You're correct in that this should work just fine. And a quick test on my webserver shows syncookies working correctly (bge driver 10-current r228438 with lighttpd).
There have been some, seemingly minor changes between 9.0-release and 9-stable in netinet but a quick look doesn't show anything...
Some, perhaps obvious, questions:
EDIT: the format of your sysctl.conf (at least as posted above) is wrong it should be:
name=value
e.g.:
net.inet.tcp.syncookies_only=1
net.inet.tcp.syncookies=1
Does your synflood tool support syncookies? Which tool are you using?
Did you reboot...
equivalence would be
$ route add -inet6 -net 3ffe:501:ffff:222::0/64 ::1 -iface em0
$ ifconfig lo0 inet6 3ffe:501:ffff:222::1/64
However you are limited to 255 addresses per interface.
You may be able to get around this limitation via interface cloning and having 255 addresses per clone.
Sounds like you just need to route the prefix in a certain way instead of assigning the entire network to an interface, unless you have 500,000 accelerators?
Please be specific in how you did this in linux.
My buest guess right now is that apears to be threading issue. Although libkvm is fully re-entrant, there is probably a subtle problem in the way or order things are being called.
I did notice that some serious cleanup has gone into the git repository in regards to this file and that way the...
I took a brief look at the code and did some debugging and didn't look like conky's freebsd.c freeing anything of the kvm's memory space. Usually dies on the 2nd or 3rd run through proc_find_top(). I'm going to take a closer look at stack vs. kernel addresses for all the vars.
Program...
Well, as you know, HTTP/1.0 without keep-alives is going to create a new connection for each element on the page. For a single html page with 9 images, you would only create 10 connections at most.
Something else is affecting your test to create 40 connections. Perhaps you are counting...
If paean's solution does not work this can be tricky to configure correctly, but the first step is to do a verbose boot so the that the driver prints out it's default pin configuration.
nextboot -o "-v" -k kernel
reboot
And after boot, you can look at the output via dmesg.
As long as...
The ipv6 addresses are aliases (see the ifconfig manual page, ifconfig(8)), by changing the network address and re-running the rc.d script, you are simply adding an address alias.
You'll need to remove the address alias to get rid of it:
ifconfig [interface] inet6 [address] -alias
My phone connects as a ugen device normally when connected (unless I select mass storage mode on the phone), but if you kldload umodem it appears as a umodem device. From there you can use ppp to connect to the network of your provider using the tty device presented by umodem / ucom.
Sounds like a memory issue. Are you building amd64?
Do you see a warning in the bootup about your AMD processor?
A long shot, but can you try rebooting and doing a buildworld with
vm.pmap.pg_ps_enabled=0
in /boot/loader.conf
This is the only way I can get my HPDL385G5 to work.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.