| 00:27.20 | macnewbold | does anyone get my posts to the mailing list? |
| 00:27.40 | macnewbold | like a reply to the GPL question? |
| 00:29.32 | macnewbold | my "receive own messages" option is set on the list too, I just checked |
| 01:28.04 | tierra | macnewbold: I don't see your reply :( |
| 01:28.50 | tierra | last email I see from you on the list was Aug 17 |
| 03:03.47 | *** join/#uphpu bigdog_ut (n=bigdog@zeus.thebigdog.biz) |
| 03:03.47 | *** mode/#UPHPU [+o bigdog_ut] by ChanServ |
| 04:13.29 | *** join/#uphpu burnin (n=burnin@42dbe40a.dsl.aros.net) |
| 07:41.07 | *** join/#uphpu cycokilr (n=cycokilr@67-40-116-40.slkc.qwest.net) |
| 13:06.02 | *** join/#uphpu RyanE (i=Piadas@rberick.dsl.xmission.com) |
| 13:07.01 | *** join/#uphpu drip (n=drip@67.96.135.254) |
| 13:37.48 | *** join/#uphpu RyanE_ (i=Piadas@rberick.dsl.xmission.com) |
| 14:10.18 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 14:10.18 | *** mode/#UPHPU [+o wps] by ChanServ |
| 14:30.13 | *** join/#uphpu _psychic_ (n=_psychic@71.32.228.156) |
| 14:30.42 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 14:30.42 | *** mode/#UPHPU [+o wps] by ChanServ |
| 14:51.18 | *** join/#uphpu bigdog_ut (n=bigdog@208.177.146.102.ptr.us.xo.net) |
| 14:51.18 | *** mode/#UPHPU [+o bigdog_ut] by ChanServ |
| 14:51.38 | drip | Arrr.. the bigdog_ut be boardin' the ship |
| 14:52.09 | bigdog_ut | who is the elubber? |
| 14:52.30 | drip | ye be the lubber. |
| 14:52.42 | drip | since mindjuju still isn't here. |
| 14:53.18 | drip | Avast! It be talk like a pirate day! |
| 15:01.28 | *** join/#uphpu Utah_Dave (n=boucha@72-166-51-162.dia.static.qwest.net) |
| 15:27.54 | *** join/#uphpu herlo (n=herlo@hq-nat2.gurulabs.com) |
| 16:47.48 | *** join/#uphpu mindjuju (n=Administ@65-121-75-195.dia.static.qwest.net) |
| 16:55.14 | mindjuju | ~rlart |
| 16:55.40 | bigdog_ut | welcome back...have you forgiven us? |
| 16:55.51 | mindjuju | forgiven? |
| 16:56.05 | bigdog_ut | you have not been around for a while |
| 16:56.18 | mindjuju | ah |
| 16:56.25 | mindjuju | mom was in town |
| 16:56.33 | mindjuju | i was doing the vpn to office |
| 16:56.35 | bigdog_ut | oh so no irc when mom is home |
| 16:56.46 | drip | he was grounded from irc |
| 16:56.53 | bigdog_ut | lol |
| 16:57.05 | bigdog_ut | too much irc i guess and mom was not happy about it |
| 16:57.09 | mindjuju | pretty much |
| 17:00.22 | bigdog_ut | welcome back mindjuju nice to see you here |
| 17:00.42 | mindjuju | thx! it's good to be back. it's boring not being on IRC |
| 17:01.12 | mindjuju | it was pretty much just log in to my box, do work, log off, spend time with fam, etc |
| 17:02.38 | mindjuju | i bet it was nice and quiet in here while i was gone |
| 17:02.44 | mindjuju | not so many larts |
| 17:03.32 | drip | lol.. pretty much |
| 17:05.55 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 17:05.55 | *** mode/#UPHPU [+o wps] by ChanServ |
| 17:06.33 | *** join/#uphpu beandog (n=sdibb@gentoo/developer/beandog) |
| 17:08.59 | synic | in other news.... python 2.5 is totally out today. |
| 17:09.12 | synic | luckily I downloaded it early... now you can hardly even get to their site |
| 17:09.13 | bigdog_ut | time to upgrade |
| 17:13.55 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 17:13.55 | *** mode/#UPHPU [+o wps] by ChanServ |
| 17:15.27 | mindjuju | anybody know any prebuilt php code that is for a garage sale? |
| 17:19.40 | bigdog_ut | nope...i have looked a couple years back and did not find any |
| 17:19.54 | macnewbold | an online garage sale? |
| 17:21.11 | bigdog_ut | actually it is not a bad idea...i thought about doing one where anyone could post their garage sale |
| 17:21.19 | macnewbold | wps: does the mailing list automatically and silently reject posts from non-members? |
| 17:44.52 | wps | mindjuju: craigslist :) |
| 17:54.21 | wps | macnewbold: it was |
| 17:54.29 | wps | I changed it from discard to reject |
| 17:54.31 | macnewbold | ah, okay |
| 17:54.33 | wps | the user now gets a notice |
| 17:54.37 | macnewbold | that would be good |
| 17:54.47 | wps | yup |
| 17:54.48 | wps | thanks |
| 17:54.49 | macnewbold | I sent one from my code greene address without knowing it, and it just vanished |
| 18:04.41 | *** join/#uphpu fungus (n=fungus@shell.aros.net) |
| 18:49.24 | *** join/#uphpu timothy_ (n=timothy@c-67-171-118-199.hsd1.ut.comcast.net) |
| 18:49.35 | timothy_ | hey guys |
| 18:50.40 | mindjuju | yo |
| 18:51.02 | timothy_ | is there a way to run a backtick command and then wait a few seconds for a response? |
| 18:52.12 | timothy_ | the program is something of a file converter, and echos "operation complete" when finished |
| 18:52.47 | timothy_ | but I don't think PHP is waiting for it to finish |
| 18:53.01 | timothy_ | because I get a blank converted file in the end |
| 18:53.36 | timothy_ | rather: a converted file that is blank |
| 18:54.10 | timothy_ | any ideas? |
| 18:56.08 | _psychic_ | unless you're forking off the execution of the converter, PHP should wait till its done to callback |
| 18:56.15 | _psychic_ | afaik :/ |
| 18:57.08 | mindjuju | sorry, stepped away for a moment. that was my experience with imagemagick from the commandline |
| 18:57.54 | timothy_ | how would I fork off the execution? |
| 19:02.04 | _psychic_ | http://us3.php.net/pcntl |
| 19:02.22 | _psychic_ | or starting some code that forked and returned I guess |
| 19:03.32 | timothy_ | **looking into it** |
| 19:04.21 | timothy_ | if the program is forking and then returning, what should I do to make PHP recognize that? |
| 19:04.39 | _psychic_ | dunno |
| 19:19.20 | beandog | oh man |
| 19:19.25 | beandog | I looked into that stuff not too long ago. |
| 19:19.27 | beandog | talk about a beast. |
| 19:19.37 | beandog | good luck. :) |
| 19:20.24 | timothy_ | thanks |
| 19:20.55 | timothy_ | so, what did you end up doing? |
| 19:25.07 | beandog | somethign completely different. |
| 19:25.20 | timothy_ | that bad, huh? |
| 19:25.23 | beandog | what are you trying to do exactly? |
| 19:25.44 | timothy_ | run pcmiler |
| 19:25.46 | timothy_ | batch pro |
| 19:26.05 | beandog | timothy_ the program is something of a file converter, and echos "operation complete" when finished |
| 19:26.09 | beandog | Okay, I was doing something similar |
| 19:26.19 | beandog | Are you making sure youre reading from stdout and not stderr? |
| 19:26.31 | beandog | Like, what command are you using to read "operation complete" |
| 19:26.43 | timothy_ | I've tried a backtick popen and exec |
| 19:26.45 | beandog | Or you could just do $foo = `whoami`; or whatever, and |
| 19:26.51 | timothy_ | yes |
| 19:26.52 | beandog | that should return the same string |
| 19:27.04 | timothy_ | I don't care about it returning anything |
| 19:27.20 | beandog | You just want to keep going once its finished executing |
| 19:27.21 | beandog | ? |
| 19:27.53 | timothy_ | when it executes, it takes one file and makes another file |
| 19:28.00 | beandog | ok |
| 19:28.17 | timothy_ | so I don't care about the output |
| 19:28.30 | timothy_ | cause I'll just open and read the file |
| 19:28.39 | timothy_ | I just need to send the command |
| 19:28.45 | timothy_ | to the command line |
| 19:28.53 | beandog | well you could fork it yourself |
| 19:29.02 | beandog | $foo = `whoami &`; |
| 19:29.07 | beandog | or just `whoami &`; |
| 19:29.33 | beandog | Either way, just using exec() or system() should work |
| 19:29.34 | timothy_ | and that won't cut the operation short? |
| 19:29.43 | beandog | no, it'll just send it in the background |
| 19:29.54 | timothy_ | **trying |
| 19:29.54 | beandog | your script will tell it to run and keep going |
| 19:30.19 | beandog | But if your script depends on the outcome of what that converter does, then you'd have to wait for it to finish. |
| 19:30.23 | beandog | In that case you can't fork it. |
| 19:30.35 | timothy_ | it doesn't |
| 19:30.37 | beandog | Also, is this a shell script or are you running it from a webserver? |
| 19:30.45 | timothy_ | webserver |
| 19:30.56 | beandog | Well the webserver is going to have a *completely* different PATH setting as well as user permissions. |
| 19:31.01 | beandog | So it might not even run it. |
| 19:31.10 | beandog | Try running echo `pwd`; for instance. |
| 19:31.16 | beandog | or `echo $PATH`; |
| 19:31.18 | beandog | you wont get much. |
| 19:31.57 | timothy_ | I believe it is running because: |
| 19:32.12 | timothy_ | the initial file is route1.in |
| 19:32.23 | beandog | also did you try passthru() |
| 19:32.23 | timothy_ | and the converted file is route1.out |
| 19:32.33 | beandog | that'll at least show you what the file is dumping to stdout |
| 19:32.49 | timothy_ | when I run the command from the command line, it creates the .out file and fills it w/ data |
| 19:33.06 | beandog | are you running in safe mode? |
| 19:33.09 | timothy_ | but from the webserver, it creates the .out file but leaves it empty |
| 19:33.34 | beandog | well how long does it usually take to run |
| 19:34.13 | timothy_ | safe_mode=off |
| 19:34.21 | timothy_ | 1 sec |
| 19:34.29 | beandog | I dunno |
| 19:34.31 | beandog | my first guess is permissions |
| 19:34.40 | beandog | what kind of OS are you using |
| 19:34.46 | timothy_ | but it does load PCMiler to do the conversion and comes back to PCBatchPro to write the .out file |
| 19:34.54 | timothy_ | Gentoo |
| 19:34.56 | beandog | I'd try dumping to a temporary directory where anyone can read/write |
| 19:34.57 | beandog | Ah, Gentoo :D |
| 19:35.13 | timothy_ | :) Thank you BigDog |
| 19:35.48 | timothy_ | (bigdog_ut hooked me up) |
| 19:35.55 | beandog | bigdog_ut: good man |
| 19:35.59 | timothy_ | yes |
| 19:36.07 | timothy_ | but I still owe him my firstborn |
| 19:36.10 | timothy_ | eek |
| 19:36.41 | beandog | what is pcmiler anyway? |
| 19:36.42 | bigdog_ut | and dinner |
| 19:36.47 | timothy_ | and dinner |
| 19:37.35 | timothy_ | pcmiler is a mileage program |
| 19:37.50 | timothy_ | city, state to city, state |
| 19:37.54 | timothy_ | in the .in file |
| 19:37.57 | timothy_ | the .out file |
| 19:38.04 | timothy_ | shows the mileage |
| 19:38.08 | macnewbold | did you get your forking problem solved? |
| 19:38.11 | timothy_ | and the time |
| 19:38.13 | timothy_ | no |
| 19:38.19 | beandog | Another thing, does your webserver have permissions to read that file? |
| 19:38.47 | timothy_ | it should |
| 19:38.52 | beandog | run stat on it |
| 19:38.55 | beandog | see what the mode is |
| 19:38.56 | timothy_ | how would I check? |
| 19:40.33 | beandog | stat foo.txt |
| 19:40.33 | macnewbold | if the program you're calling is forking and backgrounding itself, then you can look for an option to it that says not to do that, |
| 19:40.33 | beandog | Access: (0644/-rw-r--r--) |
| 19:40.33 | macnewbold | or you could watch it's output file |
| 19:40.33 | macnewbold | until it is the size you expect or something |
| 19:40.50 | beandog | that's assuming that it's executing correctly at all. |
| 19:40.50 | macnewbold | how do you usually know that the program is done running? |
| 19:41.02 | beandog | well he said it only takes a short second to run |
| 19:41.23 | timothy_ | it echos "PCBatch operation complete" and goes back to a prompt |
| 19:41.31 | timothy_ | ya, and it's quick |
| 19:41.33 | beandog | with passthru? |
| 19:41.39 | timothy_ | haven't tried |
| 19:41.41 | macnewbold | is that before or after it has generated all the output? |
| 19:41.42 | beandog | Hmm |
| 19:41.48 | beandog | Can you pastebin your code, timothy_ ? |
| 19:41.52 | beandog | or at least the relevant parts |
| 19:42.01 | timothy_ | y, 1 min |
| 19:42.22 | beandog | also, what version of php are you using? |
| 19:42.30 | macnewbold | if I understand right, the output file when finished will have the same number of lines as the input file, right? |
| 19:42.46 | macnewbold | so you could watch the output file, checking it every second or two with wc to see when it is all done, |
| 19:42.48 | beandog | I think his output file is just blank |
| 19:42.52 | macnewbold | then open it up and get your results, right? |
| 19:42.52 | beandog | but it exists. |
| 19:43.00 | timothy_ | mac: yes, same # of lines |
| 19:43.19 | timothy_ | but the output file has been blank |
| 19:43.25 | timothy_ | from the webserver |
| 19:43.33 | macnewbold | and it never grows? |
| 19:43.57 | macnewbold | now the web server probably runs as user www in group www, does that user/group have access to the output file? |
| 19:44.00 | macnewbold | (write access) |
| 19:44.09 | beandog | thats what I was asking |
| 19:44.17 | macnewbold | does this app log anything anywhere? like if it complains about something? |
| 19:44.20 | beandog | and to test writing to /tmp/foo.out or something |
| 19:44.25 | beandog | instead of ./ |
| 19:44.41 | timothy_ | mac: it never grows |
| 19:44.41 | macnewbold | sure |
| 19:44.52 | macnewbold | who owns that file? |
| 19:44.59 | macnewbold | what are the permissions on it? |
| 19:44.59 | beandog | ok, post some code, that'll help. :) |
| 19:45.05 | timothy_ | I don't know about write access... looking |
| 19:45.07 | macnewbold | like beandog said, stat will tell you, or ls -l will too |
| 19:45.24 | timothy_ | http://pastebin.com/790149 |
| 19:45.46 | beandog | man I just pulled a nasty hairball from my keyboard |
| 19:47.54 | wps | thanks for sharing |
| 19:47.58 | beandog | heh |
| 19:48.04 | beandog | anything for you, wps :) |
| 19:48.24 | bigdog_ut | i just spit up a nasty hairball |
| 19:49.06 | timothy_ | access on that file is 644 |
| 19:49.09 | beandog | Ok |
| 19:49.14 | beandog | first of all |
| 19:49.18 | timothy_ | user and group is root |
| 19:49.19 | beandog | I would do /usr/bin/pcmbatch |
| 19:49.27 | beandog | on which file |
| 19:49.42 | beandog | then, do escapeshellcmd() on the variable |
| 19:49.48 | beandog | then just run passthru() |
| 19:49.49 | timothy_ | on route1.out, but that's how each file is getting setup |
| 19:49.50 | beandog | and don't fork it |
| 19:50.06 | timothy_ | trying passthru w/o a fork |
| 19:50.18 | beandog | so $cmd = '/usr/bin/pcmbatch $args $route $etc'; |
| 19:50.27 | beandog | $cmd = excapeshellcmd($cmd); |
| 19:50.33 | beandog | passthru($cmd); |
| 19:50.44 | beandog | actually |
| 19:50.47 | beandog | yah. |
| 19:51.17 | beandog | If the directory you are running that from though is owned by root:root |
| 19:51.22 | beandog | then the webserver can't create files in there. |
| 19:51.26 | beandog | It needs to be owned by apache:apache |
| 19:51.33 | timothy_ | passthru w/o a fork gave a blank file |
| 19:51.58 | timothy_ | but it's creating the file |
| 19:52.02 | timothy_ | just not writing to it |
| 19:52.10 | timothy_ | how should I change that? |
| 19:52.21 | beandog | I dunno |
| 19:52.23 | beandog | macnewbold: what you think. |
| 19:52.32 | beandog | doesnt make sense to me. |
| 19:52.39 | beandog | the program is just plain having issues, sounds like to me. |
| 19:52.45 | beandog | Try running it as a normal user |
| 19:52.47 | beandog | does it work? |
| 19:53.54 | beandog | And can you tell it to dump the file somewhere else instead? |
| 19:54.36 | timothy_ | as a normal user? |
| 19:54.54 | timothy_ | if the same command is executed from the command line, it works |
| 19:55.05 | timothy_ | so, is that what you're asking for? |
| 19:55.07 | beandog | as what user? |
| 19:55.20 | timothy_ | sudo root |
| 19:55.28 | beandog | Well try running it as user apache |
| 19:55.30 | beandog | and see what it does. |
| 19:55.38 | timothy_ | 1 min |
| 19:55.40 | beandog | su - apache -c "$cmd" |
| 20:01.00 | *** join/#uphpu _psychic_ (n=_psychic@71.32.228.156) |
| 20:02.05 | timothy_ | it returned this "su: Authentication service cannot retrieve authentication info. |
| 20:02.06 | timothy_ | (Ignored) |
| 20:02.06 | timothy_ | No directory, logging in with HOME=/" |
| 20:03.23 | timothy_ | when I executed "su - apache -c pcmbatch P /var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route6.in 1 -t" |
| 20:03.52 | timothy_ | so it probably is an authentication thing, huh? |
| 20:04.13 | beandog | well |
| 20:04.14 | beandog | do -c "stuff" |
| 20:04.16 | beandog | put it all in quotes. |
| 20:04.22 | beandog | I think you need ot. |
| 20:04.22 | beandog | to |
| 20:05.06 | timothy_ | same result |
| 20:05.36 | beandog | Hmm. |
| 20:05.38 | beandog | -_- |
| 20:05.40 | beandog | dunno. |
| 20:05.46 | timothy_ | command was: su - apache -c pcmbatch P "/var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route6.in 1 -t" |
| 20:06.10 | beandog | no no |
| 20:06.13 | beandog | -c "all the commands here" |
| 20:06.26 | beandog | either way, shouldnt be too different. |
| 20:06.34 | beandog | Ok do this: |
| 20:06.38 | timothy_ | oh, so omit the su - apache? |
| 20:06.52 | beandog | stat /var/www/vhosts/html/apps/mylumbertradernet/pcmiler |
| 20:06.53 | beandog | no |
| 20:07.01 | beandog | su - apache -c "pcmbatch P /var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route6.in 1 -t" |
| 20:07.23 | timothy_ | stat: |
| 20:07.27 | timothy_ | <PROTECTED> |
| 20:07.27 | timothy_ | <PROTECTED> |
| 20:07.27 | timothy_ | Device: 6200h/25088d Inode: 772717 Links: 2 |
| 20:07.27 | timothy_ | Access: (0777/drwxrwxrwx) Uid: ( 1001/ timothy) Gid: ( 81/ apache) |
| 20:07.28 | timothy_ | Access: 2006-09-19 15:52:18.000000000 -0400 |
| 20:07.28 | timothy_ | Modify: 2006-09-19 15:53:23.000000000 -0400 |
| 20:07.29 | timothy_ | Change: 2006-09-19 15:53:23.000000000 -0400 |
| 20:08.11 | beandog | Ok |
| 20:08.13 | beandog | So anyone can r/w there. |
| 20:08.19 | beandog | Is that the same directory its dumping the files to? |
| 20:08.30 | timothy_ | yes, yes |
| 20:08.38 | beandog | crap. |
| 20:08.41 | beandog | I dunno. |
| 20:08.45 | beandog | Youre doing everything right, near as I can tell. |
| 20:08.57 | macnewbold | are you using suphp or suexec, timothy_? |
| 20:09.12 | timothy_ | ? |
| 20:09.15 | macnewbold | you could always try making the dir and file world read and write? |
| 20:09.31 | timothy_ | right now I'm using passthru |
| 20:09.54 | timothy_ | 777? |
| 20:10.00 | macnewbold | does your web server run your php scripts as www/www, apache/apache, root/root, or your username and group? |
| 20:10.14 | timothy_ | I don't know |
| 20:10.18 | beandog | its apache:apache |
| 20:10.31 | beandog | wait |
| 20:10.32 | beandog | Im not sure. |
| 20:10.32 | timothy_ | I think it's apache apache |
| 20:10.33 | beandog | brb. |
| 20:11.35 | timothy_ | mac: how would I find out? |
| 20:11.50 | beandog | just exec('whoami'); |
| 20:11.52 | beandog | in apache. |
| 20:12.05 | timothy_ | I don't know apache's password |
| 20:12.20 | beandog | They dont have a shell anyway |
| 20:12.22 | beandog | Just create a script |
| 20:12.29 | beandog | <? passthru('whoami'); ?> |
| 20:13.14 | timothy_ | apache |
| 20:14.18 | timothy_ | passthru('whoami') == apache |
| 20:14.40 | macnewbold | okay |
| 20:14.43 | macnewbold | well that answers that |
| 20:15.14 | macnewbold | yeah, if you chmod the directory to 777 (ugo+rwx) and the files to 666 (ugo+rw) or 777, you should be able to do it |
| 20:15.39 | macnewbold | it sounds like something is failing along the way, but you're not seeing the error message anywhere |
| 20:15.50 | timothy_ | but the command that creates and writes to the file sets the permissions to 644 |
| 20:16.01 | macnewbold | if there are any options for verbose, debugging, logging, or for not backgrounding, that might help |
| 20:16.02 | timothy_ | no errors |
| 20:16.09 | macnewbold | okay, well if it is creating it, it should be okay |
| 20:16.16 | timothy_ | I agree |
| 20:16.32 | beandog | oh yah, duh |
| 20:16.34 | beandog | stat the file too |
| 20:17.39 | timothy_ | stat /var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route1.outFile: `/var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route1.out' |
| 20:17.39 | timothy_ | <PROTECTED> |
| 20:17.39 | timothy_ | Device: 6200h/25088d Inode: 772727 Links: 1 |
| 20:17.39 | timothy_ | Access: (0644/-rw-r--r--) Uid: ( 0/ root) Gid: ( 0/ root) |
| 20:17.39 | timothy_ | Access: 2006-09-19 14:18:00.000000000 -0400 |
| 20:17.39 | timothy_ | Modify: 2006-09-19 14:17:51.000000000 -0400 |
| 20:17.40 | timothy_ | Change: 2006-09-19 14:17:51.000000000 -0400 |
| 20:18.11 | beandog | yah see |
| 20:18.14 | beandog | thats owned by root |
| 20:18.19 | beandog | gotta chown apache: on that file |
| 20:18.39 | bigdog_ut | timothy_, why is it owned by root? |
| 20:18.52 | timothy_ | how would I do that in the middle of the passthru? |
| 20:19.58 | beandog | well, delete it as root |
| 20:20.00 | beandog | then try it with apache |
| 20:22.59 | timothy_ | that's what I'm doing.... |
| 20:23.29 | timothy_ | when PHP runs the passthru command, it creates and writes to the file |
| 20:23.36 | beandog | right |
| 20:23.40 | timothy_ | one step |
| 20:23.45 | beandog | but it cant create it as root. |
| 20:23.48 | beandog | not if its running as user apache |
| 20:23.56 | beandog | so delete the file that was created before, owned by root |
| 20:23.58 | bigdog_ut | the file might already be created |
| 20:23.59 | beandog | and then try it through apache |
| 20:24.23 | timothy_ | k |
| 20:25.17 | timothy_ | my mistake, that was the command line test |
| 20:25.24 | timothy_ | here's a PHP test: |
| 20:25.27 | timothy_ | <PROTECTED> |
| 20:25.27 | timothy_ | <PROTECTED> |
| 20:25.28 | timothy_ | Device: 6200h/25088d Inode: 772730 Links: 1 |
| 20:25.28 | timothy_ | Access: (0644/-rw-r--r--) Uid: ( 81/ apache) Gid: ( 81/ apache) |
| 20:25.28 | timothy_ | Access: 2006-09-19 14:20:38.000000000 -0400 |
| 20:25.28 | timothy_ | Modify: 2006-09-19 14:20:38.000000000 -0400 |
| 20:25.30 | timothy_ | Change: 2006-09-19 14:20:38.000000000 -0400 |
| 20:25.54 | beandog | Ok |
| 20:26.01 | beandog | so it created the file at least. |
| 20:26.05 | timothy_ | sorry |
| 20:26.07 | timothy_ | yes |
| 20:26.13 | beandog | but its still empty. |
| 20:26.18 | timothy_ | yes |
| 20:26.20 | beandog | Does that program even *run* as a non-root user? |
| 20:26.32 | beandog | er, as => with |
| 20:26.43 | timothy_ | it has to, doesn't it? to create the file |
| 20:26.49 | beandog | well |
| 20:26.55 | beandog | somewhat. :) |
| 20:26.58 | beandog | I would say it's not working. |
| 20:27.17 | bigdog_ut | or the passthru is taking too long |
| 20:27.47 | timothy_ | wouldn't it be that the passthru is taking too long like bigdog_ut said |
| 20:28.02 | timothy_ | because the program works well from the command line |
| 20:28.39 | beandog | right |
| 20:28.41 | beandog | but youre running it as root |
| 20:28.44 | beandog | try it not as root. |
| 20:28.53 | beandog | if it doesnt work, then it doesnt matter who your non-root user is. |
| 20:28.54 | bigdog_ut | or you might need to write a script that does the image stuff for you |
| 20:29.00 | timothy_ | but from the command line, it does jump from the prompt to another program (pcmiler) and then back |
| 20:29.19 | timothy_ | I'll try not as root |
| 20:29.35 | bigdog_ut | you should not be running it as root? why are you running it as root? |
| 20:29.58 | timothy_ | sudo root bigdog |
| 20:30.24 | timothy_ | :) |
| 20:31.03 | bigdog_ut | unless pcmiller is crap and has setuid on it |
| 20:33.02 | timothy_ | could be dude |
| 20:33.14 | timothy_ | is there a way to test that? |
| 20:34.21 | timothy_ | I was able to run it as user timothy |
| 20:34.32 | bigdog_ut | then you are good |
| 20:34.53 | timothy_ | so it's not forcing an id |
| 20:34.55 | timothy_ | right? |
| 20:35.27 | cyco|w | passthru does need the full path, if you don't already have it.. passthru( "/usr/bin/whoami" ); |
| 20:36.20 | beandog | well setuid just means its going to run the program as the user that installed it. |
| 20:36.24 | beandog | I think. |
| 20:36.26 | beandog | :| |
| 20:36.32 | beandog | er, owns it. |
| 20:36.36 | beandog | bigdog_ut: is that right? |
| 20:37.25 | timothy_ | here' s the stat: |
| 20:37.27 | timothy_ | <PROTECTED> |
| 20:37.28 | timothy_ | <PROTECTED> |
| 20:37.28 | timothy_ | Device: 6200h/25088d Inode: 772744 Links: 1 |
| 20:37.28 | timothy_ | Access: (0644/-rw-r--r--) Uid: ( 1001/ timothy) Gid: ( 81/ apache) |
| 20:37.28 | timothy_ | Access: 2006-09-19 16:27:39.000000000 -0400 |
| 20:37.28 | timothy_ | Modify: 2006-09-19 16:27:24.000000000 -0400 |
| 20:37.30 | timothy_ | Change: 2006-09-19 16:27:24.000000000 -0400 |
| 20:37.53 | cyco|w | beandog: yes, it executes with the permissions of the owner |
| 20:38.12 | beandog | thanks cyco|w |
| 20:38.56 | bigdog_ut | setuid does system calls with evelated privs |
| 20:38.59 | bigdog_ut | commonly root |
| 20:39.12 | bigdog_ut | or whoever owns the file |
| 20:39.46 | bigdog_ut | so if root owns the file and as setuid then it can run as root |
| 20:40.31 | timothy_ | but it's working w/ root/root and timothy/apache |
| 20:41.08 | timothy_ | so, if it's running w/ timothy/apache, doesn't that mean it will run with apache/apache? |
| 20:41.19 | cyco|w | from the browser yes.. |
| 20:41.22 | cyco|w | from the command line no |
| 20:42.08 | timothy_ | you mean the opposite? since it's working from the command line but not the browser? |
| 20:42.15 | cyco|w | nope.. |
| 20:42.24 | cyco|w | from the cli it will run as the user executing it.. |
| 20:42.35 | timothy_ | could the issue be the passthru timeout instead? |
| 20:42.42 | timothy_ | oh, gotcha |
| 20:42.54 | cyco|w | also, not likely, but could be the DocumentRoot is setgid maybe.. |
| 20:43.51 | cyco|w | well, even then, that shouldn't break it.. nevermind.. |
| 20:43.56 | timothy_ | if the passthru on the browser is creating the file, but not writing to it, wouldn't it mean that it's got the necessary perms? |
| 20:44.14 | cyco|w | what is the user & group of the newly created file? |
| 20:44.22 | timothy_ | <PROTECTED> |
| 20:44.23 | timothy_ | <PROTECTED> |
| 20:44.23 | timothy_ | Device: 6200h/25088d Inode: 772744 Links: 1 |
| 20:44.23 | timothy_ | Access: (0644/-rw-r--r--) Uid: ( 1001/ timothy) Gid: ( 81/ apache) |
| 20:44.23 | timothy_ | Access: 2006-09-19 16:27:39.000000000 -0400 |
| 20:44.23 | timothy_ | Modify: 2006-09-19 16:27:24.000000000 -0400 |
| 20:44.25 | timothy_ | Change: 2006-09-19 16:27:24.000000000 -0400 |
| 20:44.29 | timothy_ | that's the stat response |
| 20:44.38 | cyco|w | this is when run from the browser solely? |
| 20:45.44 | bigdog_ut | apache might not have a shell to run the command |
| 20:46.05 | cyco|w | timothy_: so if you delete that file and run it only from the browser, that is the owner/group that the file gets? |
| 20:46.06 | timothy_ | cyco|w: yes |
| 20:46.14 | timothy_ | yes |
| 20:46.25 | cyco|w | so your browser is running as timothy.apache then? |
| 20:46.40 | timothy_ | bigdog_ut: how would I create the needed shell? |
| 20:47.31 | timothy_ | cyco|w: I pasted the wrong one, sorry |
| 20:47.51 | bigdog_ut | timothy_, try now and see what you get? |
| 20:48.23 | cyco|w | this is something really simple.. it has to do with perms and/or ownership.. i've seen this lots of times.. |
| 20:48.52 | cyco|w | just paste the file info when it's run only from the browser.. not from the cli.. |
| 20:48.56 | bigdog_ut | timothy_, did you try it again |
| 20:49.19 | timothy_ | bigdog_ut: just did, no dice |
| 20:49.33 | cyco|w | timothy_: could you paste the right one? |
| 20:49.37 | timothy_ | here's the stat of that attempt: |
| 20:49.40 | timothy_ | <PROTECTED> |
| 20:49.41 | timothy_ | <PROTECTED> |
| 20:49.41 | timothy_ | Device: 6200h/25088d Inode: 772747 Links: 1 |
| 20:49.41 | timothy_ | Access: (0644/-rw-r--r--) Uid: ( 81/ apache) Gid: ( 81/ apache) |
| 20:49.41 | timothy_ | Access: 2006-09-19 16:42:11.000000000 -0400 |
| 20:49.41 | timothy_ | Modify: 2006-09-19 16:42:11.000000000 -0400 |
| 20:49.43 | timothy_ | Change: 2006-09-19 16:42:11.000000000 -0400 |
| 20:49.56 | cyco|w | that looks better now.. |
| 20:50.07 | cyco|w | this is running the passthru("whoami"); command? |
| 20:50.51 | cyco|w | which should be passthru("/usr/bin/whoami"); or wherever whoami resides on your particular distro.. |
| 20:51.42 | beandog | Youd ont need the full path in passthru() |
| 20:51.53 | beandog | not on his system, at least. |
| 20:52.36 | cyco|w | does your webserver run with the system $PATH in it's env? |
| 20:52.43 | cyco|w | hopefully not.. |
| 20:52.48 | cyco|w | so in that case it would need it.. |
| 20:52.52 | timothy_ | it functions w/o the full path |
| 20:52.57 | cyco|w | cron needs the full path.. why wouldn't apache? |
| 20:53.11 | timothy_ | **shrug** |
| 20:53.15 | beandog | cron doesnt need the full path either. |
| 20:53.27 | cyco|w | timothy_: if you run it from the cli, do you get the disired output? |
| 20:53.28 | beandog | it is in it's $PATH |
| 20:54.02 | timothy_ | cyco|w: yes |
| 20:54.24 | cyco|w | then that's the problem.. it needs the full path for apache to know where the whoami binary is located.. |
| 20:56.25 | cyco|w | timothy_: wait a sec.. when you say it works from the cli, do you mean that it displays the info on the screen? or that it actually writes info into the file it creates? |
| 20:58.08 | timothy_ | from the command line, it creates and writes the file |
| 20:58.28 | cyco|w | alright.. that's good.. |
| 20:58.36 | timothy_ | from the browser, I have run "date" and "whoami" w/ no problem |
| 20:59.05 | timothy_ | but "pcbatch........." will create the file and close before writing to the file |
| 20:59.17 | cyco|w | ahh.. what's pcbatch? |
| 20:59.17 | timothy_ | (from the browser) |
| 20:59.26 | timothy_ | the app I'm trying to run |
| 20:59.38 | cyco|w | what is it supposed to output? |
| 20:59.53 | timothy_ | it converts a plain text file w/ an .in extention to a plain text .out file |
| 20:59.58 | timothy_ | mileage |
| 21:00.05 | timothy_ | from point a to point b |
| 21:00.18 | timothy_ | Sample .in file: |
| 21:00.35 | cyco|w | ahh.. see, and i did come in late to this one and hadn't heard about that one yet.. (tried to read back a bunch, but missed it) |
| 21:00.41 | cyco|w | so how big is the file then? |
| 21:00.59 | timothy_ | couple hundred chars |
| 21:01.10 | cyco|w | oh.. so that shouldn't be timing out.. |
| 21:01.22 | timothy_ | sample .in file: |
| 21:01.26 | timothy_ | CEDAR RAPIDS,IASIOUX CITY, IA |
| 21:01.26 | timothy_ | CEDAR RAPIDS,IAMODALE, IA |
| 21:01.27 | timothy_ | CEDAR RAPIDS,IASPENCER, IA |
| 21:01.57 | timothy_ | sample .out: |
| 21:02.00 | timothy_ | CEDAR RAPIDS,IASIOUX CITY, IA 266.7 4.8 |
| 21:02.01 | timothy_ | CEDAR RAPIDS,IAMODALE, IA 259.5 4.1 |
| 21:02.01 | timothy_ | CEDAR RAPIDS,IASPENCER, IA 242.8 4.3 |
| 21:02.28 | timothy_ | the first new column is the mileage, the second is the time |
| 21:02.34 | timothy_ | drive time |
| 21:02.35 | cyco|w | ahh.. so what code is reading in the .in file then? |
| 21:02.52 | timothy_ | it's all on the passthru...... |
| 21:03.05 | cyco|w | ahh, you're using textutils to process the file? |
| 21:03.13 | timothy_ | pcmbatch P /var/www/vhosts/html/apps/mylumbertradernet/pcmiler/route6.in 1 -t |
| 21:03.19 | timothy_ | YES |
| 21:03.34 | timothy_ | :) |
| 21:03.53 | cyco|w | what's in pcmbatch? it sounds like that's where the problem lies.. |
| 21:04.13 | cyco|w | it's probably just not handing control back to the passthru command that called it.. |
| 21:04.15 | timothy_ | pcmbatch is the program that converts the file |
| 21:04.22 | cyco|w | is it a binary or script? |
| 21:05.19 | timothy_ | binary |
| 21:05.23 | timothy_ | I guess |
| 21:05.27 | timothy_ | it's a full on app |
| 21:05.48 | cyco|w | it'll tell you if you run 'file /path/to/pcmbatch' |
| 21:05.51 | cyco|w | anyway.. |
| 21:06.23 | cyco|w | if you don't need the output from the .out file to be directly displayed in the browser, you could use exec it sounds like.. |
| 21:06.38 | cyco|w | and that might work better cause exec won't wait for the program to finish.. |
| 21:07.21 | timothy_ | file == pcmbatch: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.2.5, dynamically linked (uses shared libs), for GNU/Linux 2.2.5, stripped |
| 21:07.45 | timothy_ | I've tried exec, but I'll try again |
| 21:08.40 | cyco|w | you could set a return_var too (for the exit status..) that might help if it doesn't work right.. |
| 21:09.48 | timothy_ | no go on exec |
| 21:10.21 | timothy_ | the code is: exec($cmd, $output); |
| 21:10.40 | timothy_ | w/ $cmd being a string of the command to execute |
| 21:12.04 | cyco|w | and $cmd has the full path to the pcmbatch binary? |
| 21:12.09 | macnewbold | exec usually waits for it to finish, unless you did "&" in the command at the end, in which case it finishes really quick |
| 21:12.29 | timothy_ | no on the path, no on the & |
| 21:12.34 | timothy_ | I'll try both |
| 21:12.34 | cyco|w | yeah, you're right cause it returns the last output.. sorry.. |
| 21:14.54 | timothy_ | the ampersand didn't help |
| 21:15.08 | timothy_ | and the full path didn't create the file or write to it |
| 21:15.31 | cyco|w | interesting.. what is your php error log saying about all of this anyway? |
| 21:17.40 | timothy_ | looking |
| 21:23.04 | cyco|w | is that binary only responsible for converting those files like that? cause if it is, seems like you could have written the php code to do the same thing by now.. |
| 21:23.07 | cyco|w | ;) |
| 21:23.21 | timothy_ | :) |
| 21:23.32 | timothy_ | it pulls the mileage data from pcmiler |
| 21:23.40 | timothy_ | that's what I'm needing |
| 21:23.55 | timothy_ | to know the drive miles from point A to point B |
| 21:24.04 | timothy_ | no the way a crow flies either |
| 21:24.36 | timothy_ | so I can't do a long lat calculation |
| 21:25.18 | cyco|w | gotcha |
| 21:25.42 | timothy_ | where would I find the error logs? |
| 21:26.51 | cyco|w | depends.. some isps put them in the current working dir of the file.. if it's your own server, it'll depend on the distro, but generally under /var/log/httpd or /var/log/apache or /usr/local/apache/logs.. something like that.. |
| 21:30.53 | timothy_ | I've got a few: Error opening terminal: unknown |
| 21:32.44 | beandog | /var/log/apache2/error_log |
| 21:32.55 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 21:32.55 | *** mode/#UPHPU [+o wps] by ChanServ |
| 21:34.42 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 21:34.42 | *** mode/#UPHPU [+o wps] by ChanServ |
| 21:34.50 | timothy_ | for me it's under /var/log/apache/error_log |
| 21:35.07 | cyco|w | that's where your php errors are being sent as well? |
| 21:35.28 | timothy_ | I dunno |
| 21:35.32 | cyco|w | you can verify the php logging setup in php.ini.. just look for php_error |
| 21:35.33 | timothy_ | I guess |
| 21:35.37 | timothy_ | k |
| 21:35.57 | beandog | php logging is disabled by default |
| 21:36.00 | cyco|w | sorry, i mean error_log |
| 21:36.01 | beandog | youd have to turn it on |
| 21:36.07 | cyco|w | most isps do.. |
| 21:36.09 | beandog | then do /etc/init.d/apache2 reload |
| 21:39.02 | timothy_ | mine's off... 1 min |
| 21:39.17 | cyco|w | so they are in your apache error_log then where you were looking.. |
| 21:39.30 | macnewbold | any of you remember that gradefix project some of you helped test? well, it's fully launched now, and there's an article about it in the Daily Utah Chronicle (Univ. of Utah campus paper) about it: http://www.dailyutahchronicle.com/media/storage/paper244/news/2006/09/19/News/Students.Launch.Web.Site.To.Improve.Grades-2284314.shtml?norewrite200609191637&sourcedomain=www.dailyutahchronicle.com |
| 21:41.40 | *** join/#uphpu _psychic_ (n=_psychic@71.32.228.156) |
| 21:44.54 | timothy_ | cyco|w: just flipped it on, then saw your msg |
| 21:45.12 | timothy_ | now that it's on, I'm not sure where it's saving errors |
| 21:45.26 | cyco|w | well, that's always better anyway.. turning off the error display and logging to a file.. |
| 21:45.46 | cyco|w | you specify the path to the file on the error_log= line.. |
| 21:46.19 | cyco|w | and you can do /var/log/apache/php_errors if you want to keep it in the same place.. |
| 21:47.22 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 21:47.22 | *** mode/#UPHPU [+o wps] by ChanServ |
| 21:48.41 | *** join/#uphpu wps (n=wps@64.72.210.10) |
| 21:48.42 | *** mode/#UPHPU [+o wps] by ChanServ |
| 21:50.08 | timothy_ | doing that now........ |
| 21:53.24 | cyco|w | it may be the default, but make sure the error_reporting = E_ALL is set too.. |
| 21:53.58 | timothy_ | k |
| 21:53.58 | cyco|w | and log_errors = On as well.. |
| 21:54.05 | timothy_ | k |
| 22:01.29 | timothy_ | right now, I've got: error_reporting = E_ALL & ~E_NOTICE |
| 22:01.36 | timothy_ | is that good? |
| 22:02.03 | timothy_ | but it didn't issue an error to the file when I ran the exec command again in the browser |
| 22:02.19 | cyco|w | sure |
| 22:02.33 | cyco|w | did you restart apache? |
| 22:02.40 | timothy_ | y |
| 22:02.45 | timothy_ | but I will again.... |
| 22:03.03 | cyco|w | and you created the new log file too? |
| 22:03.21 | timothy_ | y |
| 22:03.29 | cyco|w | and set the owner.group and perms the same as the access and error _logs? |
| 22:03.39 | timothy_ | restarted |
| 22:03.42 | timothy_ | ran script |
| 22:03.45 | timothy_ | no errors |
| 22:03.56 | timothy_ | == huge headache |
| 22:03.58 | cyco|w | hmm.. strange.. unless it thinks it's doing it right.. |
| 22:04.22 | timothy_ | I think PHP does |
| 22:05.05 | timothy_ | because to begin with, I ran the backtick/popen/exec into a var |
| 22:05.18 | timothy_ | so: $operation = popen(...... |
| 22:05.33 | timothy_ | then print_r($operation) |
| 22:05.38 | timothy_ | and it showed nothing |
| 22:05.55 | timothy_ | nothing good, nothing bad |
| 22:06.31 | timothy_ | that's why I think it's a timeout thing..... |
| 22:06.56 | timothy_ | PHP runs the command and goes on to the next thing before the command line is finished |
| 22:08.05 | timothy_ | pcmbatch creates the .out file, then opens pcmiler to run the miles, then comes back to pcmbatch to write the output to the .out file, then back to the command line to say "output complete" |
| 22:08.47 | timothy_ | so, pcmbatch can write the file, but upon jumping into pcmiler, PHP thinks the deal is done and goes on w/ life |
| 22:09.16 | timothy_ | is there any way to stall PHP for a second or two? |
| 22:09.38 | _psychic_ | http://us3.php.net/sleep |
| 22:09.38 | cyco|w | sorry.. i had to make some ramen.. |
| 22:09.50 | timothy_ | .... during the execution of the backtick/popen/exec |
| 22:10.17 | _psychic_ | no the external command would need to do that, right? |
| 22:10.36 | timothy_ | ya, probably |
| 22:10.49 | timothy_ | could I use popen to open a command line |
| 22:11.05 | timothy_ | then issue a command to keep popen running for a certian time |
| 22:11.35 | timothy_ | then use fputs to issue the real command? |
| 22:14.19 | cyco|w | sorry, i have to go to the other office and help someone install php man.. |
| 22:15.21 | timothy_ | np |
| 22:34.08 | macnewbold | timothy_: your popen will return soon (as it should), but you should be able to read from it, |
| 22:34.13 | macnewbold | wait a while, read again, etc. |
| 22:34.41 | timothy_ | what do you mean? |
| 22:35.26 | macnewbold | you wanted to stall for a few seconds while the program runs, right? |
| 22:35.54 | macnewbold | I'm only catching pieces of the conversation, so I'm not sure I get it all |
| 22:36.17 | macnewbold | usually when something (like your pcmbatch) is running in the background, |
| 22:36.33 | macnewbold | you'll start it, then sleep, then check on its progress (by looking at the output file for example), |
| 22:36.40 | macnewbold | then sleep again if necessary, and check progress again, |
| 22:36.50 | macnewbold | and repeat until you're satisfied (i.e. it finished) |
| 22:37.04 | cyco|w | i don't know that that would solve the issue at all. passthru and exec should wait and that program wouldn't take long enough to time out the server.. i'm certain it has something to do with persmissions and/or ownership.. |
| 22:37.22 | cyco|w | it usually does.. |
| 22:37.44 | timothy_ | I bet you're right.... |
| 22:37.51 | timothy_ | I'm calling them in the morning |
| 22:38.19 | timothy_ | paid dang near $7000 for the thing |
| 22:39.01 | timothy_ | but I thought it was something stupid I was doing |
| 22:39.40 | cyco|w | nah.. i've see things like this a lot.. it may even be something to do with the permissions or ownership of the parent dir.. |
| 22:40.25 | timothy_ | the dir w/ the .in and .out files is set to 777 |
| 22:40.46 | timothy_ | are you talking about another dir? |
| 22:41.15 | cyco|w | just throwing out ideas from what i've seen in the past.. |
| 22:42.12 | timothy_ | oh |
| 22:42.33 | timothy_ | okay |
| 22:42.54 | cyco|w | speaking of that.. where is said server located? is it your own or at an ISP? |
| 22:43.00 | cyco|w | or hosting provider i mean.. |
| 22:43.09 | timothy_ | I have a linode |
| 22:43.16 | cyco|w | ahh.. nice.. |
| 22:43.17 | timothy_ | linode.com |
| 22:43.21 | cyco|w | oh i know.. |
| 22:43.31 | timothy_ | ya? do you have one? |
| 22:43.43 | cyco|w | nope.. but they look pretty sweet.. |
| 22:44.43 | timothy_ | it's working well for now, I have no production URLs on it yet, so we'll se |
| 22:44.44 | timothy_ | see |
| 22:45.00 | timothy_ | but it appears to be easily expandable |
| 22:45.06 | cyco|w | that's cool.. |
| 22:45.12 | timothy_ | ya |
| 22:47.08 | cyco|w | clear |
| 22:47.10 | cyco|w | sorry |
| 22:53.32 | timothy_ | is proc_open better than popen? |
| 22:53.59 | cyco|w | never used it |
| 23:06.19 | *** join/#uphpu jnbek (n=jnbek@c-24-10-178-17.hsd1.ut.comcast.net) |
| 23:42.24 | Utah_Dave | anybody have second for an SQL question? |
| 23:43.28 | timothy_ | I've got to go in 5 |
| 23:43.33 | timothy_ | but I'll help if I can |
| 23:44.27 | Utah_Dave | ok, I have an archive table and a temp table |
| 23:45.07 | Utah_Dave | we have an app that dumps new entries into the temp table, as well as updates to rows that are already in the archive table |
| 23:45.50 | Utah_Dave | I need to do some kind of SELECT INTO or INSERT INTO so that I insert new rows and update rows where the primary key already exists |
| 23:46.05 | Utah_Dave | I'm thinking I can use something I found on http://dev.mysql.com/doc/refman/5.0/en/insert-select.html |
| 23:46.09 | Utah_Dave | an insert select. |
| 23:46.18 | Utah_Dave | do you have any ideas? |
| 23:46.44 | cyco|w | update will do it automatically if the pri key already exists in the table.. |
| 23:47.09 | timothy_ | I use something like: |
| 23:47.13 | Utah_Dave | will it insert the new ones? |
| 23:47.22 | cyco|w | if the key doesn't match |
| 23:47.22 | timothy_ | upon duplicate key INSERT |
| 23:47.46 | timothy_ | hang on, let me look |
| 23:49.23 | timothy_ | ya, here's one |
| 23:50.22 | timothy_ | "INSERT INTO table (id, userId, joined, lastPolled) VALUES ($id, $this->userId, $now, $now) |
| 23:50.22 | timothy_ | ON DUPLICATE KEY UPDATE joined= $now, left = NULL, lastPolled = $now" |
| 23:50.52 | timothy_ | does that help? |
| 23:51.15 | Utah_Dave | almost. I want to do this with one SQL statement |
| 23:51.32 | Utah_Dave | so do some kind of select and insert at the same time |
| 23:51.46 | timothy_ | that runs as one statement on mine |
| 23:52.25 | timothy_ | one mysql_query() that is |
| 23:53.07 | Utah_Dave | OK, but where are you getting $id and $this->userId? |
| 23:53.31 | timothy_ | elsewhere in the code |
| 23:53.41 | Utah_Dave | yeah |
| 23:53.52 | Utah_Dave | I'm trying to do it all in one database call |
| 23:54.00 | timothy_ | so, this is in an OOP function |
| 23:54.17 | timothy_ | oh |
| 23:54.27 | timothy_ | guess I can't help then |
| 23:54.44 | timothy_ | sorry dude |
| 23:55.01 | cyco|w | you want to update the archive table if they exist and stick them in the temp table as well? |
| 23:55.13 | cyco|w | or stick them in both regardless? |
| 23:55.24 | cyco|w | (and update the archive if they exist there) |
| 23:56.08 | Utah_Dave | basically I want to empty the temp table. Some of the temp table entries are brand new and some of the entries are updates to entries that have already been inserted into the archive table |
| 23:57.01 | Utah_Dave | I'm trying to get one database call that will insert all the new entries from the temp table into the archive table AND update those entries that already exist in the archive table |
| 23:57.09 | cyco|w | ahh.. gotcha.. |
| 23:57.29 | cyco|w | yeah, i don't know that i've ever done that myself.. |
| 23:57.38 | Utah_Dave | yeah, it's kind of tough. |
| 23:58.11 | Utah_Dave | I think I need an INSERT INTO. I might be close, but I was just wondering if anyone here had any better ideas. I'm not exactly a database guru :) |
| 23:58.18 | cyco|w | would something like a select temp.userid as archive.userid type of thing maybe work? |
| 23:58.27 | cyco|w | nah, i guess that just names them.. |