I generally use search on the menu bar then advanced search. I would limit the forums to Packages and Packages > Cache/Proxy if you are looking for info about squid.

Amen. Keep it simple then add technologies one at a time if you want to experiment. But then when you are posting on a forum you are not asking someone for help with an LACP problem to solve what is really a basic issue. People get too wrapped up in configuring too many things at ones. Eg. If you are going to do 20 VLANs, do one and get it working right first.

You'll see messages like these… Alarm messages mean it's going into an alarm state, clear is coming out of one. Dec 6 10:01:59 dpinger WANGW 198.51.100.1: Alarm latency 710us stddev 620us loss 22% Dec 6 10:04:41 dpinger WANGW 198.51.100.1: Clear latency 765us stddev 741us loss 5% Dec 6 10:09:54 dpinger WANGW 198.51.100.1: Alarm latency 605us stddev 633us loss 22% Dec 6 10:12:17 dpinger WANGW 198.51.100.1: Clear latency 747us stddev 648us loss 5% Dec 9 10:49:20 dpinger WAN_PPP2_PPPOE 10.6.220.1: Alarm latency 1046us stddev 2398us loss 22%

I could be wrong but I don't think that Squid can cache something that it cannot read because of the encryption.

…..would be a closer use case to users surfing the web. This could also be a nice test fir sure and on top of this it might be also that he is able to set up that iPerf test with the -p "n"command to high up or scaling up the entire workload. As an example: iPerf Client --------------------------------------> iPerf Server IP 10.1.1.2                                          IP 10.1.1.1 #iperf -c 10.1.1.1 -P 8 so over that test it will then be used 8 streams and that could be also really filling up or saturate that line.

Also - I did find this info on the forum from others having similar issues installing pfblockerNG / php memory issues but I don't know if I need to do these edits too: for amd64 systems … change /etc/inc/config.inc (line 75) from 512M to 640M or 768M. Code: [Select] // Set memory limit to 512M on amd64. if ($ARCH == "amd64") {   ini_set("memory_limit", "512M"); } else {   ini_set("memory_limit", "128M"); } to Code: [Select] // Set memory limit to 512M on amd64. if ($ARCH == "amd64") {   ini_set("memory_limit", "768M"); # amd64 Default 512M } else {   ini_set("memory_limit", "128M"); } AND modify the /usr/local/etc/php.ini (last line) and /etc/rc.php_ini_setup files (line 303) from Code: [Select] suhosin.memory_limit = 536870912 to Code: [Select] suhosin.memory_limit = 671088640 # 640M Default 536870912 or Code: [Select] suhosin.memory_limit = 805306368 # 768M Default 536870912

I've been having the same issues.  Did you ever resolve it?

@johnpoz: So did you try turning off the dmz host mode and just forwarding specific ports, like 80 or 443, 22 and test the specific ports.. Does it source nat as well?  If so that is just MESSED UP!! Do you have any other choices for isp?  If so freaking tell them your moving to different ISP and see if that gets their attention. I did try just forwarding specific ports, but it was the same. No difference. I can't figure it out. I've never had this problem before with any previous ISP. Once explanation is that our house was built this year and it's all new infrastructure. So maybe they are still smoothing out the bugs. I'm hoping that's the case anyway. Unfortunately ISPs usually have a monopoly in specific neighborhoods. So where i live they are the only option. The only way to change ISP is to move house.

Ok, will remove it on all machines then. Sounds clear! I had the idea from the name itself: maximum transmission unit

At the moment there isn't a way to add a user to a reserved system group in a way that will be preserved. You could script a groupmod command into a shellcmd so it would happen at bootup /usr/sbin/pw groupmod proxy -m <username>That might get you closer, though it could still be clobbered if the groups get rewritten later</username>

I have got the routing setup by my ISP now. Works nicely.  :) . Thanks again for your quick reply.

It would be nice to be able to experiment with some of this stuff and right now I'm not able to. I'm not so much trying to fix a problem. Like what?  vlans - you have a managed switch..  Just not getting why you think you need a 2nd router to do vlans?  And not sure what you think having your nas in multiple networks gets you?  If you want to have your devices access files off this nas with using a different interface ok.  But then really your devices need an interface in this other network, etc. If you want to setup a management vlan and have your nas have its normal network that it serves up files on, and then a management network you access it from to admin it, ok.  But then your devices you will manage from really need to be in this managment vlan or again its pointless.

I'd try to see what a 4GHZ+ chip would do.  Maybe even overlocked with some cooling. Perhaps it would be nice to set it up in a VM, then it might be nit really bounded to the single CPU Core or am I wrong now?

Load balancing will be able to realize over two or more WAN ports with the following methods; Service based routing and/or Session based routing, this both will be the most common ways to balance traffic from server.