| 00:00.30 | goudkov | so i would have to write new tags, set up a separate application and then hope that i'll be able to reproduce the problem. |
| 00:00.38 | jasonb | Now you see how these kinds of bugs tend to stay unfixed. |
| 00:00.49 | goudkov | yep :) |
| 00:01.35 | goudkov | if i knew tomcat code base, i would just look through specific parts of code. it's quite often you can find a problem if you know what you are looking for. |
| 00:01.45 | goudkov | but it would take me more time just to figure out what's where. |
| 00:03.15 | jasonb | There's a lot of code in there. |
| 00:03.19 | jasonb | (even knowing the source) |
| 00:16.00 | goudkov | well, i as i secretly expected :( |
| 00:16.07 | goudkov | i couldn't reproduce it with a simple test case |
| 00:16.31 | goudkov | so either there is some really weir bug in my code, or the problem is with tomcat, but only occurs in more complex situations. |
| 00:19.26 | jasonb | Yeah. The worst part is: you don't know whether it's an obscure Tomcat bug or not. If it is, and if you could narrow it down, it could get fixed, which would positively affect lots of other Tomcat users / web sites. |
| 00:20.57 | goudkov | i think i know what it is :) |
| 00:21.07 | goudkov | it has something to do with some body tags returning SKIP_BODY. |
| 00:21.19 | goudkov | in those cases it looks like bodyContent isn't handled right |
| 00:21.30 | goudkov | ...or my understanding of the specs is wrong. |
| 00:21.39 | goudkov | hold on |
| 00:22.04 | goudkov | so far, i have it narrowed down to cases with body tags that return SKIP_BODY for doStartTag. |
| 00:22.16 | goudkov | if all of them always return EVAL... then everything seems fine. |
| 00:22.27 | goudkov | if some of them sometimes return SKIP_BODY then i begin to see problems. |
| 00:23.53 | goudkov | yep |
| 00:23.58 | goudkov | i'm da man! |
| 00:24.05 | goudkov | now i need to go re-read the specs. |
| 00:24.12 | goudkov | maybe tomcat does what it's supposed to do :) |
| 00:24.46 | jasonb | It's always possible. |
| 00:27.31 | goudkov | "If SKIP_BODY is returned the body, if present, is not evaluated." |
| 00:27.34 | goudkov | hmm |
| 00:27.58 | goudkov | so if you have a body tag which might return SKIP_BODY in doStartTag depending on some condition |
| 00:28.11 | goudkov | how would you know in doEndBody that the body was in fact skipped. |
| 00:28.52 | jasonb | Good question. It's been a while since I wrote a tag.. so I can't remember. |
| 00:29.06 | jasonb | But, if its your class, you can maintain your own state variables.. at the very least. |
| 00:30.21 | goudkov | ok, now i'm positive. |
| 00:30.24 | goudkov | i just reproduced it. |
| 00:31.05 | goudkov | if i have a tag that might iterate 0 or more times and it's included in the page several times. if the first instance doesn't iterate (returns SKIP_BODY) then the problems occur. |
| 00:31.16 | goudkov | but if all tags iterate at least once, then everything is ok. |
| 00:31.30 | goudkov | and it definitely has something to do with bodyCount |
| 00:31.37 | goudkov | but not with pooling or race conditions. |
| 00:31.41 | goudkov | bodyContent i mean |
| 00:32.14 | goudkov | i'm not sure how tomcat handles the stream in case of SKIP_BODY, but something isn't right there. |
| 00:32.35 | goudkov | and there is absolutely nothing in the jsp specs about doing any extra work in case of returning SKIP_BODY. |
| 00:32.40 | goudkov | at least i couldn't find anything. |
| 00:32.42 | jasonb | I do seem to remember reading some mailing list messages about SKIP_BODY behavior, and some confusion around that. |
| 00:33.47 | jasonb | So then it's getting easier for you to reproduce it? |
| 00:34.45 | goudkov | yep |
| 00:36.29 | jasonb | Very cool. |
| 00:37.34 | goudkov | there is definitely a different behavior as far as bodyContent goes |
| 00:37.49 | goudkov | depending on whether there was a tag that didn't iterate or not. |
| 00:38.02 | goudkov | but not sure if it's the cause of the problem or just a coincidence |
| 00:40.27 | goudkov | and if i have two tags and neither one iterates, then it's all fine as well |
| 00:41.24 | goudkov | and if two tags, the first iterates the second doesn't. no problem. |
| 00:41.36 | goudkov | but with two tags and first not iterating and the second one iterating, i get errors. |
| 00:43.52 | jasonb | Neat. Sounds like you've got a pretty clear picture of it now. |
| 00:44.21 | jasonb | And, that sounds easy enough to write a test case that is reproducible. |
| 00:44.32 | goudkov | yep, i'm still narrowing it down though. |
| 00:44.51 | jasonb | I have to head to dinner, but will be back in an hour or two. |
| 00:45.00 | jasonb | bbiab. |
| 02:29.06 | *** join/#tomcat terje (n=terje@vampira.scare.org) |
| 03:45.06 | *** join/#tomcat terje_ (n=terje@vampira.scare.org) |
| 04:11.07 | *** join/#tomcat bhagat (n=bhagat@203.129.255.188) |
| 04:14.22 | *** join/#tomcat acidjnk (i=acid@pD950A59F.dip0.t-ipconnect.de) |
| 04:30.43 | *** join/#tomcat McGo|Einkaufen (n=McGo@dyndsl-091-096-105-040.ewe-ip-backbone.de) |
| 04:55.56 | *** join/#tomcat tess_ (n=tess@c-24-130-52-152.hsd1.ca.comcast.net) |
| 05:01.59 | *** join/#tomcat bugz_ (n=bugz@76.194.247.21) |
| 05:10.41 | *** join/#tomcat tess_ (n=tess@c-24-130-52-152.hsd1.ca.comcast.net) |
| 05:22.10 | *** join/#tomcat rogue780 (n=rogue780@c-68-34-234-213.hsd1.md.comcast.net) |
| 05:25.23 | *** join/#tomcat rogue780 (n=rogue780@c-68-34-234-213.hsd1.md.comcast.net) |
| 05:28.59 | *** join/#tomcat rrowell (n=von_neum@c-71-205-132-208.hsd1.mi.comcast.net) |
| 07:20.58 | *** join/#tomcat tess (n=tess@c-24-130-52-152.hsd1.ca.comcast.net) |
| 08:03.41 | *** join/#tomcat dvayanu (n=another@ppp-93-104-108-167.dynamic.mnet-online.de) |
| 08:31.26 | *** join/#tomcat Serge_HID (i=Serge_HI@93.185.198.43) |
| 08:54.27 | *** join/#tomcat siliconmeadow (n=ricardo@80.194.82.66) |
| 08:54.52 | *** part/#tomcat siliconmeadow (n=ricardo@80.194.82.66) |
| 09:00.09 | *** join/#tomcat Thijssss (i=thijssss@op.tsip.nl) [NETSPLIT VICTIM] |
| 09:34.08 | *** join/#tomcat uwe_guenther (n=Notebook@p5B284D47.dip.t-dialin.net) |
| 09:44.28 | *** join/#tomcat bogor (i=3b5c3f02@gateway/web/freenode/x-phrmejwdridifewy) |
| 09:45.11 | *** join/#tomcat bhagat (n=bhagat@203.129.255.188) |
| 09:52.59 | *** join/#tomcat al (i=d@niel.cx) |
| 10:21.04 | *** join/#tomcat McGo (n=McGo@dyndsl-091-096-105-040.ewe-ip-backbone.de) |
| 11:04.26 | *** join/#tomcat \\localhost (n=david@gw1.montecristogames.com) |
| 11:14.55 | *** join/#tomcat dvayanu (n=another@ppp-93-104-108-167.dynamic.mnet-online.de) |
| 11:45.22 | *** join/#tomcat mturk (n=mturk@40-237.dsl.iskon.hr) |
| 12:09.52 | *** join/#tomcat bhagat (n=bhagat@203.129.255.188) |
| 12:26.43 | *** join/#tomcat Techdeck (n=Techdeck@bzq-84-110-152-136.red.bezeqint.net) |
| 12:27.16 | Techdeck | Hello, I have tomcat 5.5 and an SSL certificate in the key store. It seems the SSL certificate is expired, I renewed it in godaddy but I have no idea how to renew it in the keystore, any ideas anyone? |
| 12:47.57 | *** join/#tomcat acidjnk (i=acid@pD950A59F.dip0.t-ipconnect.de) |
| 13:07.32 | *** join/#tomcat mturk (n=mturk@47-110.dsl.iskon.hr) |
| 13:35.29 | *** join/#tomcat |8-D (i=0fc3b9e1@gateway/web/freenode/x-rnbxwqzrhnojsbgo) |
| 14:27.01 | *** join/#tomcat Floyd-ATC (i=c1479443@gateway/web/freenode/x-ujijrtgpmfugzcez) |
| 14:38.30 | gothos | Hey, I'm trying to write a new tomcat manager application, but can't get the associated Host data |
| 14:39.02 | *** join/#tomcat randrewj (n=u1@dyn-128-59-53-3.dyn.columbia.edu) |
| 14:40.03 | gothos | I'm using the play framework, so I am not getting any Servlet stuff. is there some other way to get that data? |
| 14:40.33 | *** join/#tomcat acarbs12 (n=acarbone@173.9.89.1) |
| 14:44.10 | *** join/#tomcat sfire||mouse (n=cwevans@adsl-71-156-55-120.dsl.irvnca.sbcglobal.net) |
| 14:51.20 | sfire||mouse | I'm trying to track down where/what code gets executed for (http://domain/servlet/foo.BarBarn) for Tomcat 4.1.31, java 1.5.0_17-b04, and solaris 10. I've checked conf/server.xml, and found the host entry for it. There are no Contexts that mention servlet. |
| 14:51.35 | sfire||mouse | Do you have any advice on where I should look |
| 14:58.10 | sfire||mouse | may have found stuff |
| 14:58.53 | sfire||mouse | Given http://domain/servlet/foo.BarBarn does it make sense the .class is foo/WEB-INF/classes/foo/BarBarn.class? |
| 14:59.35 | *** part/#tomcat randrewj (n=u1@dyn-128-59-53-3.dyn.columbia.edu) |
| 15:01.56 | *** join/#tomcat randrewj (n=u1@dyn-128-59-53-3.dyn.columbia.edu) |
| 15:10.48 | *** join/#tomcat uwe_guenther (n=Notebook@p5B284D47.dip.t-dialin.net) |
| 15:59.33 | *** join/#tomcat dvayanu (n=another@ppp-93-104-108-167.dynamic.mnet-online.de) |
| 16:04.03 | *** join/#tomcat WarDragon (n=mm@unaffiliated/war2) |
| 16:24.38 | *** join/#tomcat dob1 (n=dob@host-78-13-164-164.cust-adsl.tiscali.it) |
| 16:25.37 | dob1 | hi, i am new of servlet container/application server and programming java servlet etc, i was just wondering if tomcat can manage for me a connection pool and how can i use it from my servlet, if someone can point me to some tutorial/documentation about this |
| 17:09.25 | *** join/#tomcat tess_ (n=tess@nat.800westelcamino.lindenlab.com) |
| 17:11.41 | *** join/#tomcat Alagar (n=Administ@122.164.37.4) |
| 17:14.35 | *** join/#tomcat tess_ (n=tess@nat.800westelcamino.lindenlab.com) |
| 17:16.35 | *** join/#tomcat Alagar (n=Administ@122.164.37.4) |
| 17:22.06 | *** join/#tomcat cofeineSunshine (n=justinas@kas.tavatievs.eu) |
| 17:23.38 | *** join/#tomcat chrisjw (n=dvtt@g228211219.adsl.alicedsl.de) |
| 18:09.56 | *** join/#tomcat NilsBo (n=cczwei@ip-88-153-128-172.unitymediagroup.de) |
| 18:11.10 | *** join/#tomcat clajo04_ (n=clajo04_@74.72.55.57) |
| 18:31.01 | *** join/#tomcat sfire||mouse (n=cwevans@adsl-71-156-55-120.dsl.irvnca.sbcglobal.net) |
| 18:32.38 | *** join/#tomcat relachs (n=relachs@g229219160.adsl.alicedsl.de) |
| 19:00.13 | *** join/#tomcat Alagar (n=Administ@122.164.37.4) |
| 19:01.12 | *** join/#tomcat Alagar1 (n=Administ@122.164.37.4) |
| 19:01.39 | *** join/#tomcat jasonb (n=jasonb@m410536d0.tmodns.net) |
| 19:03.05 | *** join/#tomcat ruben23 (n=RPL@122.55.48.243) |
| 19:03.38 | *** join/#tomcat Alagar2 (n=Administ@122.164.32.44) |
| 19:04.26 | ruben23 | hi nayone can help i got this problem when i run up tomcat--->http://pastebin.com/m3768ef60 |
| 19:04.36 | ruben23 | im suign an ubuntu distro |
| 19:06.18 | ruben23 | hi anyone |
| 19:08.52 | *** join/#tomcat Alagar3 (n=Administ@122.164.32.44) |
| 19:09.32 | *** join/#tomcat Alagar2 (n=Administ@122.164.32.44) |
| 19:26.04 | cofeineSunshine | ruben23: export JAVA_HOME=/usr/lib/jvm/java-6-sun |
| 19:26.16 | cofeineSunshine | add this line to ~/.bashrc |
| 19:26.43 | ruben23 | cofeineSunshine:yes, thanks |
| 19:28.57 | *** join/#tomcat Alagar (n=Administ@122.164.32.44) |
| 19:33.03 | *** join/#tomcat ibot (i=ibot@rikers.org) |
| 19:33.03 | *** topic/#tomcat is Stable versions: 6.0.20, 5.5.28, 4.1.40. Newbies use the official binary from tomcat.apache.org, or an RPM package from http://www.webdroid.org/archives/tomcat-package. Check your Tomcat logs before you ask for an answer. SLOW MOTION CHANNEL (we all have jobs & kids): Ask your question including your TC, Java, & OS versions, then wait; check back often for answers.. Use mod_proxy_ajp, not mod_jk |
| 20:01.46 | *** join/#tomcat saml (n=sam@h-66-134-168-26.nycmny83.static.covad.net) |
| 20:02.10 | saml | hola, how do I kill tomcat? shutdown.bat only unloads app. jvm is still running |
| 20:07.18 | *** join/#tomcat dvayanu (n=another@ppp-93-104-108-167.dynamic.mnet-online.de) |
| 20:15.25 | *** join/#tomcat hesco (n=hesco@c-24-99-160-121.hsd1.ga.comcast.net) |
| 20:15.40 | *** join/#tomcat jasonb (n=jasonb@dsl027-180-244.sfo1.dsl.speakeasy.net) |
| 20:19.34 | *** part/#tomcat randrewj (n=u1@dyn-128-59-53-3.dyn.columbia.edu) |
| 20:24.47 | hesco | Tomcat: 6.0.16-0, Java: 1.4.2, OS: Centos 5.4 -- I was asked to figure out why a tomcat application is not writing to catalina.out. Its running on a staging server where I was asked to replicate our production server. I've never written any java and while I've installed it a few times, I have no experience administering it. Where would I find a good primer on this, and can anyone here please share a pointer on my immediate issue of the |
| 20:24.47 | hesco | webapps/our_app not writing to catalina.out? |
| 20:32.14 | *** join/#tomcat JeffATL (n=nouser@173-102-152-59.pools.spcsdns.net) |
| 20:33.07 | JeffATL | I need some docs/guidance as to how to connect an app under tomcat 5.5 on linux to an external DB2 server |
| 20:44.45 | *** join/#tomcat cofeineSunshine (n=justinas@kas.tavatievs.eu) [NETSPLIT VICTIM] |
| 20:44.45 | *** join/#tomcat trochej (i=trochej@noches.pl) [NETSPLIT VICTIM] |
| 20:44.45 | *** join/#tomcat wltjr (n=wltjr@unaffiliated/wltjr) [NETSPLIT VICTIM] |
| 20:44.45 | *** join/#tomcat lyken (n=dc@67-231-241-244.turnkeyinternet.net) [NETSPLIT VICTIM] |
| 20:44.45 | *** join/#tomcat ohsix (i=ohsix@66.220.111.99) |
| 20:46.53 | JeffATL | and in that context (tomcat 5.5 and DB2), what's typically used these days for connection pooling? |
| 21:07.11 | JeffATL | fwiw, do apps written for a 5.5 tomcat branch port cleanly to a 6.0 branch? |
| 21:21.17 | jasonb | JeffATL: Almost always. |
| 21:22.18 | jasonb | hesco: Tomcat 6.0.x cannot run on Java 1.4.2. |
| 21:22.49 | JeffATL | jasonb: what if the app is written for an - ahem - 5.0 branch? |
| 21:24.28 | jasonb | JeffATL: Then at least moving it to 5.5 or 6.0 it will have a well-behaved, non-broken container to run in. It should run better than ever. |
| 21:25.41 | jasonb | JeffATL: I documented configuration file migration steps from 4.1 to 5.0 to 5.5 to 6.0 in our book Tomcat: The Definitive Guide, 2nd Edition (O'Reilly). That might help you. |
| 21:27.16 | JeffATL | jasonb: wow - thanks! |
| 21:27.29 | jasonb | JeffATL: You're welcome. |
| 21:29.56 | JeffATL | as a side note - i'm trying to figure out what my mgt means by "version 6 of JVM" - I come at this from the server-build angle, not a web-devel angle so I'm not finding a puzzle piece that corresponds to anything like that |
| 21:30.59 | JeffATL | for instance, I've got a server with tomcat 5.5.27 and sun-jdk of 1.5.0.15 |
| 21:37.34 | JeffATL | jasonb: in your view, is it even adviseable to take an app from 5.0 to 6.0 in one jump w/o doing at least a little testing in 5.5? |
| 21:38.00 | jasonb | Well, there's a Java "6" (a version that Sun marketing talks about), which corresponds to Java 1.6.x. |
| 21:38.32 | jasonb | JeffATL: Yes, it makes the most sense to go straight to 6.0 without ever touching 5.5. |
| 21:38.38 | JeffATL | jasonb: ok, i wondered if that might be what's going on |
| 21:39.22 | jasonb | JeffATL: If for some reason it ends up having insurmountable trouble on 6.0 (you should ask about it in here, if you think so), then you can always fall back to 5.5 as a last resort, but I still wouldn't advise it. |
| 21:42.51 | JeffATL | jasonb: so you co-wrote *Tomcat: The Definitive Guide*, 2ed.? |
| 21:43.21 | jasonb | JeffATL: Yes. I co-authored both editions. |
| 21:44.14 | jasonb | JeffATL: I'm the only one who worked on the second edition, though. It's a derivative of the first edition, so both of us have some content in there. |
| 22:10.29 | *** join/#tomcat ech0s7 (n=ech0s7@host229-90-dynamic.57-82-r.retail.telecomitalia.it) |
| 22:10.30 | ech0s7 | hi |
| 22:11.08 | ech0s7 | i'm doing a forward from Servlet to Jsp, why the browser rewrite the url ? in teory the browser is completely unaware that it has taken place, so its original URL remains intact, why this not happens ? |
| 22:12.02 | Primer | you're doing it wrong |
| 22:12.59 | ech0s7 | Primer: i'm doing in this way |
| 22:12.59 | ech0s7 | RequestDispatcher dispatcher = getServletContext().getRequestDispatcher(toPage); |
| 22:13.00 | ech0s7 | dispatcher.forward(request,response); |
| 22:13.17 | ech0s7 | request is HttpServletRequest |
| 22:13.24 | ech0s7 | response is HttpServletResponse |
| 22:13.26 | Primer | yeah, the browser should never see that |
| 22:13.28 | ech0s7 | toPage is String |
| 22:13.34 | ech0s7 | Primer: i'm tring |
| 22:13.46 | ech0s7 | and i have seen that the browser changes url |
| 22:13.55 | ech0s7 | this happens because i'm in localhost ? |
| 22:15.36 | ech0s7 | Primer: would you seen a testcase ? |
| 22:16.56 | Primer | sorry, I don't actually like java or jsp |
| 22:17.41 | Primer | all I know is that I had to do this once before, and I noticed that the request caused the code to just jumped internally to the other page, and the browser was none the wiser |
| 22:19.48 | JeffATL | so what's the story with Sun's "marketroid" java version numbering? |
| 22:24.27 | *** join/#tomcat relachs (n=relachs@f055240022.adsl.alicedsl.de) |
| 22:30.11 | JeffATL | found answer: http://technologizer.com/2009/07/14/version-numbers/ |
| 22:33.40 | JeffATL | i need a little sense of scale. for a run-of-the-mill DB2-backed java app on tomcat, should a single 32-bit 3GHz 2GiB-RAM Xeon server support a few hundred (say, up to 1000) users at once? |
| 22:41.36 | *** join/#tomcat terje (n=terje@vampira.scare.org) |
| 22:44.05 | jasonb | JeffATL: It depends on a number of things that we don't know. Does DB2 run on this 2G RAM box also? Is the webapp a very large webapp? How much load does each of the 1000 users create on the box? How fast do the requests complete, on the average? It's pretty hard to guess, in the end. 2G RAM seems like it may be a little low for 1000 concurrent users. |
| 22:44.37 | JeffATL | jasonb: no; different (and big IBM P5) system |
| 22:45.42 | JeffATL | jasonb: i agree re the ram...i don't think it would be characterized as a "very large" app...it's a loan-application/granting/mgt app |
| 22:51.09 | *** join/#tomcat km (n=km@66-7-254-75.static-ip.telepacific.net) |
| 22:51.34 | *** part/#tomcat JeffATL (n=nouser@173-102-152-59.pools.spcsdns.net) |
| 23:04.32 | *** join/#tomcat morgajel_ (n=morgajel@d192-24-135-7.try.wideopenwest.com) |