Damn. My crystal ball is broken again.

aka from where are accessing the gui? how? directly via IP? via an URL? what kind of wan?

You get the picture.
–> http://forum.pfsense.org/index.php/topic,7001.0.html

Should be fixed in CVS now.  Thanks

maybe i should try change the WAN subnet first from /24 maybe to /30

i'll be report it soon after try this.

thanks hoba  ;) :)

I tested the changes. I didn't see any real performance difference in my testing. Response times remained within a small margin of error with the old settings and with the new on a 256MB ALIX firewall and a 1GB PIV 2.8GHz desktop firewall.

However, the new settings did reduce the amount of memory used by php. Not that big a deal on the big memory machines, but a significant improvement on lower end machines.

So I've tweaked the settings a bit more. Changes from the last version:

1. Don't disable fast_cgi on tiny boxes. I misunderstood the previous settings and having fast_cgi enabled really makes a big difference even on a tiny box.
2. More aggressive settings for max_children and max_requests for "larger" boxes.

system.inc.txt

Wonderful, thanks a lot!

Some of the guipages (like index.php or queues.php) are dynamically updating. This might cause some performance impact on lower end hardware. Not sure if you run the dashboard package but this will add some more overhead as well. Haven't seen that on non low end hardware though. On the other hand I agree that there might be some room for improvement by tweaking some webserver settings.

I created a ticket for this some time ago so we will revisit this before 1.2.1 gets released. In the meanwhile feel free to tweak the file by hand like shown here: http://cvstrac.pfsense.org/chngview?cn=16999

$opcode_cacher  .=  "apc.shm_size="8"\n";

The red number being the cache in MB. From our tests in the past you have to find the best setting as a too low number will cause the webgui to be pretty slow but even a too hogh number seems to have a negative affect. Feel free to post the best setting that you found out for your system. Note that you have to reboot the system after changing that file (you can do so easily from the webgui using diagnostics>edit file).

Let me rephrase: somebody has to be interested in doing it.

Bump the (+) icon at interfaces>assign.

Thats it. Thanks.  Shy little beggar hiding over there in the bottom right with all that empty space above and to the left!  :)

IIRC some Unix systems would parse an IP address such that an initial 0 meant the following number was octal (base 8) thus 031 (octal) was 25 (decimal) and so 201.031.161.136 meant 201.25.161.136 which is confusing until you discover whats going on. I'm not sure the convention was universally adopted, in which case there would be an ambiguity in 031.

No idea if 031 is rejected because of this possible ambiguity.

@mcrane:

If you are using firefox you can save the link to your pfsense box. Then go back to that favorite right click on it and choose properties. If you put pfsense in the keyword section and save it. Then you can go to your browser's URL bar and type pfsense and it will then replace the keyword with your link you saved in the favorites.

Another option way this can be done on more than one system is by using opendns with a free account. Once you are logged into opendns you can define keywords that will be replaced with links that you define.

Actually these are nice options that I never have thought about yet. Great ideas!  :)

perfect, thanks!

Thanks guys this has fixed it.

We haven't found any error with the code but good to hear that it works as it should.

@hoba:

This is the change: http://cvstrac.pfsense.org/chngview?cn=22213
You can manually change the file at your pfSense using diagnostics>edit file from the webgui.

I have applied changes and it is OK now.

TNX again.

Sasa

I can not make reliable 95th percentile graphs because that depends on the average over which it needs to be calculated. I can draw a line ofcourse for the 95th percentile but these would ofcourse differ per graph as most graphs have different average timeunits.
The 2 hour graph is 1 minute averages, the 16 hour graph is 5 minutes, the weeks graph is 30 minutes. etc.

Regarding your other point, no can do.
You might be interested in the bandwidthd package.

Which graph produced this error? I have not seen this error before.

Kind regards,

Seth

should be able to upload images now

I did now your test, seems that it was my fault - something wrong with my secur key:

/usr/local/sbin/lighttpd -f /var/etc/lighty-webConfigurator.conf
firewall:~#  /usr/local/sbin/lighttpd -f /var/etc/lighty-webConfigurator.conf
2008-03-28 16:25:26: (network.c.377) SSL: Private key does not match the certificate public key, reason: error:140A80B1:SSL routines:SSL_CTX_check_private_key:no certificate assigned /var/etc/cert.pem
firewall:~#

I generatet a new set key and certificate, now https workes with my certificate.