldn't update with the same error and your solution worked for me as well. I should add that after a reboot from UI, my device got a solid red status led. After power cycling it, it started working again with latest version!

To be clear; no you can't add another mSATA device. Only one slot supports SATA. Steve