Perl 6 - the future is here, just unevenly distributed

IRC log for #webwork, 2013-09-10

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:20 mgage_ joined #webwork
02:00 mgage_ joined #webwork
03:15 rbeezer joined #webwork
03:28 rbeezer_ joined #webwork
09:20 eee-blt_ joined #webwork
09:20 aubreyja joined #webwork
09:20 mgage joined #webwork
09:20 ChanServ joined #webwork
11:55 mgage_ joined #webwork
16:56 theseb joined #webwork
16:57 theseb anyone got completely new ver2.7 webwork to install?  i'm having issues with apache2 default dir it seems... "File does not exist: /var/www/webwork2" and "File does not exist: /var/www/webwork".
16:57 theseb I've checked many times that my webwork.apache2-config is exactly
16:57 theseb like webwork.apache2-config.dist by comparing hashes.
16:57 theseb (what i posted on forum)
16:57 theseb I  have set $server_root_url in site.conf to the full URL ..
16:57 theseb http://x.x.x.x.compute-1.amazonaws.com .
16:57 theseb (If I leave $server_root_url equal to an empty string I get a different error instead
17:19 aubreyja hi theseb - still around?
17:19 theseb yes!
17:19 aubreyja just a sec - lemme read your forum post too
17:19 theseb thanks
17:20 theseb aubreyja: say i rem you from mailing list....i think your Jason no?
17:20 aubreyja yep - did you just post your question?
17:20 theseb last nite
17:21 aubreyja do you remember the title?
17:21 theseb you're one of the most helpful contributors...that's how i remembered your name
17:21 aubreyja thanks :)
17:21 theseb it is 1st one of main forum
17:21 theseb let me find link
17:21 theseb hen trying to start Apache2 with fresh install of ver2.7..."File does not exist: /var/www/webwork2" and "File does not exist: /var/www/webwork".
17:21 aubreyja found it - thanks
17:21 theseb http://webwork.maa.org/moodle/mod/forum/discuss.php?d=3090
17:23 aubreyja ok - looks like your cod is at /opt/webwork/webwork2, but apache is trying to resolve the url .../webwork2 to /var/www/webwork2
17:23 theseb yes
17:24 aubreyja what os is the amazon share?
17:24 theseb ubuntu 12.04 LTS
17:24 aubreyja ah - was this from following the aws set up instructions on our wiki?
17:24 theseb i tried that days ago but the AWS image was ver2.5
17:25 theseb so i thought it'd be better to try a virgin install from source myself using ver2.7
17:25 aubreyja ok, and do you have any other websites, apps, etc., running on the machine?
17:25 theseb no
17:25 aubreyja kk - just a sec
17:27 rbeezer joined #webwork
17:27 theseb aubreyja: say..i'm going to blow this image up and do it over anyway...if you can post your RSA public key somewhere I can give you access to the image
17:28 theseb if necessary
17:28 aubreyja sure - an ssh key?
17:28 theseb yea..i think i need to add an RSA public key to $HOME/.ssh/authorized_hosts
17:29 theseb then anyone can ssh to it
17:29 theseb err....anyone w/ that key
17:29 theseb you'll still hold the private key...don't give me that
17:29 theseb probably in your $HOME/.ssh/id_rsa.pub if you use linux
17:30 theseb be back in a few min
17:30 aubreyja sure - I've done this before - just a sec
17:33 aubreyja ok, I think I know what the problem is, but I don't know why it isn't typically a problem.
17:34 aubreyja in /etc/apache2/sites-enabled there is a file called 000-default
17:34 aubreyja it sets the DocumentRoot to /var/www
17:36 aubreyja Try commenting out that line and restarting the server.
17:37 aubreyja You may need to specifically set a different document root in the webwork.apache2-config, but let's try this first...
17:43 aubreyja Also, if you want to reinstall you can use my installation script: https://github.com/aubreyja/ww_install
18:23 theseb ok i'm back
18:24 theseb ok i'm trying what you said now
18:36 theseb aubreyja: didn't work..i tried a couple of different things....also do you set $server_root_url   = ""; ?
18:36 theseb when i do that I get..Syntax error on line 45 of /opt/webwork/webwork2/conf/webwork.apache2-config:
18:36 theseb unable to determine apache server url using course environment |WeBWorK::CourseEnvironment=HASH(0x7fdb9db40b28)|.check that the variable $server_root_url has been properly set in conf/site.conf\nCompilation failed in require at (eval 1324) line 2.\n\t...propagated at /usr/share/perl/5.14/base.pm line 93.\nBEGIN failed--compilation aborted at /opt/webwork/webwork2/lib/WeBWorK/ContentGenerator/Instructor/PGProblemEditor3.pm line 20.\nCompilation failed in requir
18:36 theseb e at (eval 1323) line 2.\nBEGIN failed--compilation aborted at /opt/webwork/webwork2/lib/WeBWorK.pm line 85.\nCompilation failed in require at /opt/webwork/webwork2/lib/Apache/WeBWorK.pm line 35.\nBEGIN failed--compilation aborted at /opt/webwork/webwork2/lib/Apache/WeBWorK.pm line 35.\n
18:37 theseb Action 'configtest' failed.
18:37 theseb wow that was long...sorry
18:37 aubreyja hi - you should probably try localhost for that and then try the public url
18:38 theseb public url gives the original error
18:38 theseb i'll try localhost now
18:40 theseb aubreyja: http://localhost has original prob still... stuff like File does not exist: /var/www/webwork2
18:40 aubreyja ok, I think there is something nonstandard about your apache set-up.
18:41 aubreyja It's trying to resolve everything to /var/www
18:41 theseb yea i didn't do much with it..i can delete and reinstall apache..the only change i made was /etc/apache2/httd.conf includes the webwork conf file
18:41 aubreyja is this a vanilla ubuntu 12.04 - i.e. could amazon have modified it in some way to make it 'easier' to put up a website on aws?
18:42 theseb possibly
18:42 theseb aubreyja: want to ssh in and take a peak? others may have the same issue if they use Amazon
18:43 aubreyja sure, but I'm on a new computer and haven't made any new ssh keys.  I have a meeting coming up in 15 min. too. So, I'll try to find you this afternoon and we can try it.
18:43 theseb ok
19:25 goehle joined #webwork
20:10 theseb aubreyja: must leave for today..will be back tomorow...if you want me to try something my email is cseberino@gmail.com
22:08 mgage_ joined #webwork
22:11 goehle hey mgage_
22:12 goehle (or mgage)
22:12 goehle have you seen this
22:12 goehle http://bugs.webwork.maa.org/show_bug.cgi?id=2760
22:12 mgage_ yes -- I've been watching it -- I don't have a great answer -- at least not yet
22:12 goehle I just put an update up
22:13 mgage_ yep -- reading it now
22:16 mgage_ I don't have a better answer at the moment. It seems to me that trying to fix HTML::Scrubber is still the quickest way to resolve this.  There might be a more elegant way.
22:17 goehle how quick do you want the fix
22:17 mgage_ but I don't see it at the moment.
22:19 mgage_ well we probably need to do something fairly quickly -- because existing problems are breaking.  Just disabling HTML::Scrubber puts us back where were for years  ---hypothetically ripe for XSS nattiness -- but as far as we know it never actually happened.
22:20 mgage_ If we don't see a quick fix that has a reasonable chance of working then backing off HTML::Scrubber and designing something elegant might be the best option.  If there is "one more thing for try" -- let's do that before we give up.
22:20 goehle sigh
22:21 mgage_ exactly --http://xkcd.com/1172/
22:22 mgage_ make sure you read the pop up when you hover over the graphic
22:23 goehle :D
22:24 goehle well I may have stumbled on a fix
22:24 goehle I backed off scrubber to allow html comments
22:24 goehle and its letting things through
22:24 goehle no promises that it doesnt' break other stuff
22:24 goehle but this bug will go away
22:26 mgage_ kk -- I think that is a reasonable thing to try -- meantime let's alert Davide (for MathObjects) and think hard about whether there is another way to prevent html getting into the database that is more robust.  or a way (Perhaps using taint) reliably making sure that it is never rebroadcast without HTML encoding.
22:27 goehle yeah
22:27 goehle thats a good idea
23:35 goehle_ joined #webwork
23:36 goehle_ hey mgage / mgage_
23:36 mgage_ yep -- I'm reading Davide's response (and writing my own) -- he has good points, but I'm not sure has caught the problem of saving weird stuff to the past_answer database
23:36 mgage_ I'm looking now
23:37 mgage_ where is that save to the database and to the log file made?
23:37 goehle_ its made via ProblemUtils.pm
23:37 goehle_ in Problem.pm
23:39 goehle_ although its also done in gateway
23:39 goehle_ seperately
23:39 goehle_ I put the scrubber where I did because it would remove <script> tags immediately
23:40 goehle_ before they could be processed by anything
23:40 mgage_ what kind of answers are being passed -- could they be scrubbed here?
23:40 goehle_ here?
23:40 mgage_ and in the PGbasicmacros as Davide suggests
23:40 mgage_ hang on
23:44 mgage_ they are being printed from $pg->{answers}  -- if we have scrubbed those answers from within pg then I don't think anything bad reaches the database.
23:46 goehle_ sign
23:46 goehle_ sigh
23:46 goehle_ I dont think I'm for fixing it in any case
23:46 goehle_ I'll just remove the scrubber stuff
23:46 mgage_ I think that is best for now -- this is a bigger can of worms than I can tackle at the moment as well.

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary