00:48:54 <Spork> lol
00:48:57 <AaronSw> * AaronSw reconnects
00:50:44 <Spork> * Spork sporks off another process
01:07:14 <davb> davb has quit
01:07:14 <docwolf> docwolf has quit
01:07:14 <rbm> rbm has quit
01:07:35 <docwolf> docwolf has joined #openacs
01:07:35 <rbm> rbm has joined #openacs
01:07:35 <davb> davb has joined #openacs
01:18:28 <Spork> wb
02:10:42 <davb> davb has quit
04:41:26 <Spork> anyone awake?
04:45:27 <docwolf> docwolf has quit
08:10:38 <ola> ola has joined #openacs
09:52:34 <rzolf> rzolf has joined #openacs
09:52:40 <rzolf> anyone alive?
09:53:28 <rzolf> rzolf has quit
11:43:00 <ola> ola has quit
11:57:36 <ola> ola has joined #openacs
12:20:54 <ola> ola has quit
12:43:10 <ola> ola has joined #openacs
12:52:11 <ola> ola has quit
12:56:29 <ola> ola has joined #openacs
13:03:55 <davb> davb has joined #openacs
13:05:57 <davb> davb has changed the topic to: Free Web Toolkit http://openacs.org
13:07:58 <ola> morning davb.
13:10:47 <john> john has joined #openacs
13:12:45 <davb> Hi ola.
13:13:02 <davb> Hi john
13:14:19 <ola> * ola goes to eat
13:27:12 <john> hey davd
13:34:15 <john> I'm having trouble getting postgres to dump and load with a fresh OACS4 install...
13:34:35 <john> the syntax looks pretty straightforward,  so I'm wondering if there's anything special I should be doing
13:35:59 <john> pg_dump --file=yadayada.tar --format=t openacs4
13:37:29 <davb> What happens or does not happen when you use that command?
13:48:26 <davb> and what postgresql version?
13:48:37 <john> the dump works fine.
13:48:43 <john> I've tried two restore methods
13:48:53 <john> the first:  loading restore.sql into psql
13:49:02 <john> (after untaring the archive)
13:49:52 <john> I'll find that error in a sec... but I read in the docs that I should just be using pg_restore with the tar file
13:50:20 <john> When I do that,  I get a error message on the COPY function.
13:50:51 <john> PG 7.1
13:51:05 <davb> just 7.1 or 7.1.x?
13:51:24 <davb> I think there was a bug in 7.1.1 or 7.1.2 with dump/restore
13:52:07 <john> err,  PG 7.1.2 to be precise
13:52:46 <john> aaaah.  ok.  do you think  I should go to  7.1 instead... or just wait for a patch
13:52:58 <davb> No, I think it is OK.
13:53:24 <davb> I am not the postgresql expert around here.:)
13:54:51 <davb> You might want to try #postgresql.
13:56:36 <john> good advice,  thanks.  I though there might be an easily solved problem specific to OACS,  but that was wishful thinking that
13:57:19 <davb> I haven't tried dumping or restoring the database yet. It could be an OpenACS issue.
14:38:14 <Spork> bah!
14:38:18 <Spork> :)
14:38:20 <Spork> morning
14:39:49 <davb> Hi Spork
14:39:58 <Spork> hey dave
14:41:34 <ola> hey spork!
14:41:56 <Spork> Hola Ola
14:42:25 <Spork> can someone go to www.neverhere.com and tell me if it works?
14:42:42 <davb> looks ok to me
14:42:48 <Spork> cool
14:42:51 <Spork> :)
14:42:55 <Spork> What about the site itself
14:44:17 <ola> is it yours?
14:44:39 <Spork> yeah
14:45:02 <Spork> hasn't been updated for a while though
14:45:14 <Spork> I'll update it tonight or tomorrow 
14:45:17 <Spork> gotta add to the blog
14:45:24 <ola> the animated fountain is way cool!
14:45:26 <AaronSw> * AaronSw disconnects
14:45:28 <ola> :-)
14:45:57 <davb> not bad
14:46:12 <Spork> :)
14:46:14 <davb> ARGH. Anyone know how to use GET DIAGNOSTICS in postgresql?
14:46:44 <davb> I can't find it in the postgresql docs...
14:47:06 <AaronSw> * AaronSw reconnects
14:47:44 <ola> heh. I thought "Nick B" was fake for "nickname B".
14:49:15 <Spork> lol
14:53:21 <Spork> i like that site
14:53:27 <Spork> simple but useful (kinda)
14:54:15 <davb> rebuild it with OpenACS 4!!(:))
14:55:24 <Spork> uh huh
14:55:34 <Spork> I don't think my P200 can hold Oracle or PG
14:55:48 <davb> aha.
14:56:00 <davb> yeah that will slow you down.
14:59:22 <davb> Hey, I tried to install oralcle on a P200 with 64MB ram. after like 18 hours I killed it and bought a new machine with 256MB. Then I switched to Postgresql anyway.
14:59:53 <davb> althought it is running nicely as a mail/dns/apache web server.
15:07:48 <irc.openprojects.net> Disconnected from irc.openprojects.net (ERROR :Closing Link: loggy[xcdfddb76.ip.ggn.net] by asimov.openprojects.net (Ping timeout for loggy[xcdfddb76.ip.ggn.net]))
15:07:53 <loggy> loggy has joined #openacs
15:07:53 <benford.openprojects.net> topic is: Free Web Toolkit http://openacs.org
15:07:53 <benford.openprojects.net> Users on #openacs: loggy john davb ola rbm abbaJ Spork GEM_ chump AaronSw 
15:07:53 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
15:14:35 <ola> davb: I think you should declere a VARIABLE int4 := 0; and then put: "GET DIAGNOSTICS VARIABLE = ROW_COUNT;" after your query or dml.
15:15:41 <ola> (just guessing though)
15:17:27 <davb> ok. I just declared the variable in the DECLARE section in pl/pgsql. thanks
15:18:59 <ola> does it work?
15:19:20 <davb> not the way I did it.
15:20:07 <ola> this is what I checked: http://www.linuxgazette.com/issue69/nielsen.html
15:20:16 <davb> thanks.
15:20:33 <davb> Oops, my declaration appears to have not survived. let me try again.
15:27:23 <davb> nope. Whats the latest version of postgresql?
15:27:28 <davb> I am running 7.1.2
15:28:04 <ola> 7.1.3
15:28:14 <davb> aha.
15:28:49 <davb> Wow, not bad.
15:29:26 <davb> didn't help, but at least I am updated.
15:32:59 <Spork> doh
15:34:06 <davb> argh. 
15:34:18 <davb> according to all the docs I am doing it right.
15:37:06 <davb> aha. It needs to be DECLAREd
15:39:10 <davb> the GET DIAGNOSTICS works! Now that I am done debugging my debugging statements I can debug my function :)
15:39:44 <davb> I think this is it: ERROR:  Referential Integrity: attempting to delete live_revision: 2469
15:39:55 <davb> That is not showing up in the logs though...
15:40:58 <davb> Oh crap. I think its that Postgresql referential integrity in a transaction problem again.
15:43:05 <ola> davb, what do you mean by; "It needs to be DECLAREd"?
15:43:41 <davb> just doing v_rec_affected :=0; apparently is not creating the variable. Either that or the variable name I picked postgresql didn't like. 
15:43:56 <davb> so I put v_rec_affected integer; in the declare section.
15:44:02 <ola> ok. I see.
15:44:21 <davb> doesn't matter. the CR won't let me delete a live revision.
15:48:43 <ola> Spork, can I name my company "Spork"? the authorities didn't approve of "Utilus" - it was too close to another name...
15:49:06 <davb> I suspect it is Trademarked at least in the US
15:49:28 <davb> aha DanW took out an RI trigger for just this problem, so why is it still occuring?
15:49:47 <ola> bummer.:-)
15:50:16 <davb> oh wait, he rewrote the triggers, so either they are still broken or I am doing something wrong.
15:51:07 <davb> maybe I just have to unspecify the live revision.
15:51:16 <ola> have you updated lately?
15:51:25 <davb> last week i think.
15:51:29 <davb> friday.
15:52:49 <davb> ack.
15:53:05 <davb> now it says I am deleteing the LATEST revision. How can I get rid of it...
15:53:16 <Spork> ugh
15:53:20 <Spork> I hate not having order
15:53:58 <davb> aha. There are two columns, live_revision and latest_revision.
15:54:49 <ola> is it the drop script you're working on? (twenty questions, sorry!)
15:55:17 <Spork> me?
15:55:21 <davb> no, I am afraid to touch that.
15:55:23 <davb> :)
15:55:40 <davb> I am just working on the function that removes a page frm the database if it is removed frm the filesystem.
15:56:53 <Spork> oh
15:57:09 <Spork> can one of you guys tell the main chief not to have all packages checked by default?
15:57:18 <Spork> It's not efficient ;)
15:57:25 <davb> use UNCHECK ALL
15:57:36 <Spork> yes I know
15:57:41 <Spork> but it takes an extra 2 minutes
15:57:46 <Spork> plus, who installs ALL the packages?
15:57:56 <davb> Don't CHECKOUT all the packages. 
15:58:04 <Spork> :P
15:58:10 <davb> heh
15:58:36 <davb> it is a pain. Thats why I am just using acs-core for development. took 5 mins of my install time.
15:58:55 <davb> the installer checks every package .info file 
16:01:27 <ola> Spork: help me come up with a company name, will you? Dave suspects that "Spork" is Trademarked ;-)
16:02:06 <ola> I need a name by next week!
16:02:16 <Spork> heh
16:02:20 <Spork> What kind of a name?
16:02:50 <ola> a company name that's being globally useful.
16:03:15 <davb> Aha. ourstuffworks.com?
16:03:28 <ola> not so shabby.
16:03:35 <Spork> MidgetSoft...We cater to your midget needs!
16:03:41 <ola> heh
16:03:50 <davb> globallyuseful?
16:04:28 <ola> well, so that swedes can understand, something latin perhaps?
16:04:31 <Spork> awww crap
16:04:48 <Spork> Whats' that command to list username in Solaris mark mentioned last time
16:04:57 <ola> or a phrase from a movie..
16:05:39 <ola> whois?
16:06:30 <Spork> no
16:06:37 <Spork> it told you your user id and group id
16:08:00 <Spork> how about joopentalogy.com
16:08:26 <davb> brb
16:08:29 <davb> davb has quit
16:12:19 <davb> davb has joined #openacs
16:12:38 <davb> I had to reboot to install the smartmedia floppy adapter driver.
16:15:11 <Spork> this sucks
16:18:27 <ola> wow! Redhat 7.2 is out. woo-hoo!
16:18:48 <davb> Wow, does it work this time? :)
16:19:48 <ola> hopefully:-) 7.1 worked quite well actually.
16:20:33 <davb> cool. I have to upgrade to a 2.4 kernel. My last compile was less than sucessful.
16:21:15 <Spork> You guys sure did a lot in 1 week
16:21:29 <Spork> aside from breaking all of my stuff :P
16:21:33 <davb> heh
16:22:04 <davb> were you relying on buggy behavior?
16:22:10 <Spork> yes
16:22:47 <davb> what is broken?
16:24:03 <Spork> ummm
16:24:06 <Spork> all my links
16:24:11 <Spork> even though they point to the right spots
16:25:52 <davb> hmmmm...weird.
16:26:11 <davb> I think it has to do with the host_map. How is your hostname set in the tcl config file?
16:26:33 <Spork> *hits head on desk repeatedly*
16:26:45 <Spork> I set it to cv.net
16:27:34 <davb> OK. Is the machine called cv.net?
16:27:43 <Spork> probably not
16:28:04 <Spork> i'm really starting to dislike my job
16:28:28 <Spork> see
16:28:32 <Spork> I have this
16:28:34 <davb> could be screwing it up. That host_map patch is for domain based subsites or soemthing. So it redirects to the host name which != the hosts name.
16:28:43 <davb> or something like that.
16:28:45 <Spork> ./packages/snmp/www
16:28:52 <Spork> where the link points to
16:28:52 <davb> * davb wings it again
16:28:57 <Spork> and no files are in there
16:29:03 <Spork> i mean it won't open them
16:29:17 <davb> I think that is the wrong way to program it...
16:29:36 <davb> the links should just go to "pagename"
16:29:51 <Spork> how can that be?
16:30:03 <davb> or whatever. The pages are mounted in the tree somewhere. not necssarily /snmp
16:30:05 <Spork> I mean...I have an index in the /web/oacs/www/index.*
16:30:16 <Spork> which is the main page to the site
16:30:23 <Spork> from there I have an administer site link
16:30:34 <davb> hmmmmm
16:30:36 <Spork> add,edit,remove, modem poll/history, change pass and log out
16:30:49 <davb> hmmmm.
16:30:49 <Spork> last two and administer site links work
16:31:16 <Spork> the others always pointed (and worked) in /web/oacs/packages/snmp/add-user*
16:31:21 <Spork> and so forth
16:31:25 <Spork> now I can't go to it
16:31:36 <davb> weird.
16:31:47 <davb> the links pointed to the filesystem path??
16:32:01 <davb> not the web-site path?
16:32:03 <Spork> yes
16:32:09 <Spork> here's my link
16:32:38 <davb> I wonder how you do that.
16:32:57 <davb> Link to packages that are mounted at some arbitrary path.
16:33:02 <Spork> <li><a href=/snmp/www/user-add>Add </a> Users</li> 
16:33:02 <davb> from the index page.
16:33:12 <Spork> worked fine before
16:33:15 <Spork> now it's broken
16:33:18 <davb> I am surprised that worked. 
16:33:28 <Spork> how am I supposed to do it then
16:33:32 <davb> Not sure.
16:33:37 <davb> where is the package mounted?
16:33:44 <Spork> ./snmp
16:34:00 <davb> so probably just /snmp/user-add
16:34:09 <Spork> tried that
16:34:12 <Spork> same results
16:34:16 <davb> the www is the filesystem path. the Request processor converts URLs to filesystem paths.
16:34:26 <davb> Sounds like the host_map wackiness.
16:35:32 <davb> try using the IP address
16:36:44 <Spork> i am using the ip address
16:37:03 <davb> aha. 
16:37:12 <davb> hmmmm. you updated just now?
16:37:37 <Spork> last night
16:37:51 <davb> I see where it is definitely breaking if you use the IP address, but there seems to be a patch...
16:38:04 <Spork> It worked fine up to now
16:38:09 <davb> that was applied to CVS 10/19
16:38:17 <Spork> where
16:39:07 <davb> sounds like you are getting this error: content_item__delete works correctly.  Are you running the
16:39:07 <davb> latest code from cvs?  Also what version of postgresql are you running?
16:39:13 <Spork> oracle
16:39:16 <davb> oops
16:39:26 <davb> I looked into this. It's not supposed to insert an entry for the main site. However, the code is a bit fragile - it depends on your host as determined by AOLserver when you hit the site be the same as the host defined in your AOLserver initialization file. 
16:39:47 <Spork> ?
16:40:11 <davb> and this: Further info on this issue for consideration: I now can get the problem "solved" when accessed using the "correct" URL, such as http://testacs.grcomputing.net:8000 but the issue is still there if the same server is accessed by IP address as http://192.168.1.1:8000 instead. 
16:40:26 <Spork> where are you posting this from
16:40:29 <Spork> I might as well go rad it
16:40:34 <Spork> s/rad/read
16:40:50 <davb> here's thje whole thread
16:40:51 <davb> http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0002wj&topic_id=14&topic=OpenACS%204%2e0%20Testing
16:40:52 <chump> A: http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0002wj&topic_id=14&topic=OpenACS%204%2e0%20Testing from davb
16:41:12 <davb> A:|Thread on host_node_map
16:41:12 <chump> titled item A
16:43:37 <Spork> don't think that's the same problem
16:43:59 <davb> are you getting any sort of error message in your browser or the log?
16:44:23 <Spork> not the browser
16:44:28 <Spork> let me check/make a log
16:47:16 <Spork> [22/Oct/2001:12:42:55][22797.1][-main-] Notice: nssock: listening on 167.206.9.143:8080
16:47:18 <Spork> see
16:47:21 <Spork> it listens on the ip
16:47:23 <Spork> I use ip
16:47:26 <Spork> i can get to main site
16:47:28 <Spork> not my subsite
16:51:14 <Spork> you know
16:51:20 <Spork> if I wrote this in java I'd be done already
16:51:28 <davb> heh. Sure.
16:51:36 <docwolf> docwolf has joined #openacs
16:51:37 <Spork> Well, I'd have a lot less problems
16:51:47 <davb> maybe. I have never programming in Java.
16:51:48 <docwolf> Hi Everybody!
16:51:51 <davb> hi docwolf
16:51:53 <Spork> hello
16:52:33 <Spork> i think I spend every monday trying to figure out WHAT got changed in OpenACS and fix my stuff accordingly
16:53:43 <davb> I don't think anything else really changed.
16:54:00 <docwolf> http://digitalmass.boston.com/news/globe_tech/at_large/2001/1022.html
16:54:01 <chump> B: http://digitalmass.boston.com/news/globe_tech/at_large/2001/1022.html from docwolf
16:54:41 <rbm> moo
16:56:30 <davb> hi rbm
16:56:58 <Spork> hey Roberto
16:58:00 <Spork> I think I might as well delete everything and start from scratch
16:59:57 <davb> just go back to before you updated.
17:00:20 <davb> I think I need to do that. My system is now also hosed.
17:01:03 <davb> weird, my server isn't even starting now. That sounds like an un-openacs problem.
17:01:56 <rbm> what's up everyone/
17:02:14 <rbm> * rbm had a good chat with talli over MSN
17:02:33 <davb> I am trying to track down a problem in static-pages, I cvs updated and now my aolserver is not starting or logging anything.
17:02:39 <davb> rbm: cool
17:02:55 <davb> using a MS operating system?
17:02:57 <davb> :)
17:03:02 <rbm> You cvs updated AOLserver or OACS?
17:03:18 <davb> no, just openacs.
17:03:23 <rbm> davb: No. I use everybuddy as IM, so that I can use MSN, AIM and ICQ at the same time.
17:03:34 <davb> cool.
17:03:41 <rbm> But I was over at a computer lab on campus and I decided to take a look at the windoze client
17:04:18 <davb> I also upgraded postgresql to 7.1.3. Is that a bad idea?
17:05:55 <rbm> No. It's a good idea.
17:06:14 <rbm> But something may have gotten screwed up in your upgrade, especially if you let your packaging system handle it.
17:08:28 <davb> It seemed to be working OK, until I updated OpenACS
17:08:55 <davb> now I can't login to the server. ack.
17:09:02 <rbm> * rbm goes to eat something. brb
17:09:06 <docwolf> ugh
17:09:16 <docwolf> so it looks like aD might really be going closed source
17:09:45 <davb> docwolf: according to those quotes, yeah.
17:10:19 <davb> Well we all know poorly thought out well marketed closed-source products already sell very well.
17:10:44 <davb> * davb goes to eat also
17:27:25 <Spork> time to recreate everything
17:29:19 <Spork> did I ever tell you all I hate mondays?
17:30:29 <davb> I hate it when I can't SSH into my machine.
17:30:41 <Spork> can you ping it?
17:31:18 <davb> good question. I'll try.
17:31:47 <davb> nope. it didn't restart for some reason. I hate when that happens.
17:32:07 <Spork> aha
17:32:11 <Spork> Time to make a trip eh?
17:32:20 <Spork> or phone calls
17:32:21 <Spork> ehehe
17:32:59 <davb> i'll just wait until tonight.
17:33:04 <Spork> :)
17:33:50 <Spork> ok
17:33:57 <Spork> should I be making my thing a subsite now?
17:35:15 <davb> I don't think so. Shouldn't you just mount it under the root?
17:36:41 <Spork> I don't think I can do that without screwing up a bunch of things
17:37:04 <Spork> i need the main page to look a specific way
17:37:26 <Spork> and links go to a ~/packages/snmp/www/files*
17:37:48 <Spork> and no access to any other parts of the site unless you're admin
17:37:59 <Spork> i had all of this working
17:39:09 <davb> aha
17:40:50 <davb> hmmmmm.
17:42:02 <Spork> ?
17:42:12 <davb> so all requests to /snmp/ go to the subsite?
17:42:54 <Spork> technically...yes
17:42:58 <Spork> but see
17:43:01 <Spork> not all
17:43:09 <Spork> changing password goes to /admin something or other
17:43:17 <Spork> same thing with log out and re-login
17:43:42 <davb> ok. 
17:43:48 <davb> The main site, is also a subsite.
17:43:54 <davb> you already know that.
17:43:55 <Spork> and the links for add,edit,remove go to hacked versions of /admin-*
17:44:00 <davb> right.
17:44:08 <Spork> which are in my /snmp
17:44:17 <Spork> at least until this whole thing is finalized
17:44:43 <Spork> the new ACS lets me add roles to the subsite stuff too
17:44:51 <Spork> so I'm not sure what goes where
17:45:02 <Spork> right now, I'm reinstalling the data model
17:45:15 <davb> fun.
17:45:29 <Spork> not really
17:45:38 <davb> that might be why it was busted. It adds a table called host_node_map
17:46:01 <Spork> yeah so I heard
17:46:05 <Spork> how do I fix this
17:46:25 <Spork> I'm thinking I can change so everything is in /web/oacs/www
17:46:29 <Spork> no links to /snmp
17:46:38 <Spork> so I won't even have to mount anything
17:46:47 <davb> Well you have to mount it anyway I think.
17:47:01 <Spork> why
17:47:05 <Spork> No files are in /snmp anymore
17:47:08 <davb> ETP does it like that.
17:47:17 <Spork> I'll give it a shot
17:47:19 <Spork> can't hurt
17:47:29 <davb> Right, but you need to mount the package or it won't serve pages.
17:47:39 <Spork> no I don't :)
17:47:58 <Spork> i'm going to wind up turning the front page stuff into "my site"
17:48:26 <Spork> the sql-create scripts can be run by hand
17:48:41 <davb> aha.
17:48:45 <Spork> ?
17:48:47 <davb> OK that might work.
17:49:02 <Spork> i mean, how often do we run those anyway
17:51:42 <davb> sure that should work in a one-time installation. 
17:51:55 <Spork> You plan on reinstalling your package daily? :)
17:52:03 <davb> so far...
17:52:08 <Spork> heh
17:52:57 <Spork> I think I'll become a farmer
17:56:07 <Spork> i have to go to school in 4 hours
17:56:09 <Spork> *sigh*
17:56:46 <davb> http://www.alistapart.com/stories/emen/
17:56:47 <chump> C: http://www.alistapart.com/stories/emen/ from davb
17:56:57 <davb> C:|The Trouble with EM 'n EN
17:56:57 <chump> titled item C
17:57:01 <davb> C: from A List Apart
17:57:02 <chump> commented item C
17:57:08 <Spork> lovely
17:57:16 <davb> C: proper typesetting marks in HTML 4.0?
17:57:16 <chump> commented item C
17:57:29 <Spork> I just ran my snmp-create and it didn't register
17:57:32 <Spork> but I got no errors
17:58:19 <Spork> odd
18:00:38 <Spork> manager isn't a valid privilege
18:00:38 <Spork>     while executing
18:02:26 <Spork> * Spork hits head on desk
18:02:28 <Spork> SQL> /
18:02:28 <Spork> PRIVILEGE               PRETTY_NAME             PRETTY_PLURAL
18:02:28 <Spork> ----------------------- ----------------------- -----------------------
18:02:28 <Spork> read
18:02:28 <Spork> write
18:02:29 <Spork> create
18:02:31 <Spork> delete
18:02:33 <Spork> admin
18:02:35 <Spork> supervisor
18:02:37 <Spork> manager
18:08:55 <Spork> Where's mark when you need him
18:21:59 <talli> talli has joined #openacs
18:22:03 <talli> hey guys
18:22:08 <Spork> hi talli
18:22:14 <Spork> Dave, you still with the living?
18:23:52 <Spork> greaaat
18:23:56 <Spork> my scripts won't work anymore
18:29:43 <donb> donb has joined #openacs
18:29:53 <donb> Hey ... is talli here actually?
18:30:37 <donb> Is *anyone* here?
18:31:37 <Spork> yes
18:31:52 <Spork> I'm here trying to find out what OACS changed that my thing is totally fubarred
18:32:10 <Spork> talli was awake about 10 minutes ago
18:34:07 <donb> Well ... do you have more details?  Was it working with an earlier OACS (4? 3?) or Classic (4? 3?)
18:34:13 <Spork> It worked a week ago
18:34:26 <donb> OACS 4?
18:34:28 <Spork> yes
18:34:35 <donb> What are the symptoms?
18:34:47 <Spork> It's saying manager is not a valid privilege
18:34:58 <donb> Which package?
18:34:59 <Spork> Although my snmp-create.sql clearly creates one
18:35:07 <Spork> the one I'm writing :(
18:35:08 <donb> Oh your package ... hmmm...
18:35:10 <Spork> or was...
18:35:22 <donb> So ... can you select it out of the perms table by hand?
18:35:30 <Spork> lemme check
18:36:28 <donb> My first approach to debugging any weirdness involving the RDBMS is to jump right into PSQL (or SQL*Plus) and verify that I can do by hand what I think my code should be able to do by script...
18:36:59 <Spork> well, yes I can
18:37:07 <Spork> but now they're getting messed up object_ids
18:37:15 <Spork> -3       2198 supervisor
18:37:15 <Spork>         -3       2195 manager
18:37:18 <Spork> -2?
18:37:22 <Spork> ack -3...
18:37:51 <Spork> isn't that default context?
18:38:02 <donb> Something like that, yes.
18:38:12 <Spork> ok, well, they are being created
18:38:15 <donb> One of the magic pre-defined objects 
18:38:39 <donb> Did you restart AOLserver after installing/mounting your package???
18:38:46 <Spork> I haven't mounted it 
18:39:01 <davb> Spork is -3 the id? or is 2198 the id and -3 the context or whatever?
18:39:15 <Spork> yeah 2198 is the id and -3 is the context
18:39:18 <donb> Hmmm...OK.  Are perms cached by anything? (I don't know offhand) ...
18:39:27 <Spork> I'm restarting aolserver again but it didn't work the first time
18:39:38 <Spork> and...it bombs out
18:39:42 <donb> OK If you restarted once that should be enough anyway :)
18:39:45 <donb> Right ...
18:39:49 <Spork> You guys didn't make mounting a package a requirement now did you?
18:40:02 <donb> No, shouldn't be a requirement.
18:40:07 <Spork> that's what I thought
18:40:15 <Spork> but it still bombs out :(
18:40:26 <donb> What URL is bombing out one in the perms code itself?
18:40:45 <Spork> well, what I did (now that everything is broken)
18:41:07 <Spork> is take my old index.adp/tcl that I used (custom) and moved it into a /www/myindex
18:41:21 <Spork> so when I go to server.com/myindex it should show me my files instead
18:41:32 <Spork> but it just chokes with "manager isn't a valid privilege"
18:41:55 <Spork> and I have my Manager and Supervisor groups as well
18:42:14 <Spork> i'm at a loss...no clue what messed up
18:42:23 <donb> I'm grep'ing to see just where that error comes from ...
18:42:46 <Spork> k
18:43:14 <Spork> gotta love it though...
18:43:23 <Spork> now I have a lot more to write in my status report
18:43:53 <Spork> * Spork looks forward to doing this all next monday
18:45:41 <talli> donb: has there been any resolution as to how SDM will be made available?
18:47:03 <donb> Check ad_permission_p in acs-tcl/tcl/acs-permissions-procs.tcl.  It does a "search count(*) from acs_privileges where privilege = :privilege" query before giving that error message.  Make sure that SQL logging's turned on for your DB pools and check your log to make sure that query's being issued.  It indicates that the select ain't finding no rows which is odd since you can do so from PSQL (or SQL*Plus, I don't know which RDBMS you're using).  This is
18:47:15 <donb> THere's no chance one or more of your db pools is misconfigured?
18:49:17 <donb> Talli - not that I know of.  Someone should look into the "gatekeeper" package to see if it will do what it claims it can do, i.e. glue two things together.  We could also look into writing ACS 3.x cookies when folks log in to the  new site or dropping in code to read 4.x style cookies.  Getting the SDM ported over is a high priority for me, too, I'd sure like to get a couple of folks engaged in that task soon.
18:49:51 <talli> i asked vinod about it and he said he was swamped but wanted to get to it eventually
18:50:11 <talli> we are bidding on a project that would like to use a MS project like system for tracking progress
18:50:25 <Spork> don, I don't think it's misconfigured...are you referring to the aolserver startup script
18:50:28 <talli> the project probably doesn't need something like that at all, but they want some answer at least
18:50:31 <donb> Talli - also folks are reporting that the CVS version of bboard only allows the site admin to post to a bboard.  I know the alpha site allows anyone to post so you guys have fixed it, rigged permissions or whatever's necessary...
18:50:45 <talli> yeah, we haven't seen that issue
18:51:06 <donb> Well ... the SDM isn't really a project manager at this point, but it would sure be nice to grow it into something more complete.  Integrated with an Intranet...
18:51:15 <talli> but then have you ever tried to give an individual the permission to post tot he bboards?
18:51:28 <talli> yeah, i don't really think SDM is appropriate for their need
18:51:42 <talli> project tracker purpotedly does this, but then again it really is vapor ware
18:51:49 <donb> I know Vinod's swamped, I'm bummed.   As far as bboard goes I've not done any significant testing myself.
18:52:00 <talli> the admin UI is an embarrassment
18:52:13 <donb> project tracker is indeed vaporware and I have no idea if they even attempted intranet integration, since that's vaporware, too.
18:52:26 <talli> there are about 10 types of permissions when all you really want is read, write and admin
18:53:04 <donb> Spork - I'll hang around a bit to try to figure out what's causing your problem if you're going to work on it right now ...
18:53:15 <Spork> I am working on it right now :/
18:53:19 <donb> Great ..
18:53:29 <Spork> I don't have anythign else to do...
18:53:34 <Spork> Aside from OpenACS stuff
18:53:46 <talli> so i wouldn't be surprised if these guys just are having rtouble figuring out how to assign correct permissions
18:54:02 <Spork> Maybe I should run this whole site thing by you so you can point me in the proper way of developing it :(
18:54:35 <donb> Spork - you can probably get a lot of help in the forums, too, unless it's too proprietary to share ...
18:54:40 <Spork> because Dave Mark Vinod or Ola weren't very accertive
18:54:52 <Spork> I don't know if it is or not
18:55:14 <Spork> My boss wanted me to add on a monitoring package type thing 
18:55:34 <donb> Talli - you're probably right.  Maybe you or someone could check the forum and give a quick bit of help?  I'd like to know if it is a bug or cockpit error soon...
18:55:52 <talli> sure, will do
18:55:56 <donb> Spork - I'm paying attention keep firing away...
18:55:58 <Spork> He wanted to use OpenACS user management facilities and add functionality where a tech support representative can run a query on a cable modem (we have a tcl script that does it and I ported to some extent) 
18:56:34 <Spork> then the results of this SNMP query are returned to the tech support user and he can add a few words to comment what he saw...then all that is shoved into Oracle
18:57:04 <Spork> It's really not a lot...it took me a while to get up to speed on a few things and start doing it
18:57:30 <Spork> Not sure if I should be doing that as a subsite or not
18:57:40 <Spork> Because we're having a whole server dedicated to it
18:57:57 <Spork> like snmp.hostname.com should show the user a login screen 
18:58:08 <Spork> S/He can log in and start running the queries
18:58:15 <donb> I probably woudn't bother making it a subsite ...
18:58:25 <Spork> The more important catch was user management
18:58:29 <donb> but then again it's easy so why not?
18:58:35 <Spork> There are 3 user types
18:58:41 <Spork> regular, supervisor, manager
18:58:50 <Spork> supervisor can create regular users but not managers
18:58:59 <Spork> managers can create regular, supervisor, and manager
18:59:22 <Spork> if a user creates a subuser he can only edit his information
18:59:36 <Spork> so you can't have regular users editing manager info and adding managers
19:00:01 <Spork> And of course, when I was kinda up to the fun stuff my install blew up
19:00:45 <donb> So you're creating perms for the three roles as well as user groups to contain users in the three roles?
19:01:08 <Spork> i was creating a manager group and a supervisor group
19:01:15 <Spork> regular users were registered users
19:01:25 <donb> Right makes perfect sense ...
19:01:40 <Spork> Also, users can't register themselves, they had to be added (I had that done too)
19:02:03 <donb> Makes sense too ... wouldn't want people like me to register themselves ;)
19:02:05 <Spork> The only problem with my scheme ^^ was that users couldn't edit their created users..I was getting up to that
19:02:25 <Spork> I had a employee hierarchy table that I was planning on using
19:02:29 <Spork> never got that far...
19:02:43 <Spork> and my setup was kinda weird
19:02:50 <Spork> * Spork wonders if that's why it blew up
19:02:59 <Spork> I had my index page in /oacs/www
19:03:19 <Spork> and the rest of the files (mostly hacked admin pages for adding, removing, editing) in /oacs/packages/snmp
19:04:00 <donb> Well ... the log file should verify whether or not the query's running correctly.  I mean it's an awfully simple query that preceeds the error message you're getting.  It either sees your privilege or not ...
19:04:10 <Spork> right
19:04:18 <Spork> I verified that the permissions and groups are there
19:04:37 <Spork> I was about to run diff on the acs-permissions-procs from the new cvs from the ones I have from a week ago
19:05:22 <donb> So then why does that query return "0" as the count?  That's the magic question.  The log file will show the value being compared to (i.e. the value of the Tcl variable "permissions").  I'd just start with the log file myself ... make sure that query's there and if it is start scratching your head.
19:05:39 <Spork> ok
19:05:43 <Spork> hang on 
19:05:50 <Spork> let me create a log (I always run it in ft mode)
19:05:52 <donb> Perms seem to be working for standard packages on my system so they can't be entirely broken...
19:06:59 <donb> Dave - did you see my note about using the schema browser to list what tables have RI checks on another table?  To help track down who's referencing your CR item/revision when you delete it?
19:07:06 <Spork> ah..Don, when was the host_node_map added?
19:07:21 <davb> Don: yes, thanks. I crashed my server, so I have to wait until I get home...
19:07:44 <donb> Just recently ... oh, I see, you've created a subsite but didn't do the host_node_map hook, is that what you're about to tell me, Spork?
19:07:55 <Spork> perhaps
19:08:08 <Spork> :/
19:08:11 <Spork> is it fixable?
19:08:12 <Spork> lol
19:08:29 <donb> I don't think offhand it will matter I mean the perms error message is searching the entire database for your permission type ...
19:08:32 <Spork> is recently in the last 8 days?
19:09:12 <donb> Actually acs subsite will set it up when you mount the package and (again) it shouldn't matter regarding this particular problem.  But it was in the last week or two.
19:09:30 <talli> donb, davb or rbm: have you guys ever seen a discussion online about a comparison between Oracle and PG, if all things finnacial were equal?
19:09:59 <donb> The host_node_map is part of the code that lets you point different IPs to different subsites and keep the world straight.
19:10:07 <talli> i am interested in arguments for an against oracle and PG if oracle were free as well
19:10:10 <Spork> right
19:10:14 <Spork> I got that once I saw it
19:10:24 <Spork> I just don't understand why it broke my site completely
19:11:26 <donb> Talli, no I haven't but Oracle's more featureful, of course, and if you need some of its features, well, you need them.  But OACS doesn't...PG's easier to administer, and doesn't require 512MB RAM to install :)
19:11:47 <talli> yeah, that's a good point
19:12:00 <talli> but if money were no object, what would you prefer?
19:12:01 <Spork> hmmm
19:12:12 <Spork> I think that node_map is breaking it
19:13:08 <talli> anyway, i will pose this as a question to the bboards. i'm intersted in it as a marketing argument
19:13:08 <donb> So Talli ... my personal experience with Oracle is that it tends to drive me nuts so unless I were to need its extra features I would rather use PG.  It's just easier to live with, configure, etc.  I would pick Oracle if I were running a site doing massive amounts of inserts/updates because PG doesn't implement an overwriting storage manager, meaning you need to vacuum analyze frequently (this will change circa PG 7.3, though)
19:13:24 <talli> ok, sounds cool.
19:13:25 <donb> (actually still won't be overwriting but it will reclaim space in the background)
19:14:12 <talli> brb. need a latte
19:14:17 <donb> However websites don't do massive amounts of updates/inserts typically.  Everything Oracle does seems to take more RAM and frequently is a bit slower.  PL/SQL is nicer than PL/pgSQL, though.
19:15:07 <donb> Spork - well ... maybe so but how does it break "select count(*) from acs_permissions where permission = 'manager'"???  That's the query that appears to be breaking judging by the error message you're getting ...
19:15:07 <Spork> odd
19:15:20 <Spork> beats me
19:15:24 <donb> uh-huh thus the need for a log file...
19:15:57 <Spork> uh huh
19:16:02 <Spork> I think you guys have a bug
19:16:05 <Spork> I just hit reload
19:16:06 <Spork> and it worked
19:16:26 <donb> How about browser-side caching?
19:16:27 <Spork> mind you I made 0 modifications
19:16:37 <Spork> I hit reload
19:16:42 <donb> And it worked ... 
19:16:58 <Spork> i hit reload before :)
19:17:02 <Spork> multiple times
19:17:37 <Spork> i think the world just hates me :)
19:17:46 <donb> Hmmm...OK.  without the log file to see what query was being executed it's impossible to see what was going on...I always leave "verbose" on in my db pools when developing for this reason.
19:18:08 <Spork> I have verbose on
19:18:19 <Spork> ok, now let me ask you a question
19:18:22 <donb> Great ... oh but you weren't logging to a file???
19:18:27 <Spork> right
19:18:33 <Spork> I run the server in -ft 
19:18:35 <Spork> not -it
19:18:46 <Spork> since it's not on all the time
19:18:49 <Spork> just when I dev
19:18:51 <donb> Damn ... would've been nice to go back through the log to try to see what was going wrong...
19:19:02 <donb> Anyway ... on to your question.
19:19:06 <Spork> ok
19:19:19 <Spork> I have a few hacked admin files for adding removing users
19:19:28 <Spork> I've been keeping them in /packages.snmp
19:19:58 <Spork> now since the new OACS cvs has subsite things is it better for me to just keep it in a subdir under /www
19:20:04 <Spork> instead of /packages/snmp
19:20:28 <Spork> as well as my files
19:20:32 <Spork> for querying the modem
19:21:21 <donb> Physical directory?  I prefer keeping as much as possible in the packages subtree and let the request processor map to wherever I mount it.  As far as mounting things I just mount under "/" so users have short URLs.
19:21:32 <donb> (I've never written a site using subsites)
19:21:40 <Spork> heh ok
19:21:53 <Spork> is it going to be a problem with spanning subsites though?
19:21:57 <donb> (though I'll be helping a bit with dotLRN and subsites will be one of the main structuring tools there, with each class having its own subsite).
19:21:59 <Spork> because the main site is a subsite
19:22:36 <Spork> and I'll wind up having a snmp subsite as well (which apparently now comes with it's own set of relation/group management tools)
19:22:51 <Spork> aha
19:22:55 <Spork> I hit reload again
19:22:59 <Spork> it now wants me to log in
19:23:15 <donb> Strange ...
19:23:31 <Spork> indeed
19:23:33 <donb> What browser are you using?
19:24:16 <Spork> IE
19:24:45 <Spork> very odd
19:24:47 <donb> Hmmm...usually works well enough ...
19:25:17 <donb> There's nothing weird between the two machines, I assume (squid, etc)?
19:25:52 <Spork> nope
19:26:39 <donb> I haven't seen anything weird with login cookies, most of that code's eons old...
19:27:20 <donb> Anyway ... subsites etc ... you have more stuff going on than the snmp stuff, then, and are wondering how to structure the site?
19:27:47 <Spork> yeah pretty much
19:27:54 <Spork> site navigation is very simple too
19:27:57 <Spork> you go to the main site
19:28:03 <Spork> you log in
19:28:11 <Spork> you're taken to your custom menu (depending on the level)
19:28:25 <Spork> From there, you can run a query or get a history from db
19:28:37 <Spork> if you're manager or supervisor, you can add, delete, edit users
19:28:51 <Spork> the last thing is already done with hacked ACS files
19:29:01 <Spork> like I have no need for "ban user"
19:29:20 <Spork> So I just edited the proc to build that so it only shows delete, undelete
19:30:03 <Spork> I'm not sure if it's best to separate that or just shove it into www
19:30:13 <Spork> The www thing would be the simplest way probably
19:30:33 <donb> Now ... if these managers and supervisors deal with user management for the site at large this function shouldn't be part of the snmp package.  Normally when I've hacked ACS-based sites I've just whacked the files in-situ, but that's in part historical (aD was changing stuff so fast and with no regard to backward compatibility that adopting updates was almost impossible anyway).  
19:31:00 <donb> That's not a negative statement, BTW ... the toolkit was very primitive in its beginning and aD needed to whack at it daily to meet customer needs.
19:31:25 <Spork> well, optimally, I can always just hack the files for the "final" revision before all the Q/A is done
19:32:03 <donb> Right these files probably haven't changed all that much.  Besides I don't expect much change in this area in the next two-three months.
19:32:21 <Spork> ok
19:32:48 <Spork> well, another reason I haven't done that is because on cvs things I always grab the whole tree not just update it
19:33:01 <Spork> So I'll always have to remember what files got hacked
19:33:31 <donb> Anyway ... I'd separate out user management from the snmp package.  Are all users on the site users of the snmp site as well, or are just some of them?  I ask because there's the concept of "joining" a subsite which can be used with permissions to limit who can access a given subsite.
19:33:45 <Spork> all users
19:34:03 <donb> OK that's really simple ...
19:34:04 <Spork> everyone uses the snmp queries
19:34:13 <Spork> We just want to limit who can do user management
19:35:09 <donb> I don't think you really need to use the subsite concept at all, frankly.  The main reason to do it is to provide multiple sites under one roof, so to speak - each with their own look-and-feel, for instance.
19:36:17 <Spork> right
19:36:24 <Spork> so that's what I was thinking
19:36:31 <Spork> I don't even need to create a package
19:36:35 <donb> Own look and feel, own IPs, own separate sets of users - say implementing a gigantic snmp site that several cable companies could use, all on one server, but each "pretending" to belong to each separate company.
19:36:45 <Spork> right
19:37:15 <donb> No, you don't but packages are easy to build and deploy.  Who knows, you guys might find use for it elsewhere...
19:38:05 <Spork> right
19:38:15 <Spork> I can always create a package from my completed stuff
19:38:28 <Spork> It just gets mounted in / 
19:38:42 <Spork> thank you very much Don!
19:38:43 <Spork> :)
19:38:48 <donb> I don't think we'll solve your mystery (the failure) without log files, but feel free to e-mail me if it crops up again or if you have other problems.   I'm going to run off ...  yeah, packaging afterwards is easy enough ... right mount in / ignore subsiting is what I'd do.  ANyway, good luck!
19:39:11 <donb> Talli et al talk to you later!
19:39:14 <donb> donb has quit
19:39:15 <talli> later donv
19:39:27 <Spork> he's helpful :)
19:42:14 <Spork> I think I'll get fired anyway though
19:44:39 <davb> no way. 
19:45:39 <Spork> ?
19:45:42 <Spork> no way what?
19:47:02 <davb>  get fired 
19:47:32 <Spork> yeah right
19:54:04 <Spork> ok, so how would I create a list of stuff from an array
19:56:13 <Spork> so an array is like a record in java/c
19:56:16 <Spork> ack
19:56:18 <Spork> in c/modula
19:57:30 <davb> set list_var [array get $arrayname] I think
19:57:39 <davb> which creates a list of lists
19:58:12 <Spork> k
19:58:29 <Spork> i still don't get this array thing
19:59:19 <davb> wow, the web sucks today
19:59:24 <Spork> lol
19:59:33 <davb> or its roadrunner
20:00:07 <davb> "Tcl arrays are actually hash tables and have nothing in common with the data structures called arrays in other programming languages . A Tcl array provides a rapid answer to the question "is there a value associated with this key". Here is a rat-simple example: "
20:00:37 <davb> think like that and everything makes more sense.
20:01:04 <Spork> set table(count) 0
20:01:07 <Spork> i got that far
20:01:24 <Spork> that makes an array table and sets index count to 0
20:01:37 <Spork> set table(header) { ifIndex ifDescr }
20:01:50 <Spork> and this adds a header key with a list as a value 
20:01:55 <davb> yes.
20:02:09 <davb> you are decoding someone else's Tcl?
20:02:15 <Spork> yah
20:02:23 <davb> Did you see that someone loaded Scotty into AOLserver?
20:02:34 <Spork> where?
20:02:42 <Spork> was it Vlad?
20:03:08 <davb> not sure. it was on the AOLserver mailing list and I can't check my email right now.
20:03:37 <Spork> was it this  http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0002rJ&topic_id=11&topic=OpenACS
20:04:49 <ola> http://www.pinds.com/acs-tips/tcl-data-structures
20:04:50 <chump> D: http://www.pinds.com/acs-tips/tcl-data-structures from ola
20:04:57 <ola> D:|Datastructures in Tcl
20:04:57 <chump> titled item D
20:07:34 <talli> D: lars is smart
20:07:34 <chump> commented item D
20:07:43 <Spork> heh
20:10:56 <davb> this is it: http://listserv.aol.com/cgi-bin/wa?A2=ind0110&L=aolserver&P=R4915&X=3E49AF28ABEB09B672&Y=dave@deepskydesign.com
20:11:06 <davb> the link might not work though..
20:11:50 <davb> First of all I suppose checking whether NS entries for such a domain
20:11:52 <davb> exist is a good option. I'm going to test whether loading scotty (using
20:11:52 <davb> Tcl 'load' in ns_thread begindetached :) is possible. Maybe it'll help
20:11:52 <davb> me.
20:12:01 <davb> oops. thats a quote from another message from this guy
20:12:40 <rbm> hello there
20:12:43 <davb> Talli: I see you are hosting it now 
20:13:02 <Spork> oh
20:13:09 <Spork> I think he did something else
20:13:16 <Spork> I'm using a diff snmp thing :0
20:15:31 <davb> hi rbm
20:16:27 <rbm> hey davb
20:17:24 <talli> yeah, we're hosting pinds.com
20:17:38 <talli> lars got the boot and he asked for some room
20:18:21 <rbm> cool.
20:18:29 <rbm> Does he still want to work on ACS?
20:18:42 <talli> i hope so :)
20:18:48 <talli> he may work on some ETP stuff
20:18:48 <rbm> I was hoping he would like to continue helping the project
20:18:51 <rbm> Nice!
20:19:03 <talli> but he has designs on a new project and starting a company for himself
20:19:11 <rbm> oh
20:19:54 <davb> <talli> he may work on some ETP stuff: Cool!
20:21:54 <talli> yeah
20:22:08 <talli> davb you can access your email?
20:22:32 <Spork> nope hehe
20:22:36 <Spork> his machine didn't boot
20:23:42 <davb> I will in an hour or so.
20:23:54 <davb> I have to leave in about 10 mins.
20:24:08 <talli> luke sent a long email about how to implement photo functionality in ETP
20:24:20 <talli> lars said he had some notes about improving the usability of ETP
20:25:30 <davb> great!
20:25:35 <davb> on all that stuff.
20:29:47 <rbm> no kidding
20:34:07 <davb> bbl
20:34:10 <davb> davb has quit
20:35:31 <Spork> Does TCL start indexing from 0 or 1?
20:35:35 <Spork> my guess is 0
20:37:43 <talli> rbm: in debian, how can i find a package to download?
20:38:05 <Spork> I think there's a search engine on the debian site
20:38:07 <talli> IOW, if i want to download the source for the 2.4.12 kernel, how do i find the package name?
20:38:25 <Spork> search for 2.4.12
20:38:26 <Spork> :)
20:38:33 <talli> but is there a way to do it using apt-get?
20:38:59 <talli> smart guy that spork
20:39:29 <Spork> see if apt-get has a keyword arg
20:40:15 <rbm> talli: what do you mean? wanna search for a package?
20:40:20 <rbm> apt-cache search foobar
20:40:48 <rbm> to see packages you have installed" dpkg -l
20:41:00 <rbm> to see files in a package: dkpg -L package
20:41:08 <rbm> s/dkpg/dpkg/
20:41:21 <rbm> apt-cache is very useful btw
20:41:52 <rbm> another cool thing is "deborphan". It'll show you packages you have installed that have nothing depending on them.
20:44:30 <talli> what if i want to know what the name of the package is on the debian server
20:45:12 <Spork> * Spork loves ports
20:49:20 <Spork> ugh
20:50:03 <Spork> 603000000 / 1000000 should be 603.0
20:50:09 <Spork> right?
20:53:59 <rbm> talli: apt-cache show package
20:54:03 <talli> oh, cool
20:54:05 <talli> thanks
20:54:08 <rbm> no
20:54:49 <Spork> set modem_us_freq [format %.2f [expr [lindex $cmdata 4] / 1000000.0]] 
20:54:53 <Spork> how does this not divide?
20:56:35 <rbm> do you get an error?
20:56:39 <Spork> no
20:56:42 <Spork> it just does not divide
20:56:47 <rbm> What's on lindex $cmdata 4?
20:57:04 <Spork> value before
20:57:07 <Spork> 603000000
20:57:07 <rbm> What's on lindex $cmdata 4?
20:57:17 <rbm> let me see
20:57:24 <Spork> errr
20:58:08 <rbm> % format %.2f [expr 603000000 / 1000000.0]
20:58:09 <rbm> 603.00
20:58:13 <rbm> It divides for me
20:58:41 <Spork> yeah I see the problem I have
20:58:45 <Spork> odd...very very odd
21:00:00 <ola> ola has quit
21:27:49 <docwolf> docwolf has quit
21:32:23 <Spork> anyone good with ns_fmttime and ns_time
21:44:39 <Spork> anyone still awake?
22:22:10 <talli> talli has quit
23:54:27 <docwolf> docwolf has joined #openacs
23:56:41 <rbm> 'sup doc?
23:56:50 <docwolf> howdy