Getting at server's MaxClients setting?
Hi, is it possible to get at the configured MaxClients setting from within my handler? I checked Apache::Server which gives you some configuration information but there's no accessor for MaxClients. Any other, easy, way?
Re: Getting at server's MaxClients setting?
Geoff, On Wednesday, November 27, 2002, at 03:07 PM, Geoffrey Young wrote: in this case, MaxClients looks like it calls ap_daemons_limit to make the setting, so you may be able to create a subclass that offers up that API to perl-land. for more information on how, you can check out the slides from my recent ApacheCon talk: http://www.modperlcookbook.org/~geoff/slides/ApacheCon/oo-mod_perl-printable. ppt.gz hmm... sounds like another good idea to add to my next talk :) Well, for the moment, I'll grep for it in the current httpd.conf ;) Just thought I'd ask in case I was missing something obvious. Thanks, Bas.
MySQL prepared sql statements?
Hi, I'm wondering if I would gain (some) performance by preparing my sql statements once during ChildInit when using MySQL? Does it cache execution plans for sql statements? I know it makes a big difference on Oracle but I'm not sure what MySQL is like in this respect. Regards, Bas.
IPC suggestions sought/talking between children?
Hi all, I'm looking for some suggestions on a problem I'm looking at. I have a mod_perl handler running that sends out certain push messages (could be e-mail, could be an http post etc.) for several customers that are identified by a unique identifier. Examples: http://my.server.com/myhandler?target=customerAdata=sendthis http://my.server.com/myhandler?target=customerBdata=sendthat Now what I want to do is restrict the number of concurrent running requests per customer. For example, if 5 apache children are running my handler on behalf of customer A, I want my handler to do something else with a new request for customer A. I don't want to throw away the 6th request but slap it in some queue for later processing. I looked at mod_throttle and similar offerings but they tend to throw away requests when there are too many concurrent requests or when bandwidht is higher than some defined max.. That's not what I want. What I really need is some form of IPC between apache children so one child can find out what other children are currently doing. I've been thinking about IPC::Shareable but I have had some bad experience with it (longish ago though). Anyone got some pointers for me to look at? Is there something hidden in the Apache API that I might use from mod_perl? Regards, Bas.
FYI: MySQL default max_connections
Hi all, just recovering from a major spike on our MySQL server and thought I'd share something *I* didn't know... We got a db box with MySQL on it and 2 (mod_perl) webservers on separate boxes, the mod_perl servers have MaxClients set to 160 for one machine and 40 for the other. Now we had a little spike this evening and what happened was this: DBD::mysql::st execute failed: Too many connections at /opt/ttgp/prod/ttgp/lib/TTGP/ObjectServer.pm line 245. After some digging around it appears that MySQL default max. number of connections is 100, no matter how much RAM you stuck in your server! So even though I had no real problems with my webservers, MySQL didn't allow new connections. I added this to /etc/my.cnf (and restarted mysqld): [mysqld] ... set-variable = max_connections=300 ... Now I can have my 200 connections again and actually use that precious RAM in my db server ;) Just thought I'd mention it if you ever get caught with this problem. Regards, Bas.
Apache/mod_perl still not ready for OS X?
Hi, I've been postponing to move my Linux Apache/mod_perl development to my highly appreciated iBook running Mac OS X due to all the required tweaks until now. I would imagine things have been sorted out by now so I downloaded apache 1.3.24 to give it a go. The system I'm working on has a self-contained build-script which fetches everything from CVS, executes the right build commands etc. without me having to think. This has a built-in step to compile mod_perl statically (i.e. not DSO as i don't want that) into Apache. Now I find that out that the part that compiles Apache bails out with this: env LD_RUN_PATH=/opt/ttgp/dev/applications/perl/lib/5.6.1/darwin/CORE cc -c -I.. -I/opt/ttgp/dev/applications/perl/lib/5.6.1/darwin/CORE -I../os/unix -I../include -DDARWIN -DMOD_PERL -DUSE_PERL_SSI -pipe -fno-common -DHAS_TELLDIR_PROTOTYPE -fno-strict-aliasing -DUSE_HSREGEX -DNO_DL_NEEDED -pipe -fno-common -DHAS_TELLDIR_PROTOTYPE -fno-strict-aliasing `../apaci` alloc.c alloc.c: In function `spawn_child_core': alloc.c:2291: `STDOUT_FILENO' undeclared (first use in this function) alloc.c:2291: (Each undeclared identifier is reported only once alloc.c:2291: for each function it appears in.) alloc.c:2297: `STDIN_FILENO' undeclared (first use in this function) alloc.c:2303: `STDERR_FILENO' undeclared (first use in this function) make[4]: *** [alloc.o] Error 1 make[3]: *** [subdirs] Error 1 make[2]: *** [build-std] Error 2 make[1]: *** [build] Error 2 make: *** [apaci_httpd] Error 2 I almost find this appalling. It can't find something basic as STDOUT_FILENO (which is in /usr/include/unistd.h)... So I went back to Google to find solutions and the first hit sends me to stepwise.com to a tutorial that tells me in detail what commands to type in. Great, but it forces me to use DSO which I don't want! So, what is missing in the Apache configuration part that will make this work in a sensible way? Regards, Bas. ps. I know I'm a programmer that loves to tweak but when it comes to something as basic as this I'm just a user that's looking for the any key;)
Inline generation of error document?
Hi all, I'm a little confused (honestly). I want to handle parameter errors in a content handler. When there's a parameter missing in the URL, my handler returns HTTP_BAD_REQUEST. Now Apache sees the HTTP_BAD_REQUEST return value from my handler and generates an error (HTML) document. How can I create this document right from within my handler? I could create another handler of course and use the ErrorDocument directive to point to that but I am wondering if I can do it in my handler directly. Somehow I can't find this in the eagle book. Any tips? Bas.
Re: Non-web use for Apache/mod_perl
Issac, On Thursday, March 21, 2002, at 11:01 AM, Issac Goldstand wrote: Bas A.Schulte wrote: Hi, I've been meaning to write an article about how I used Apache/mod_perl to implement a mobile SMS application platform as it demonstrates use of Apache/mod_perl outside the Web realm, something I hadn't seen so far. Time constraints (as always) have prevented me from doing this properly, however, I'll try to give a short description of the system. The goal kinda is to see if others are using Apache/mod_perl in a similar way to share experiences, discuss issues that arise or discuss alternatives (I'd love someone to tell me I'm an idiot for handling all this lowlevel stuff myself and move to J2EE/java at once, if he can convince me I could have implemented this with the same budget and time frame). [snip] I actually implemented a similar system, although it's not SMS specific. Basically, I have a central system in Apache/mod_perl. The eventual design is for full I/O in many protocols, including HTTP, SMTP, FTP and a handful of other protocols. Well, this proves again that everything we software developers do has been done umteen times before ;) The system I described actually isn't tied to SMS as well; it has been designed with other devices in mind as well. The virtual SMSC component as I described it really should have a better name as it really is a protocal handler and serializing/deserializing component that is capable of handling other protocol and formats. In fact, I am considering enhancing it to be able to use it in a WAP or i-Mode context (if we can only find a way to get paid by the mobile subscriber using WAP which is easy with SMS). So far the output is pretty much ready, and I think I'm going to wait for Apache 2 before I do the inputs. I _could_ write mod_perl handlers to understand other protocols, but Apache 2 is being designed around this, as far as I understand, so I'm going to do that around Apache 2 (as soon as I get off my ass and learn the new API :-)) Yep; Apache 2 combined with mod_perl sure sounds interesting, now if it only had the stability that 1.x has ;) Regards, Bas.
Re: Non-web use for Apache/mod_perl
Hi,
On Thursday, March 21, 2002, at 11:39 AM, Ilya Martynov wrote:
On Thu, 21 Mar 2002 00:40:24 +0100, Bas A.Schulte
[EMAIL PROTECTED] said:
BAS To handle a large number of concurrent transactions in a
BAS transaction-safe environment without me having to worry too much
about
BAS concurrency issues and referential integrity I will slowly move to
BAS Oracle. $dbh-do('LOCK TABLE USER, INSTANCE, APP_DATA') just plain
BAS sucks unless you want to create a very large distributed
*single-user*
BAS system running on multiple machines.
Without changing SQL backend you can have transactions with
MySQL. Just use InnoDB table type. It is faster than default table
type for read/write intensive applications because it doesn't lock
whole tables but provides Oracle style row-level locking.
I knew I shouldn't have mentioned MySQL and it's, possibly perceived,
shortcomings ;)
I know about InnoDB, I know about Postgresql too, it's just that I
really like Oracle for lots of reasons. I have no personal experience
with running MySQL or Postgresql in a high-volume concurrent read/write
transaction situation. It seems most MySQL success stories are about
websites with 98% read-only transactions on the database which just
isn't similar to my needs. I used MySQL myself for that type of problem
and it sure worked.
If someone really wants to do another DB discussion, let's move that to
a different discussion ;)
Bas.
Re: Non-web use for Apache/mod_perl
Hi,
On Thursday, March 21, 2002, at 12:57 AM, Drew Taylor wrote:
Have you tried Postgresql? It's a free, fast, ACID compliant database.
I have it compiled running within a few hours, not knowing much of
anything about it before hand. From what I've read the speed is
comparable w/ mysql for most applications.
Drew
At 12:40 AM 3/21/2002 +0100, Bas A.Schulte wrote:
To handle a large number of concurrent transactions in a
transaction-safe environment without me having to worry too much about
concurrency issues and referential integrity I will slowly move to
Oracle. $dbh-do('LOCK TABLE USER, INSTANCE, APP_DATA') just plain
sucks unless you want to create a very large distributed *single-user*
system running on multiple machines.
I kinda was afraid about people getting excited about my RDBMS remarks
which really is a subject in itself ;)
I have zero personal experience with Postgresql even though I've read
good things about it so that ruled it out for a project with a very
tight time frame. And I just happen to really, really like Oracle so
unless the application generates some real money, I will probably always
go far Oracle.
Regards,
Bas.
