Forums

Resolved
0 votes
Hi All

Running COS 6.7 pro and getting these errors

Jan 19 20:50:25 cos kernel: dansguardian-av[13747]: segfault at fffffffffffffff8 ip 00007fca9571b4f3 sp 00007fff5061a050 error 4 in libstdc++.so.6.0.13[7fca9567e000+e8000]
Jan 19 23:38:22 cos kernel: dansguardian-av[13745]: segfault at fffffffffffffff8 ip 00007fca9571b4f3 sp 00007fff5061a050 error 4 in libstdc++.so.6.0.13[7fca9567e000+e8000]
Jan 20 01:01:29 cos kernel: dansguardian-av[13670]: segfault at 414 ip 0000000000408283 sp 00007fff50612220 error 4 in dansguardian-av[400000+94000]
Jan 20 05:32:14 cos kernel: dansguardian-av[7735]: segfault at fffffffffffffff8 ip 00007fca9571b4f3 sp 00007fff5061a050 error 4 in libstdc++.so.6.0.13[7fca9567e000+e8000]


I have logged a ticket, but due to time zones I was wondering if anyone here has any advice on this?
Wednesday, January 20 2016, 08:26 AM
Share this post:
Responses (6)
  • Accepted Answer

    Wednesday, January 20 2016, 09:16 AM - #Permalink
    Resolved
    0 votes
    Hi Duncan, I've not got any segfaults here, these are usually a symptom of a hardware (RAM?) issue. What hardware are you using?
    The reply is currently minimized Show
  • Accepted Answer

    Wednesday, January 20 2016, 09:31 AM - #Permalink
    Resolved
    0 votes
    Hi Tim

    Running on a Cisco c 220 M4 32 Gb RAM

    EDIT I enabled kernel logging in the rsyslog conf file
    The reply is currently minimized Show
  • Accepted Answer

    JohnF
    JohnF
    Offline
    Sunday, November 20 2016, 07:58 PM - #Permalink
    Resolved
    0 votes
    Hello all,
    For me, this issue was not related to RAM, but instead to having zeroed out template files. The exact errors looked like
    Nov 20 02:27:21 router kernel: dansguardian-av[853]: segfault at fffffffffffffff8 ip 00007fe053ed0fcc sp 00007ffebdd028e0 error 5 in libstdc++.so.6.0.19[7fe053e12000+e9000]

    The elements which remained the same were
    router kernel: dansguardian-av

    segfault at fffffffffffffff8 ip

    sp

    error 5 in libstdc++.so.6.0.19

    The error arose from zeroing out template files so user report is null, to wit:
    $ > /etc/dansguardian-av/languages/ukenglish/fancydmtemplate.html
    $ > /etc/dansguardian-av/languages/ukenglish/template.html

    To achieve the same goal and stop the segfaults:
    $ echo '<html></html>' > /etc/dansguardian-av/languages/ukenglish/fancydmtemplate.html
    $ echo '<html></html>' > /etc/dansguardian-av/languages/ukenglish/template.html
    The reply is currently minimized Show
  • Accepted Answer

    Monday, November 21 2016, 06:15 AM - #Permalink
    Resolved
    0 votes
    @JohnF Thanks

    Will give it a try, as I still get these seg faults every now and then
    The reply is currently minimized Show
  • Accepted Answer

    JohnF
    JohnF
    Offline
    Monday, November 21 2016, 09:06 AM - #Permalink
    Resolved
    0 votes
    Hi Duncan,
    I did not expect for you to still be having this problem as the original post was 10 months ago. I did this mostly for my records if I run into the problem in the future. However, I will expound upon how I diagnosed this in hopes it may provide you (or anyone else) with some help. First, locate installed packages which are related to "dansguardian" or "content filtering" with this command
    $ yum list installed | grep -iP 'content|dans'
    app-content-filter.noarch 1:2.1.26-1.v7 @clearos-updates
    app-content-filter-core.noarch 1:2.1.26-1.v7 @clearos-updates
    dansguardian-av.x86_64 2.10.1.1-11.v7 @clearos-updates
    dansguardian-phraselists.noarch 2.9-3.v7 @clearos-updates

    Next, find files from these packages which are different from the original. Like this
    $ rpm -V dansguardian-av dansguardian-phraselists app-content-filter app-content-filter-core
    S.5....T. c /etc/dansguardian-av/dansguardian.conf
    S.5....T. c /etc/dansguardian-av/dansguardianf1.conf
    S.5....T. c /etc/dansguardian-av/languages/ukenglish/fancydmtemplate.html
    S.5....T. c /etc/dansguardian-av/languages/ukenglish/template.html
    S.5....T. c /etc/dansguardian-av/lists/bannedsitelist
    S.5....T. c /etc/dansguardian-av/lists/bannedurllist
    S.5....T. c /etc/dansguardian-av/lists/exceptioniplist
    S.5....T. c /etc/dansguardian-av/lists/exceptionsitelist
    S.5....T. c /etc/dansguardian-av/lists/exceptionurllist
    S.5....T. c /etc/dansguardian-av/lists/filtergroupslist
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/banned_portuguese
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_chinese
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_danish
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_dutch
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_french
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_german
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_italian
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_japanese
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_malay
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_norwegian
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_portuguese
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_russian
    S.5....T. /etc/dansguardian-av/lists/phraselists/pornography/weighted_spanish

    Here is a good page to explain the nomenclature. https://www.novell.com/coolsolutions/feature/16238.html
    Since the lists and conf files have been updated after being installed, they can be ignored. This leaves the fancydmtemplate.html and template.html, and deduced a missing resource (zeroed out file) could very likely cause a segfault. More than happy to help if this does not remedy your segfaults.
    Actually, the first thing I did was to run mprime overnight to make certain it was not the RAM, as Tim had suggested. Tim Burgess is very knowledgeable with ClearOS -- may even be one of the developers -- usually Tim's advice is correct when I have an issue, it just happened to not be the case in my situation this time. If you have not run memtest86 or mprime against the hardware for 10+ hours without fault, then you may also want to check this. Hope this adds some clarity and help!
    Sincerely,
    John
    The reply is currently minimized Show
  • Accepted Answer

    Tuesday, November 22 2016, 05:45 AM - #Permalink
    Resolved
    0 votes
    Hi John

    I had wondered how you had arrived at the solution!

    Thank you for the help - I will keep an eye on it and post here my findings
    The reply is currently minimized Show
Your Reply