00:44.03 | *** join/#tomcat andyt (n=andyt@THOMPSON.ADMIN.SCPS.NYU.EDU) |
01:44.03 | *** join/#tomcat oxi (i=trallall@gve-gix-bbcs-dynip-128-048.vtx.ch) |
02:27.06 | *** join/#tomcat Quiddity (n=steve@c-98-198-199-37.hsd1.tx.comcast.net) |
02:38.05 | *** join/#tomcat Quiddity (n=steve@c-98-198-199-37.hsd1.tx.comcast.net) |
02:51.56 | *** join/#tomcat albert_kam (n=albert_k@222.124.39.108) |
03:23.51 | *** part/#tomcat albert_kam (n=albert_k@222.124.39.108) |
03:26.13 | *** join/#tomcat albert_kam (n=albert_k@222.124.39.108) |
04:17.53 | *** join/#tomcat acidjnk (i=acid@pD9509E4D.dip0.t-ipconnect.de) |
04:24.32 | *** join/#tomcat albert_kam (n=albert_k@222.124.39.108) |
04:54.52 | *** join/#tomcat kanai (n=kanai@124.32.114.226) |
05:29.12 | *** join/#tomcat cppexpert (n=cppexper@c-98-207-85-93.hsd1.ca.comcast.net) |
06:13.00 | *** join/#tomcat andyt (n=andyt@THOMPSON.ADMIN.SCPS.NYU.EDU) |
06:41.17 | *** join/#tomcat cppexpert (n=cppexper@c-98-207-85-93.hsd1.ca.comcast.net) |
06:55.43 | *** join/#tomcat rambo123 (n=rocky@59.96.165.157) |
07:52.27 | *** join/#tomcat |Chris| (n=cdupont@ip-129.net-80-236-80.issy.rev.numericable.fr) |
07:56.51 | *** join/#tomcat Piet (n=dsdf@77.61.181.235) |
07:57.04 | Piet | MSHL did you read <Piet> MSHL i did configured a context but still the same |
07:57.05 | Piet | <Piet> http://pastebin.com/m25262187 |
07:57.05 | Piet | <Piet> please modify it its wrong i have to go now |
08:06.56 | *** part/#tomcat albert_kam (n=albert_k@222.124.39.108) |
08:13.27 | *** join/#tomcat kjkoster5489 (n=kjkoster@233-250-103-145.cust.glaslokaal.nl) |
08:24.21 | *** join/#tomcat vikumar (n=Vimal@nat/redhat-in/x-1deb105c674b47df) |
08:25.14 | *** join/#tomcat Wello (n=Wello@stargate.fastweb.it) |
08:33.42 | Piet | <PROTECTED> |
08:36.58 | kjkoster5489 | Piet: If you access Tomcat directly (i.e. not through Apache) it works well or not? |
09:03.08 | *** part/#tomcat remiss (i=roger@remiss.org) |
09:03.17 | Piet | i can access tomcat directly but if i do for example www.example.com:8180 it goes also to the tomcat index page |
09:21.54 | kjkoster5489 | Piet: if you access Tomcat directly, what page do you see? What is the webroot of your webapp? Wat is the installation directory of your webapp? What is in your context file, if you use one? What URL do you use for each test? How the hell do you expect us to help you without actually providing some actual information about your problem? And no "it does not work" is not information. |
09:28.00 | *** join/#tomcat albert_kam (n=albert_k@222.124.39.108) |
09:31.13 | Piet | webroot = /var/www/ install dir webapp = /var/lib/tomcat5/webapps i use java.com configured in my hosts to test. i dont use a context file, if i access tomcat directly i see the " If you're seeing this page via a web browser, it means you've setup Tomcat successfully." page i use a worker to tell that tomcat must handle the jsp page |
09:36.05 | kjkoster5489 | What is the path of your jsp page? |
09:36.53 | Piet | my index.jsp is in /var/www/java/ |
09:37.14 | kjkoster5489 | What webapp is that JSP part of? |
09:39.01 | kjkoster5489 | I don't think you can just hand jsp's over to tomcat like that. You have to make a webapp that contains it. |
09:41.18 | *** join/#tomcat spmccann (n=spmcann9@nat/intel/x-a6727d1de42bdac5) |
09:43.01 | Piet | how? could you give me a example? |
09:43.34 | *** join/#tomcat milestone (n=hoffmann@213.157.14.203) |
09:44.02 | milestone | hi all |
09:44.32 | kjkoster5489 | Piet: follow a Tomcat tutorial to make a webapp and put your JSP in there. |
09:45.02 | milestone | i am running tomcat 5.5.25 from ubuntu gutsy gibbon behind an apache 2.2.4 |
09:45.33 | milestone | wihtin my turbine web application i have a page which reads a logfile and keeps reading it and writes the lines to the browser |
09:46.09 | milestone | this works fine when accessing the tomcat http connector on port 8080 but not when accessing the webapp through mod-jk-1.2.26 |
09:46.26 | milestone | the page seems to load forever |
09:46.39 | milestone | as if mod-jk is waiting for tomcat to end its response |
09:46.43 | milestone | which never happens |
09:46.56 | milestone | because tomcat keeps writing to the output stream |
09:47.43 | milestone | is there a special config option to get mod-jk to flush its buffers immediately and not to wait for tomcat to finish its response? |
09:51.58 | milestone | maybe recovery_options is the way to go? |
10:44.01 | *** part/#tomcat albert_kam (n=albert_k@222.124.39.108) |
10:48.24 | *** join/#tomcat Evert (n=dsdf@77.61.181.235) |
10:48.33 | *** join/#tomcat albert_kam (n=albert_k@222.124.39.108) |
11:00.31 | ca | do tomcat 6.0.10 support national characters(åæø) in a form ldap authentication? |
11:06.17 | *** join/#tomcat Dwayne_Hicks (n=elsie@adsl-68-74-112-144.dsl.emhril.ameritech.net) |
11:40.38 | *** join/#tomcat victori_ (n=victori@cpe-75-80-141-237.san.res.rr.com) |
12:07.40 | *** join/#tomcat rambo123 (n=rocky@59.96.120.245) |
12:22.10 | *** join/#tomcat matt7676 (n=matt@popper.yospace.com) |
12:44.48 | *** join/#tomcat rambo123 (n=rocky@59.94.104.15) |
12:50.57 | *** join/#tomcat MHSL- (i=Linux@116.88.6.47) |
12:57.06 | *** join/#tomcat Evert (n=dsdf@77.61.181.235) [NETSPLIT VICTIM] |
12:57.06 | *** join/#tomcat rchern-afk (n=stlgirl@unaffiliated/rchern) [NETSPLIT VICTIM] |
12:57.06 | *** join/#tomcat fifo_ (n=quintin@196.38.199.139) [NETSPLIT VICTIM] |
12:57.06 | *** join/#tomcat fix (n=fix@86.39.152.5) [NETSPLIT VICTIM] |
12:57.23 | *** join/#tomcat rchern (i=stlgirl@unaffiliated/rchern) |
13:06.06 | *** join/#tomcat supernova (n=Miranda@213.200.64.123) |
13:06.35 | *** join/#tomcat hstrauss (n=chatzill@p5B0A66E6.dip.t-dialin.net) |
13:06.41 | *** part/#tomcat hstrauss (n=chatzill@p5B0A66E6.dip.t-dialin.net) |
13:06.46 | ca | do tomcat 6.0.10 support national characters(åæø) in a form ldap authentication? |
13:07.27 | milestone | ca: why shouldn't it |
13:08.35 | ca | I have no idea, but in out application users that have national characters in the username or password cannot logg in, if we change the username or password so that none of them has national characters they can log in.. |
13:09.46 | ca | in ldap the names are sored as utf8 (atleast slapcat shows them as utf8 when it is exported) and all our jsp files/etc use utf8 and the system character encoding 8linux) is set to utf-8 |
13:11.41 | ca | and we ave this in the web.xml file <web-app character-encoding="UTF-8"> |
13:12.12 | ca | but from what i see in the ldap log, it looks as if the jndi realm athentication sends iso-8859-1 |
13:15.24 | supernova | Tomcat 5.0.x Redhat - I have a DB layer library that is deployed in the commen/lib folder. The Problem: I have two Tomcats running on different databases, so I need to compile two versions of the DB layer and deploy the one or the other. For automation, it would be great if the server could decide to take the one or the other on deployment (or let every server instance ignore one of them). Is this somehow possible? What would be the best way to do this? |
13:22.24 | *** join/#tomcat matt7676 (n=matt@popper.yospace.com) |
13:24.44 | ca | milestone, do you know if there is any way to configure which charaterencoding to be used in the jndi realm? |
13:31.04 | MHSL | supernova, you're not supposed to put any libraries in common/lib folder unless you need both tomcat and your web app to be able to access the libraries, e.g. jdbc driver |
13:33.55 | MHSL | supernova, however if you still need to put your library under common/lib, e.g., to sharing the same library across different web apps without deploying it multiple times, you can use a configuration file, or anything, to differentiate between the two. this is app design problem, you cant expect tomcat to do this separation for you. |
13:44.30 | supernova | ca, MHSL, yes I know it's not the nicest solution as there are wars and more. But it would be a nightmare to change this by now. The configuration solution does not work for us because the the classes are identically named, but differently enhanced by JDO. I think I'll go for a dynamic deployment script as this seems the most transparent solution for me. THANKS! |
13:45.09 | kjkoster5489 | supernova: you're opting to prolong the agony. |
13:45.14 | kjkoster5489 | Your choice, of course. |
13:46.15 | MHSL | yep, that's not solving the problem, but avoiding it |
13:46.49 | kjkoster5489 | One might even argue that he even fails to avoid the problem. He has to put effort into working wround it. |
13:47.04 | ca | supernova, I did not answer your question, but since you included me, i thought I'd best answer .-) why cannot you put your DB layer library in WEB-INF/lib ? |
13:48.53 | ca | any of you know if it is possible to set which character encoding the jndirealm is using? |
13:51.41 | supernova | The problem is not where to put it. WEB/INF and common/lib work both, but depending on our build processes it's common/lib, so it's accessible for all webapps. The true problem is that this HOLE solution (ALL webapps + ALL libs) needs to be deployed on another machine. And the only difference is the enhancement of these lib classes. So I do not know, why my idea is "agony". Even putting the lib into WEB/INF would not solve the problem, but generate a bu |
13:54.03 | ca | supernova, ah ok sorry I missread your question (missed the part with 2 tomcats) form the looks of it this migt be best handled with your deployment scripts (if you have any) |
13:54.47 | supernova | ca, no problem. Thanks you all for thinking over my problems :-) |
13:55.34 | *** part/#tomcat matt7676 (n=matt@popper.yospace.com) |
13:56.24 | *** join/#tomcat matt7676 (n=matt@popper.yospace.com) |
13:56.44 | kjkoster5489 | supernova: true, I also misunderstood. But if it's two tomcats, why put anythinh in common/lib? |
13:57.28 | ca | kjkoster, so that it accessiible to ALL webapps in that tomcat .-) |
13:57.52 | ca | othervice he would have to put that library in every WEB-INF/lib directory of that tomcat |
14:01.02 | kjkoster5489 | ca: so? What does diskspace cost these days? Not as much as manhours lost in resolving inter-webapp dependencies and versioning. |
14:03.54 | supernova | kjkoster5489, this is not for disk space, but you'll ensure the same lib version is used across all webapps. Other way round, you are forced to ensure intregrity by you build process. |
14:04.16 | supernova | That also works, but depends on your environment ;-) |
14:04.35 | ca | kjkoster, perhaps... but if ALL webapps are supposed to use same version updating the db layer for one webapp means you need to update all webaps, puting it in common/lib ensures that they all use same version all the time |
14:04.36 | kjkoster5489 | I always strive to make my webapps self-containted and deploys scripted. |
14:04.53 | ca | I'm not saying this is either good or bad, I just say that there is reasons for it .-) |
14:05.13 | kjkoster5489 | Yes, and replacing Tomcat suddenly becomes a project, where I just replace a symlink. |
14:05.30 | kjkoster5489 | I'm saying lib/common is bad. :) |
14:09.35 | ca | kjkoster, I'm also using symlinks :-) but I understand why some persons put things in commons/lib .-9 |
14:10.03 | ca | kjkoster, btw hoe does symlinks work if your tomcat is running in windows? .-) |
14:12.52 | *** join/#tomcat Nicke (n=niclasa@ua-83-227-140-135.cust.bredbandsbolaget.se) |
14:13.41 | kjkoster5489 | ca: Ah, you got me there. No symlinks on WIndows. My. Too bad for my grandma, I say. ;-) |
14:14.15 | kjkoster5489 | Actually, scratch that. It works in cygwin. :) |
14:14.59 | ca | kjkoster, if you know how to use cygwin .-) |
14:15.06 | ca | which tomcat does not require .-) |
14:15.40 | ca | ut who would run tomcat on windows as a server, I don't know, for deevelopment I can undertand it but not for server |
14:16.09 | *** join/#tomcat deepjoy (n=dj@122.167.243.106) |
14:17.10 | deepjoy | Hi I want to aggregate log4j and catalina logs from multiple tomcat instances into a single place. Could anybody point me to resources on how to do this? |
14:18.29 | deepjoy | I'm using tomcat 6.0.16 on jdk1.5.13 |
14:18.44 | deepjoy | on centos 5 boxes |
14:18.54 | kjkoster5489 | deepjoy: look at the syslog appender in log4j and check out chainsaw. |
14:19.42 | deepjoy | wow thanks that looks awesome |
14:19.45 | ca | deepjoy, I'm not that familiar with log4j to know if this can be done out of the box, bit if it cannot, create a logging server that recieves logg messages using tcp/ip/rmi/soap or whatever, then create a log4j implementation that connects to that server and sends the data .-) |
14:19.58 | ca | deepjoy looks like you got our answer .-9 |
14:20.02 | kjkoster5489 | or use the JDBC appender |
14:20.12 | kjkoster5489 | Just dump everything in a database somewhere. |
14:20.31 | deepjoy | I was considering syslog appender and cryptcat before now. I was not aware of chainsaw |
14:20.58 | kjkoster5489 | I never used chainsaw, but I hear good mumblings about it. |
14:21.10 | deepjoy | I looked at some DB options but was not convinced |
14:21.35 | kjkoster5489 | What is syslog other than a weird database? Or a file for that matter. :) |
14:21.46 | deepjoy | kjkoster5489: thanks I'll give it a spin and report back :-) |
14:22.18 | deepjoy | ca: thanks I am using jmx+rmi for some other stuff. not sure if it'll scale to real time logging |
14:22.37 | deepjoy | seemed to have a bit of lag if you ask me |
14:23.32 | ca | any of you know if it is possible to set which character encoding the jndirealm is using? (still wanting to know) |
14:26.37 | kjkoster5489 | Still don't know the answer, sorry. |
14:30.51 | *** join/#tomcat Vanuatoo (n=Vanuatoo@host-85-114-227-35.adsl.caucasus.net) |
14:34.55 | *** join/#tomcat randrew (n=raj@dolmen.cc.columbia.edu) |
14:48.18 | *** join/#tomcat jkl (n=jkl@sistemas.uis.edu.co) |
14:51.31 | *** join/#tomcat {Chris} (n=cdupont@LMontsouris-152-61-3-84.w80-13.abo.wanadoo.fr) |
15:36.17 | *** join/#tomcat aplata (n=aplata@CPE00173fee2ecf-CM0011e6c417e1.cpe.net.cable.rogers.com) |
15:37.20 | aplata | I was wondering if anyone can help me. I would like to learn more about application support in relation to tomcat can any one tell me how I can get information on monitoring, troubleshooting log files, etc. |
15:39.33 | randrew | aplata: http://www.oreilly.com/catalog/tomcat/ perhaps |
15:44.47 | aplata | thanks randrew |
15:54.26 | *** join/#tomcat Quiddity (n=steve@ng1.cptxoffice.net) |
16:35.41 | *** join/#tomcat |Chris| (n=cdupont@LMontsouris-152-61-3-84.w80-13.abo.wanadoo.fr) |
16:44.40 | *** part/#tomcat deepjoy (n=dj@122.167.243.106) |
16:58.12 | *** join/#tomcat Zack1403 (n=Zack1403@64.122.14.80) |
17:05.58 | *** join/#tomcat spmccann (n=spmcann9@nat/intel/x-e49268c9c999ecfa) |
17:13.00 | *** join/#tomcat roue (n=roue@afterthought.ACS.Carleton.edu) |
17:25.47 | *** join/#tomcat rambo123 (n=rocky@59.94.99.235) |
17:28.33 | *** join/#tomcat _hen (n=hen@c-67-183-122-76.hsd1.wa.comcast.net) |
17:35.20 | *** join/#tomcat jasonb (n=jasonb@75.144.23.117) |
17:36.17 | *** join/#tomcat quadmasta (n=bdcribbs@74.92.194.125) |
17:37.23 | quadmasta | Tomcat 5.5.20, Java 1.6.0_03, Win XP Pro |
17:38.20 | quadmasta | I followed the directions on this site http://tomcat.apache.org/tomcat-5.5-doc/ssl-howto.html to try to enable SSL on my local machine for an application |
17:39.00 | *** join/#tomcat jwisher (n=jwisher@unaffiliated/jwisher) |
17:39.43 | quadmasta | is there a way to automatically set up a trigger or a forward so if someone tries to access the SSL port using the http protocol the server will make sure the user ends up on the SSL port with the https protocol? |
17:42.37 | pfn | no |
17:42.40 | pfn | that's not how it works |
17:42.47 | pfn | if you go http to 443 then you're screwed |
17:42.48 | pfn | period |
17:42.58 | pfn | although, some servers will detect and print out a message |
17:43.16 | pfn | tomcat may or may not... |
17:43.33 | quadmasta | it tries to make Firefox download something it says is an xoctet-stream |
17:43.39 | quadmasta | so I'm figuring it wont' work |
17:43.45 | quadmasta | that's better |
17:43.53 | quadmasta | crap, I can't change my nick |
17:44.25 | quadmasta | pfn: is there a way to secure just a particular servlet? |
17:44.45 | pfn | use <security-constraint> |
17:45.25 | pfn | <user-data-constraint><transport-guarantee> |
17:45.37 | pfn | google on any/all of those terms will get you the info you need |
17:45.42 | quadmasta | from what I've read online that seems to use a security realm, of which I only have one and it's already in use |
17:46.05 | pfn | what do you mean by "secure a particular servlet" |
17:46.16 | quadmasta | I have one servlet that needs to run SSL |
17:46.16 | pfn | you can enforce transport-guarantee without requiring a realm |
17:46.27 | pfn | that's generally a bad idea |
17:46.35 | pfn | because other confidential information *will* be in plaintext |
17:46.36 | quadmasta | explain please |
17:46.37 | pfn | e.g. session id |
17:47.24 | quadmasta | even if that one servlet handles everything that the user will ever need access to? |
17:47.38 | pfn | then you might as well apply the security constraint to the entire webapp |
17:48.03 | quadmasta | okay, I think I should explain the webapp a bit more |
17:48.24 | pfn | if that servlet already handles everything, there's no sense in differentiating |
17:48.32 | quadmasta | but there's more than just that servlet |
17:48.44 | pfn | then it's a bad idea |
17:48.52 | pfn | as soon as you need to use session ids, etc. |
17:48.56 | quadmasta | there's an admin portal which doesn't need to be secure and a customer portal which does need to be secure |
17:49.29 | quadmasta | all of the admin servlets are mapped to /admin/ and customers servlets are mapped to /customer/ |
17:50.02 | pfn | so put a security constraint on /customer |
17:51.17 | quadmasta | but if I use form based authentication, won't I need a security realm? |
17:51.25 | pfn | no |
17:51.29 | pfn | you don't specify an auth-constraint |
17:51.35 | pfn | just user-data-constraint |
17:54.45 | quadmasta | so if someone hits serverhttp://customer/index.jsp it'll force them to https? |
17:55.22 | pfn | yes |
17:55.57 | quadmasta | actually, it gives me 403 errors |
17:56.02 | quadmasta | I must have done something incorrectly |
17:56.10 | *** join/#tomcat [2]Zack1403 (n=Zack1403@64.122.14.80) |
18:01.56 | quadmasta | http://rafb.net/p/QsayeT85.html |
18:02.02 | quadmasta | what did I do incorrectly? |
18:20.06 | *** join/#tomcat rambo123 (n=rocky@59.94.99.160) |
18:32.20 | pfn | sounds like it should work ok |
18:32.27 | pfn | what happens if you go to https? |
18:32.42 | *** join/#tomcat tadamski (n=tadamski@74.94.194.205) |
18:33.32 | tadamski | I'm using tomcat 5.5.9, java1.6, running windows xp, is there a way to test the timezone tomcat is using? |
18:33.45 | pfn | it uses the default system timezone |
18:33.52 | pfn | which correlates into the default java timezone |
18:37.13 | tadamski | this is probably an obvious question...bu I guess to change the timezone tomcat uses I would need to change the timezone java is using |
18:37.33 | tadamski | I can't just define a setting in a tomcat config file |
18:37.48 | pfn | that's right |
18:37.52 | pfn | you set the timezone in the system |
18:38.46 | tadamski | alright, thank you for your help |
18:42.09 | quadmasta | pfn, I think it's the stupid Jetty plugin that's causing the issue |
18:42.25 | pfn | jetty plugin? heh, what does that have to do with tomcat? |
18:43.12 | quadmasta | nothing :) |
18:43.33 | quadmasta | I thought we were using the cargo plugin to deploy to tomcat but it's jetty |
18:43.36 | quadmasta | testing in Tomcat now :) |
18:44.29 | quadmasta | what do ya know, it works in Tomcat |
18:49.38 | *** join/#tomcat dryrot (i=10539@tsunami.OCF.Berkeley.EDU) |
18:50.00 | dryrot | so i have successfully gotten tomcat to work on port 80... i would like to use HTTPS and port 443. how do i do that? |
18:50.28 | dryrot | (tomcat 6) |
18:58.00 | quadmasta | thanks for the help pfn |
18:58.15 | quadmasta | dryrot: search SSL enable Tomcat |
19:05.07 | *** join/#tomcat Neo0 (n=t7DS@200-170-149-242.xf-static.ctbcnetsuper.com.br) |
19:12.39 | *** part/#tomcat Neo0 (n=t7DS@200-170-149-242.xf-static.ctbcnetsuper.com.br) |
19:14.27 | *** join/#tomcat Neo0 (n=t7DS@200-170-149-242.xf-static.ctbcnetsuper.com.br) |
19:27.21 | *** part/#tomcat quadmasta (n=bdcribbs@74.92.194.125) |
19:29.54 | *** part/#tomcat Neo0 (n=t7DS@200-170-149-242.xf-static.ctbcnetsuper.com.br) |
19:33.07 | *** join/#tomcat gverig (n=mikle@12.44.117.104) |
19:34.01 | gverig | Tomcat was able to read v0 cookies with |
19:34.51 | gverig | dang... with '=' signs in it. 5.5.15 still works, 5.5.26 does not. Was this accidental side effect of clean up or was this done on purpose to be 'standards compliant'? |
19:45.47 | *** join/#tomcat {Chris} (n=cdupont@LMontsouris-152-61-3-84.w80-13.abo.wanadoo.fr) |
19:59.45 | *** join/#tomcat JuanCraft (n=deeede@sistemas.uis.edu.co) |
20:12.16 | *** join/#tomcat Nopik (n=chatzill@83.142.73.113) |
20:12.23 | Nopik | hi there |
20:14.10 | Nopik | I have quite complex application, which at startup (in contextInitialized()) tries to log in into some remote service.. sometimes this login fails, I would like to destroy my context and stop application.. how to do that? |
20:15.04 | *** join/#tomcat plexq (n=aturner@70.91.39.126) |
20:15.25 | plexq | Please help - Tomcat isn't picking up changes to my web.xml, and I can't figure out why or how to fix it |
20:16.02 | plexq | oh shit - I think I figured it out |
20:16.08 | plexq | It's eclipse screwing up my web.xml |
20:24.18 | *** join/#tomcat Zack1403 (n=Zack1403@64.122.14.80) |
20:26.25 | *** join/#tomcat _hen (n=hen@c-67-183-122-76.hsd1.wa.comcast.net) |
20:26.53 | *** part/#tomcat rambo123 (n=rocky@59.94.99.160) |
20:39.36 | *** join/#tomcat iostream (n=iostream@89.221.166.136) |
20:40.27 | *** join/#tomcat [1]Zack1403 (n=Zack1403@64.122.14.80) |
20:54.20 | *** join/#tomcat [2]Zack1403 (n=Zack1403@64.122.14.80) |
20:54.47 | *** join/#tomcat pierz (n=pierz@213.151.171.220) |
20:55.39 | pierz | Hello, where to put individual context file for servlet in tomcat 6, $TOMDIR/conf/ ? |
20:55.52 | *** join/#tomcat Zack1403 (n=Zack1403@64.122.14.80) |
21:21.01 | *** join/#tomcat [1]Zack1403 (n=Zack1403@64.122.14.80) |
21:44.23 | *** join/#tomcat Zack1403 (n=Zack1403@64.122.14.80) |
22:28.40 | *** part/#tomcat dryrot (i=10539@tsunami.OCF.Berkeley.EDU) |
22:38.18 | *** join/#tomcat cyth (n=cythrawl@c-98-212-147-21.hsd1.il.comcast.net) |
22:38.45 | cyth | hi I'm wanting to set some custom http headers on a few static files, what's the best way to do this? |
22:38.46 | cyth | create a filter? |
23:00.45 | jasonb | cyth: Yes. |
23:02.18 | cyth | jasonb, okay was kinda wondering if there wasn't some setting of the connector or web.xml one could do, but it seems a filter makes a lot of sense |
23:02.29 | cyth | just kinda like... wait... I have to write this? |
23:02.42 | jasonb | heh |
23:03.29 | jasonb | There should probably be a Tomcat valve for doing that.. but there isn't. |
23:04.56 | cyth | yeah it's not a huge deal, just something one would think would be trivial but common enough that there would be a configuration somewhere. |
23:07.02 | jasonb | Yup. It is. |
23:07.18 | cyth | heh just found that you can configure compression pretty easily with coyote, that's nice |
23:08.16 | jasonb | Yes, and it's a little strange that it isn't configured that way by default in the stock server.xml. |
23:21.24 | pfn | enabling compression by default is a bad idea if your app is processing heavy |
23:21.49 | pfn | it's a tradeoff, of course, network latency vs. processing latency |
23:22.05 | pfn | in a hosted environment, you'll probably run into the latter before the former (for many types of content) |