inserting handler/filter in main server vs. virtual host configuration?

For httpd.conf in Apache 2.x / mp2, does it matter if I insert a handler in the main server section vs. a virtual host section? My question pertains specifically to performance/memory, as I already understand the logical implications. PerlChildInitHandler +MyStuff::InitHandler vs. PerlChi

Re: mp2 shared vs. static build - pros & cons?

That's helpful information... I guess I'll stick with the DSO. At one point I did benchmark mod_perl2 with vs. without threads, and without threads was marginally faster. I've since started using some new CPAN modules, so I'll have to double-check that none of them use Perl threads. thanks! JB

mp2 shared vs. static build - pros & cons?

In anticipation of upgrading my mp2 build to the latest/greatest, I would like to consider switching from a DSO build to a static build. Could folks here offer their opinions on pros/cons in terms of performance/memory, etc? FYI my environment is httpd 2.0.55, prefork MPM, Perl 5.8.6 (w/threads),

[mp2] pnotes-like facility for $c connection object?

I've made great use of $r->pnotes, and now finding myself in need of a similar way to stash objects within the connection object. While there are regular 'notes' offered by the connection ($c->notes), unfortunately there is no 'pnotes'. Any suggestions for a workaround? The requirement is as fol

Re: bucketbrigades with html filter

I think the only conclusive answer is to try both and benchmark them... there may be some generalizations one can make about memory, hook/subroutine invocation overhead, etc., but ultimately you need to test both methods to see which is faster. And please share your findings if you decide to do t

Re: bucketbrigades with html filter

Yikes indeed :) I should have clarified that in our app we don't actually process embedded tags. Our app lets users mangle the source HTML using RegEx, and since users can (and often do) perform filtering like s/()(.*?)(<\/body>)/$1$2<\/center>$3/, we need to buffer it all. The main benefit I se

Re: bucketbrigades with html filter

The way to deal with this is to buffer as much content as you need (maybe the whole page) and then do your work on the buffer. Our application (HTTP output filter) buffers the stream as we go and stores it on the filter context ($f->ctx) across filter invocations. When we've seen $f->eos, then we

Re: API to mime.types?

Ok, that worked... my only gripe is that it doesn't correctly parse the type for a WinZip (.zip) file! Oh well... thanks :) JB On 9/23/05, John ORourke <[EMAIL PROTECTED]> wrote: > Geoffrey Young wrote: > > http://search.cpan.org/dist/File-MMagic-XS/

API to mime.types?

Through my MP2 handler, I need to be able to set Content-Type based on the extension of any file that I'm reading from disk and then writing to the client. I'd like to somehow (programmatically) map from file extension to MIME type, and it seems like Apache's mime.types file is a good place to do

Re: httpd consumes 220mb of RAM when printing alot of output

M as the size of the table! (?!) I suppose a workaround would be to do multiple select's for different key ranges, and print out each subset iteratively. JB On 9/22/05, Malcolm J Harwood <[EMAIL PROTECTED]> wrote: > On Thursday 22 September 2005 11:06 am, Jeff Ambrosino wrote: &g

httpd consumes 220mb of RAM when printing alot of output

I'm exporting a database table through a mod_perl2 handler. The problem is that for large tables, the size of the httpd process balloons to consume alot of RAM. For example, a 299mb MySQL table (size of .MYD file), which creates a 35mb export, causes httpd to consume about 220mb of RAM! My code

modperl_filter.c bug causing hung httpd child? (w/backtraces)

I've observed recurring hung Apache children in my mod_perl http output filter. The problem is intermittent, and while I can't cause it to happen, our site gets alot of traffic and the problem is happening 10+ times per day. The immediate impact is that the page requested never gets served to the

Re: hanging CPU-sucking httpd procs that say "..reading.."

some backtraces and then head over to http-dev and/or mod_ssl lists. Any last-ditch advice from MP folks would be welcome, but not expected since this doesn't appear to be a MP/Perl issue (although it could be libapreq since I believe mod_ssl uses that.) thanks Jeff On 9/8/05, Jeff Ambrosino

Re: hanging CPU-sucking httpd procs that say "..reading.."

n http output filter that processes the proxied response (and unfortunately the mod_proxy list is basically dead). thanks JB On 9/8/05, Jeff Ambrosino <[EMAIL PROTECTED]> wrote: > Hi mod_perl folks, > > I'm getting occasional hanging httpd children (prefork) that suck up >

hanging CPU-sucking httpd procs that say "..reading.."

Hi mod_perl folks, I'm getting occasional hanging httpd children (prefork) that suck up alot of CPU and bog down my server. When I inspect Apache Status (ExtendedStatus On), I see something like this: 434-3 11031 0/29/612 R 0.49 1341 0 0.0 0.20 4.09 ? ? ..reading.. There's no client IP, no vi

Re: get # of response bytes sent for large/streaming content?

Yup, that works great. thanks! JB On 9/1/05, Geoffrey Young <[EMAIL PROTECTED]> wrote: Jeff Ambrosino wrote:> I'm running a mod_perl cleanup handler from a reverse proxy> ($r->pool->cleanup_register), and> I need to know the total # of bytes sent in a response so I

get # of response bytes sent for large/streaming content?

I'm running a mod_perl cleanup handler from a reverse proxy ($r->pool->cleanup_register), and I need to know the total # of bytes sent in a response so I can store this info in a database. A real-world example would be the end user (browser) streaming a Quicktime movie, and I want to know how long

delete a specific Set-Cookie header within a proxied response?

We have the need to delete a specific Set-Cookie header found in the reponse within a reverse proxy [Apache2, libapreq2, mod_perl2, mod_proxy with an HTTP output filter]. Looking at the docs for APR::Table [1] it doesn't appear that this is possible. Although you can "unset" and "delete" based on

Re: how to trap connection reset/abort in Apache::Filter?

Ok, this is going to be tough for me to test because I'm still on RC4, so it's not yet converted to the new Apache2 naming. I'll still try to make/test the svn snapshot, though (which also gives me a good reason to get more familiar with subversion.) I'll email what I find out. thanks Stas! Jef

Re: Graphics / Pie Charts

If, by chance, you also have use for a general purpose templating system, I've had great luck with Template-Toolkit (www.template-toolkit.org.) Among [many] other things, it embeds GD so you can write little templates that slurp in the data and produce graphics with graphs, charts, etc. The learn

Re: advice needed: mod_perl reverse proxy

I've had some similar issues, but gotten around them by using erro_headers_out. This worked for me, so you might give it a try: $f->r->headers_out->unset('Content-Length'); $f->r->err_headers_out->set('Content-Length' => length($NewBody)); Obviously "$NewBody" is whatever you're

Re: how to trap connection reset/abort in Apache::Filter?

the error by myself, I know when our scanning service will do so, and I could put some additional code in place to watch for it. If there are additional debug/traps that you could suggest, I'd like to try. Any ideas? thanks Jeff On 5/2/05, Stas Bekman <[EMAIL PROTECTED]> wrote: &g

Re: Activating/decativating PerlOutputFilterHandler "on-the-fly"

>From within an existing Perl handler/filter, you can dynamically insert ('register') a new filter with the add_output_filter method on the request object: $r->add_output_filter(\&callback); And, you can remove a downstream filter (from within an upstream filter) with: $next_f = $f->next; $next_

Re: how to trap connection reset/abort in Apache::Filter?

Ok, I tried this approach, and it only partially worked; unfortunately my httpd procs are still dying, and APR::Error doesn't seem to pass anything. Here's the error I got _before_ wrapping $f->print in the eval: [Sun May 01 05:12:21 2005] [error] [client X.X.X.X] Apache::Filter::print: (104

Re: how to trap connection reset/abort in Apache::Filter?

Actually, it looks like using APR::Error [1] and wrapping $f->print in an eval would do the trick...? Jeff [1] http://perl.apache.org/docs/2.0/api/APR/Error.html On 5/1/05, Jeff Ambrosino <[EMAIL PROTECTED]> wrote: > [Sun May 01 05:12:21 2005] [error] [client X.X.X.X] > Apache

how to trap connection reset/abort in Apache::Filter?

I've got an HTTP output filter running on mp2-RC4 + httpd2.0.54 + libapreq2-2.04_03-dev, and occasionally I see these messages in my error_log: [Sun May 01 05:12:21 2005] [error] [client X.X.X.X] Apache::Filter::print: (104)Connection reset by peer at /var/httpd/lib/perl/MyOutputFilter.pm line 453

Re: setting http output filter priority?

I'm logging some closure to this one... I was able to get this to work by modifying the priorities of mod_perl/src/modules/perlmod_perl.c's http output filter to AP_FTYPE_CONTENT_SET-1, and then httpd-2.0.53/modules/experimental/mod_cache.c's cache_save and cache_out filter to AP_FTYPE_CONTENT_SET

Re: setting http output filter priority?

>From looking at the code, it appears that I may need to modify mod_perl and either of mod_cache or mod_deflate: mod_perl.c - line ~839 -- ap_register_output_filter(MP_FILTER_REQUEST_OUTPUT_NAME, MP_FILTER_HANDLER(modperl_output_fil

Re: setting http output filter priority?

Thanks for the reply Geoff... that isn't quite the problem. Here's a further explanation that should better articulate it. I've turned on debug on Apache, and this is what's currently happening when content is found in the cache: [Fri Apr 01 14:14:13 2005] [debug] mod_disk_cache.c(371): disk_ca

setting http output filter priority?

Hi mod_perl folks, background: I have an http output filter working on Apache 2.0.53 with mod_proxy, mod_cache, deflate and mod_perl 2.0.0-RC4. My functionality is that I'm stripping out certain types of advertising content, IMG tags, etc. The problem I'm seeing is that if a request can be fulfi