the issue is as I warned for and is now an exploit
-
Re: Ubuntu 18.04 on XG-7100 platform
First the 7100 device ships with no seal. So then I direct open the device to check physicall.
So what answered marvell on the switch? As the tech design says clear a 802.1Q VLAN switch on port levell. With a quad core intell , which is to use the rss offload on cpu by this netgate used 2 intell 10 gig direct on board ,so no rj45,just on the centos boot the 2 10 gig becommes 4 links with full duplex so the 5 gig is diveded on 2.5 gig, so this is lagg0
Now a lag is a Link Aggregation Group, So vlan WAN was 4000 and LAN 4001
Now those vlans are created after lagg0, so you have a link group that becommes lagg0.1, lagg0.2
remember the appliance says 8 ports , so thats wrong, due those ports are not even bound on seperate pci lanes, so it's 1 WAN in vlan lagg0.1 tag 4000 on software , with 7 lan ports pointing on lagg0.2 vlan 4001
Now what happens, the BSD kernelll makes the 2 10 gig links in 4 ix 5 gig that uses a P dye on board to try ssegment the WAN out the LAN.
so lagg0 is parent of all 4 links, but gets vlan tagged as those 4 links are designed against the quad core intell cpu. But the CPU is not operating in a virtuall gen1
So what happened? lagg0 is parent so the 2.5 gig up and downstream bounced on the CPU which is a SoC, As I filed BSD considering the design and got every fabric stage contacted. marvell had just a order to soldied the frame, no more. and what is the result? The device got attacked and is the uart in memory corruption, so the BIOS, GONE.
The device boots pfSense just as a linux BSD , the pfSense has no drivers, no device controller, that's BSD Kernell task that boots in a main system Centos.
Vlans in linux are just dev that maps on the ethernet NIC. In a quad port VLAN system, we have always a BASE MAC and the NICS are functions, but every function has on vlan id the burned in MAC, so traffic on link off ISP may never run in layer 2, as that's etherchannel, layer3 is the MAC gone and runs IP the transmission, and NO tcp/ip in a link from outside the LAN runs NEVER Vlan, as the ingress is covered by PVID, but egress is mtu 1500
and no IEEE shall a pakket in 802.1Q vlan placing, as the ISP never sees those 4 bytes extra,
Now the appliance is a ask if netgate, rubicon can deliver a intell xeon , with 2 10gig fibre , duall path and short range. So the diode must send with a return path , as that can run the disk on same way so the controller can operate in a SAS backend and the second fibre runs the SATA backplane. The only cruciall is then a broadcomm quad port vlan 802.1Q PCI , and netgate has a Server product for enterprise levell.
As the esxii boots in a SD micro chip which we can lock on flipping the card so the boot always runs on ROM. , and just a usb flash disk is used for the config changes.
and to inform the case , I have 35 years experience so I needed just disconect the pfSense disk, which the boot stopped in BSD. And dmesg is enough to see the bootstrap, a firewall is not a router , pfSense is a patented system by rubicon, a site project off Google LLC. The kernell off BSD update gave all problems, as the distro forked drivers in ip6,WLAN (WiFi) and made also devices in obsolete.
What Intell Atom uses to run mods in a ARM7 mode, and those mods are Amiga files, so spectre meltdown was activated and the C ,P ACPI states were used to offload data in ip6 on a x86 CPU, As x64 only works with Intel Vt-d. and the memory in 3.5 GIG reserved for the base controller. Those idiots that uses a hp thinstation are unaware that those links are 100mbs half duplex. if that get's a DOCSIS 3.0 as "bridge", then the cable is cat5e while pfSense runs in a cat3 device. Should I have made a wrong conclusion then ask blackbox for confirmation. And should I get the microcode for the xg 7100, so the BIOS is corrupted, so i have no way to connect a telnet as viewport, but if I could repair that device it can run as a extender in a link with NAS , as SAS is not possible in that system, but can run as a PXE server, so my devices in compute power can then leave my development room
as all we need is a KVM but as machine., And fyi, I have Google LLC as full controll on all in outside. So to clarify , this is not Google for public as I also signed all data stored in USA and by that I refer on the pattents violations against EU. So all I do is for non commerciall, as I closed all roadmaps, and have no physicall links to outside , but once coder always coder : - )) -
Tried to read that twice now and I'm still not sure what you are saying here. You are running Ubuntu? CentOS?
You think you've found some exploit?
Steve
-
@stephenw10 You are not alone. It seems that there are many different topics and no clear question. Much of it is gibberish.
-
@AndyRH said in the issue is as I warned for and is now an exploit:
Much of it is gibberish.
Its all gibberish if you ask me..
Its like they went to https://loremipsum.io/
There are not more than a few words that string together that make any sense..
His other post from 2020 is also pure gibberish nonsense..
If I had to guess, he has some sort of Aphasia, maybe drug induced?
-
Could this be AI generated or something?
Can you state the name of the exploit you found and CVE number?
-
@JonathanLee said in the issue is as I warned for and is now an exploit:
Could this be AI generated or something?
Artificial Ignorance?
-
@provels Is this the extent of AI capability?
Ted
-
@johnpoz said in the issue is as I warned for and is now an exploit:
If I had to guess, he has some sort of Aphasia, maybe drug induced?
hahaha.
