Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    [solved] - Traffic Graphs: Cannot get data about interface [2.0.2]

    Scheduled Pinned Locked Moved Problems Installing or Upgrading pfSense Software
    66 Posts 13 Posters 30.3k Views
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • S
      Slam
      last edited by

      Thanks for the information Jim, I would really appreciate the new images as one of my location consists of an alix and heavily relies on captive portal and its new fixes , it would be very much appreciated when you have the time to upload.

      @jimp:

      gitsync does work on NanoBSD, but it will be rather slow and involve a lot of writes to the CF. So it'll be a pretty bad idea, unless you don't love your CF, in which case it's fine.

      I've got a new set of images built, I'll upload them in a bit and post the url when they're done.

      Sorry I forgot to mention the initially errors, its been over a year since I ran it.

      I have been watching the traffic graphs for the last 6 hours and for me its fixed and no errors being reported about fastcgi crashing.

      @tim.mcmanus:

      Thanks, Abdsalem!

      I ran the command.  Got a few errors about git not being installed and having some older dependencies installed, but the process of updating the pfSense stuff ran without issue.

      I'll open up the web portal now and report back.  The UI usually fails within 30 minutes, I'm hoping it works out this time.

      Thanks again for the suggestion!

      1 Reply Last reply Reply Quote 0
      • T
        tim.mcmanus
        last edited by

        I'm happy to report that after 5 hours running, I haven't had any issue with the graphs.  No fastcgi errors in the /var/log/lighttpd.error.log either.

        pfSense version listed is:  2.0.3-PRERELEASE (amd64) built on Fri Dec 7 22:39:16 EST 2012 FreeBSD 8.1-RELEASE-p13

        I think we have a winner!

        Thanks again to Abdsalem for the gitsync suggestion and commands, and thanks to the mods for taking the time to help with this.  Not sure what the root cause was, but the gitsync seems to have fixed it and not caused any new errors.  I ran it at 9AM today in case you need to refer to the code base that included the fix.

        1 Reply Last reply Reply Quote 0
        • jimpJ
          jimp Rebel Alliance Developer Netgate
          last edited by

          If you need a "real" image instead of gitsync, I've temporarily re-enabled 2.0.x snapshots.

          http://snapshots.pfsense.org/FreeBSD_RELENG_8_1/i386/pfSense_RELENG_2_0/?C=M;O=D
          http://snapshots.pfsense.org/FreeBSD_RELENG_8_1/amd64/pfSense_RELENG_2_0/?C=M;O=D

          Remember: Upvote with the 👍 button for any user/post you find to be helpful, informative, or deserving of recognition!

          Need help fast? Netgate Global Support!

          Do not Chat/PM for help!

          1 Reply Last reply Reply Quote 0
          • S
            Slam
            last edited by

            @tim.mcmanus:

            No fastcgi errors in the /var/log/lighttpd.error.log either.

            Just a heads up Tim, lighty outputs its errors to syslog now so you can view it in the webgui under https://pfsense/diag_logs.php ….. /var/log/lightyd.error.log will contain errors from prior the gitsync update we applied, so any information lightyd-error.log is now irrelevant
            More info here

            http://redmine.pfsense.org/projects/pfsense/repository/revisions/6f0d237e757ab25a8d333af4caf42a7d77239e5b

            I am coming across a small warning but bear in mind I havent pulled the latest using gitsync so this warning may go away once its quiet enough to do

            Jan 3 19:53:09 lighttpd[25142]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT - configured request variable name length limit exceeded - dropped variable 'amp;{google:RLZ}{google:acceptedSuggestion}{google:originalQueryForSuggestion}{google:assistedQueryStats}{google:searchFieldtrialParameter}sourceid' (attacker '10.0.0.218', file '/usr/local/captiveportal/index.php')

            1 Reply Last reply Reply Quote 0
            • T
              tim.mcmanus
              last edited by

              Thanks, I went to look at the system.log and it's clean of any fastcgi errors.  It didn't show the output from the gitsync earlier today, but I'll assume that stuff isn't logged to save space (or because I have the logger set at it's default and not in dev mode to robustly log everything).  I prefer to read the logs via CLI because I like the breeze from my propellor hat.  :)

              Safari crashed so I didn't make it past 5h worth of logging.  I restarted it and will let it run all day.  I assume at this point it's fixed.

              1 Reply Last reply Reply Quote 0
              • J
                jikjik101
                last edited by

                After gitsync, my dashboard hangs. Need to reset webconfigurator.
                Been running the browser for 30 minutes and no 500 error yet but many errors in system logs.

                
Jan 4 08:22:23	lighttpd[40989]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT - configured request variable name length limit exceeded - dropped variable 'yui:3_5_1/widget-position-constrain/widget-position-constrain-min_js' (attacker '111.111.111.111', file '/usr/local/www/sgerror.php')
Jan 4 08:22:23	lighttpd[40989]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT - configured request variable name length limit exceeded - dropped variable 'yui:3_5_1/widget-position-constrain/widget-position-constrain-min_js' (attacker '111.111.111.111', file '/usr/local/www/sgerror.php')
Jan 4 08:11:37	lighttpd[40989]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT - configured request variable name length limit exceeded - dropped variable 'yui:3_5_1/build/querystring-stringify-simple/querystring-stringify-simple-min_js' (attacker '111.111.111.111', file '/usr/local/www/sgerror.php')
Jan 4 08:11:37	lighttpd[40989]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT - configured request variable name length limit exceeded - dropped variable 'yui:3_5_1/build/querystring-stringify-simple/querystring-stringify-simple-min_js' (attacker '111.111.111.111', file '/usr/local/www/sgerror.php')
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 1 load: 391
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 1 load: 391
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-0
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-0
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 0 load: 391
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 0 load: 391
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-1
Jan 4 08:06:40	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-1
Jan 4 08:06:31	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:31	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:31	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:31	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 1 load: 391
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 1 load: 391
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-0
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-0
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 0 load: 391
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.3002) backend died; we'll disable it for 1 seconds and send the request to another backend instead: reconnects: 0 load: 391
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-1
Jan 4 08:06:28	lighttpd[34122]: (mod_fastcgi.c.1732) connect failed: Connection refused on unix:/tmp/php-fastcgi.socket-1
Jan 4 08:06:26	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:26	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:26	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
Jan 4 08:06:26	lighttpd[34122]: (mod_fastcgi.c.2760) fcgi-server re-enabled: 0 /tmp/php-fastcgi.socket
                1 Reply Last reply Reply Quote 0
                • S
                  Slam
                  last edited by

                  I had a little freeze a moment ago when trying to browse to 'Status/Traffic Graph', a reboot fixed that as I hadnt rebooted all day since I GitSynced, I returned to 'Status/Traffic Graph' to make sure it wasnt the cause. Reboot if you havent done so.

                  If youre referring to

                  FastCGI-stderr: ALERT

                  I am also seeing this, the cause of mine is the portal page I have uploaded that has a link to google.com, I notice in your log that the cause of the alert is dashboard widgets, both show that the variable name length is set too short and causing that alert.

                  When it happens to me, I am guessing my captive portal user clicks the www.google.com link and gets redirected to the captive portal login page and because cp appends additional info to the url, that causes the error, I'm no coder but could a dev confirm if this might be the case?

                  1 Reply Last reply Reply Quote 0
                  • J
                    jikjik101
                    last edited by

                    My system log is now full of```
                    lighttpd[35079]: (mod_fastcgi.c.2676) FastCGI-stderr: ALERT -

                    
3 hours running and my traffic graphs are okay for the time being.
Except for that above error.
I have no redirect page or captive portal.
My packages: squid, squidguard, lightsquid, vnstat2, imspector, mailreport and sarg.
Although i have traffic limiter for my network, 2 WANs and 2 LANs.
                    1 Reply Last reply Reply Quote 0
                    • S
                      Slam
                      last edited by

                      Like I said earlier, in your case it seems to be a dashboard widget, what widgets do you have enabled? also I noticed in the original log you posted the attacker IP is '111.111.111.111' I am guessing this is the client you use to administer pfsense.

                      1 Reply Last reply Reply Quote 0
                      • T
                        tim.mcmanus
                        last edited by

                        A lot of whiskey, beer, and hours later and my graphs are plugging away swimmingly.

                        IMHO, this is resolved for my installation.

                        Many thanks again, and I raise a glass!

                        1 Reply Last reply Reply Quote 0
                        • J
                          jikjik101
                          last edited by

                          @Abdsalem:

                          Like I said earlier, in your case it seems to be a dashboard widget, what widgets do you have enabled? also I noticed in the original log you posted the attacker IP is '111.111.111.111' I am guessing this is the client you use to administer pfsense.

                          My widgets are: traffic graphs, gateways, interfaces, IPSec and System Information.
                          The 111.111.111.111 is a bogus IP I edited in replacement to my WAN IP. The log shows my WAN IP but I don't like to post it so I made it to 111.111.111.111

                          1 Reply Last reply Reply Quote 0
                          • S
                            Slam
                            last edited by

                            I have a feeling if you remove the IPsec widget the error will disappear….

                            The original bug posted in this thread is also fixed for me and the bug we both experience I feel is something seperate, I think it needs its own thread.

                            1 Reply Last reply Reply Quote 0
                            • J
                              jikjik101
                              last edited by

                              you are correct, the IPSec and the System Information widgets created the whole errors.

                              I think the Uptime bar on the system information is the culprit, because I just installed a 2.0.2 on a new box and on IE, it showed some error.

                              input style="border: 0px solid white;" size="30" name="uptime" id="uptime" value=" Fatal error: Call to undefined function get_uptime() in /usr/local/www/widgets/widgets/system_information.widget.php on line 159
                              1 Reply Last reply Reply Quote 0
                              • J
                                jikjik101
                                last edited by

                                I want to edit my first post to put "SOLVED" on it but it seems that option is not available.
                                I checked and rechecked again where the edit/modify button is, but i cant find it any longer.

                                Anyway, Abdsalem's solution seems to work:

                                pfSsh.php playback gitsync RELENG_2_0

                                Although the 500 errors don't show up, the solution produces different errors on the syslog, but that is another issue which is here:
                                http://forum.pfsense.org/index.php/topic,57363.msg306384.html#msg306384

                                Thanks again to everyone for your help.
                                Cheers.

                                EDIT: it's weird, i can edit my last post but there is no MODIFY button on my previous posts. ???

                                1 Reply Last reply Reply Quote 0
                                • W
                                  wallabybob
                                  last edited by

                                  @jikjik101:

                                  i can edit my last post but there is no MODIFY button on my previous posts. ???

                                  The "modify" option expires on "old enough" posts?

                                  1 Reply Last reply Reply Quote 0
                                  • J
                                    jikjik101
                                    last edited by

                                    i dunno. how old is old? ???

                                    cmb edited the first post to put the word "solved".
                                    tnx cmb.

                                    1 Reply Last reply Reply Quote 0
                                    • W
                                      wallabybob
                                      last edited by

                                      @jikjik101:

                                      i dunno. how old is old? ???

                                      Evidence suggests the Modify link disappears from the author's view of a post after a time less than 36 60 hours.

                                      1 Reply Last reply Reply Quote 0
                                      • C
                                        carcar
                                        last edited by

                                        hope the dev team will not consider this BUG as solve, as it still exist even in NEW installation(2.0.2)….

                                        1 Reply Last reply Reply Quote 0
                                        • jimpJ
                                          jimp Rebel Alliance Developer Netgate
                                          last edited by

                                          It is solved in 2.0.3.

                                          We haven't made the new release yet.

                                          Remember: Upvote with the 👍 button for any user/post you find to be helpful, informative, or deserving of recognition!

                                          Need help fast? Netgate Global Support!

                                          Do not Chat/PM for help!

                                          1 Reply Last reply Reply Quote 0
                                          • J
                                            jikjik101
                                            last edited by

                                            @jimp:

                                            It is solved in 2.0.3.

                                            We haven't made the new release yet.

                                            alright!!! thanks!  ;D

                                            1 Reply Last reply Reply Quote 0
                                            • First post
                                              Last post
                                            Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.