Re: Forbidden access
Paul Cotter wrote: Also there's a typo: It has been confirmed by the original author that the typo was only in the email sample. There was also a /Locatiom but I guess this was a transcription error also. As a general rule, when posting slices of config, code etc. to the lists: ALWAYS use your window-manager's cut'n'paste utility to grab the code. NEVER transcribe it by typing it in again... If you try to type it in, you will only type in what you THINK it says - not what it actually says, i.e. you might unconciously correct an error in the code or introduce a new error which becomes red-herring. Rgds, Owen Boyle.
rational.com
is www.rational.commodperl based or java/jsp ? [frank@proxy frank]$ telnet www.rational.com 80Trying 207.25.225.84...Connected to www.rational.com.Escape character is '^]'.HEAD / HTTP/1.0 HTTP/1.1 302 FoundDate: Thu, 08 Mar 2001 13:03:18 GMTServer: Apache/1.3.12 (Unix) mod_perl/1.21Location: http://www.rational.com/index.jspConnection: closeContent-Type: text/html; charset=iso-8859-1
Re: rational.com
They are running Apache with Mod_Perl installed, but the page that you are visitng is a .jsp page.
[Apache::Cookie] Segv in Cookie.so ?
Anyone else experiencing segv's in Cookie.so? I'm using mod_perl 1.25, Apache 1.3.14, and libapreq (Apache::Cookie) 0.31.3 with HTML::Mason 0.89. I tried a rebuild of libapreq etc., but it still segv's. I've an authen handler and a request handler, and both use Apache::Cookie. I'm not certain where it's seg faulting... My error log is full of messages about apache's dieing with sig 11, and the screens in the browser are often missing graphics, or coming up with a "document contains no data" dialog at that point. Perhaps it serves one request and then segv's? If this is not a common problem, let me know, and I'll spend some time trying to get a backtrace and recipe to reproduce the bug. -- mailto: (Karl M. Hegbloom) [EMAIL PROTECTED] http://www.microsharp.com phone://USA/WA/360-260-2066
RE: Easy Newbie mod_perl installation question
layouts seem to be defined in: ../apache_1.3.x/config.layout John -Original Message- From: Andy Emmerich [mailto:[EMAIL PROTECTED]] Sent: Thursday, March 08, 2001 2:48 AM To: [EMAIL PROTECTED] Subject: Easy Newbie mod_perl installation question When mod_perl's Makefile builds the new httpd in the /src directory, where does it get the apache layout from? Currently, my build with modperl is looking for httpd.conf in /usr/local/apache, which is not my apache source directory. I've put the actual directory in every place I can, from the apache source dir's layout file to the Makefile.PL's APACHE_SRC value. What am I missing? Thanks, and my apologies if this is already in the archive somewhere and I missed it! Andy Emmerich CS/Engineering Undergrad University of Illinois
Fwd: Different code in different children
I have a site that is up and running and serving several customers using Mod_Perl, Apache, and MySQL on a Freebsd 4 machine. I am trying to add enhancements to my code. I created a new directory called dev and copied all of my scripts into that directory to work on them. Several of my scripts 'require' a package I made called init.pr. As I do my testing on the dev directory, and the live people are in my live directory, it seems that some Apache children have the dev init.pr loaded, and some have the live init.pr loaded. This is true even if I give the whole path to the require file. (yes, I am restarting the server...) Is this a common problem? Do I HAVE to have an entirely different server for development? Any help will be appreciated... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Kelly Dodge http://www.edurus.com IT Director [EMAIL PROTECTED] edurus, Inc. (775) 885-8333 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Kelly Dodge http://www.edurus.com IT Director [EMAIL PROTECTED] edurus, Inc. (775) 885-8333
Can't find index.html
We are running Apache 1.3.17 and mod_perl 1.25 on Solaris 7. The problem I am having is that when I type: http://my.domain.name I get a error message: The document contained no data. Try again later, or contact the server's administrator And in the error log I get: child pid 15965 exit signal Bus Error (10) Yet when I type: http://my.domian.name/index.html The page loads no problem. This problem only started after I have loaded mod_perl. John Whitnack [EMAIL PROTECTED] Gilbert Laustsen Jung
Re: Can't find index.html
John Whitnack ([EMAIL PROTECTED]) said something to this effect on 03/08/2001:
We are running Apache 1.3.17 and mod_perl 1.25 on Solaris 7.
The problem I am having is that when I type:
http://my.domain.name
I get a error message:
The document contained no data.
Try again later, or contact the server's administrator
And in the error log I get:
child pid 15965 exit signal Bus Error (10)
Yet when I type:
http://my.domian.name/index.html
The page loads no problem.
This problem only started after I have loaded mod_perl.
It sounds like you have something like:
Location /
SetHandler perl-script
PerlHandler My::Handler
/Location
in your httpd.conf. Try either:
FilesMatch "*\.html$"
SetHandler perl-script
PerlHandler My::Handler
/FilesMatch
Or, in your handler:
sub handler {
my $r = shift;
if ($r-content_type eq 'text/html' || -d $r-finfo) {
return DECLINED;
}
# the rest as usual...
In other words, decline to handle directory listings. Let mod_dir
or mod_autoindex do it's job.
(darren)
--
Your freedom to swing your fist ends at the tip of my nose.
RE: Make test hangs on httpdconf
-Original Message-
From: Stas Bekman [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 06, 2001 10:08 AM
To: John Walker
Cc: '[EMAIL PROTECTED]'
Subject: RE: Make test hangs on httpdconf
Actually, it seems I've screwed things up worse than that.
[...]
try to run the test under strace
% strace perl t/TEST ...
[...]
Alternatively, you can attach to the hanging process
fails on the same test still and I've reaquired the source, and done the
simplest of build from you guide. (And there aren't any errors in the
build.)
I've attached a txt file containing the output of strace -p. It doesn't mean
much to me, but perhaps you will see something meaningful. One line did
strike me as odd, but I got it again even after downloading a new tarball.
(No the file is not there...why do we test for it?)
open("/usr/src/mod_perl-1.25/t/.htaccess", O_RDONLY) = -1 ENOENT (No such
file or directory)
In the mean time, I have done a few other things to try to resolve the
problem. I am using Debian linux for the first time, and since I am familiar
with CPAN module shell, I uninstalled all the .deb perl files (except the
required base package) and then reinstalled everything through CPAN. CPAN
tells me (cpan r) that I should upgrade Test::Harness, but when I try this
also fails. I don't know if this is related, but the name coincidence
strikes me as odd.
What could underly both of these and cause my problems? (I think my gcc
envirnment is pretty good... I was able to build a new 2.4 kernel with it
anyway...)
Thanks for the help,
John
From: root [[EMAIL PROTECTED]]
Sent: Thursday, March 08, 2001 2:42 PM
To: John Walker
accept(16, {sin_family=AF_INET, sin_port=htons(32845),
sin_addr=inet_addr("127.0.0.1")}}, [16]) = 3
rt_sigaction(SIGUSR1, {SIG_IGN}, {0x809f294, [], SA_INTERRUPT|0x400}, 8) = 0
getsockname(3, {sin_family=AF_INET, sin_port=htons(8529),
sin_addr=inet_addr("127.0.0.1")}}, [16]) = 0
setsockopt(3, IPPROTO_TCP1, [1], 4) = 0
alarm(300) = 0
read(3, "GET /test.html HTTP/1.0\r\nHost: l"..., 4096) = 79
rt_sigaction(SIGUSR1, {SIG_IGN}, {SIG_IGN}, 8) = 0
time(NULL) = 984080384
alarm(300) = 300
alarm(0)= 300
rt_sigaction(SIGALRM, NULL, {0x809def4, [], SA_RESTART|0x400}, 8) = 0
dup2(15, 2) = 2
rt_sigprocmask(SIG_BLOCK, NULL, [], 8) = 0
stat("/usr/src/mod_perl-1.25/t/docs/test.html", {st_mode=S_IFREG|0644, st_size=566,
...}) = 0
open("/usr/src/mod_perl-1.25/t/.htaccess", O_RDONLY) = -1 ENOENT (No such file or
directory)
open("/usr/src/mod_perl-1.25/t/docs/.htaccess", O_RDONLY) = 4
fstat(4, {st_mode=S_IFREG|0666, st_size=1, ...}) = 0
fstat(4, {st_mode=S_IFREG|0666, st_size=1, ...}) = 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x40015000
read(4, " ", 4096) = 1
read(4, "", 4096) = 0
read(4, "", 4096) = 0
close(4)= 0
munmap(0x40015000, 4096)= 0
open("/usr/src/mod_perl-1.25/t/docs/test.html", O_RDONLY) = 4
old_mmap(NULL, 566, PROT_READ, MAP_PRIVATE, 4, 0) = 0x40015000
alarm(300) = 0
alarm(0)= 300
alarm(300) = 0
alarm(300) = 300
alarm(0)= 300
close(4)= 0
select(4, [3], NULL, NULL, {0, 0}) = 0 (Timeout)
write(3, "HTTP/1.1 200 OK\r\nDate: Thu, 08 M"..., 839) = 839
alarm(30) = 0
shutdown(3, 1 /* send */) = 0
select(4, [3], NULL, NULL, {2, 0}) = 1 (in [3], left {2, 0})
read(3, "", 512)= 0
close(3)= 0
alarm(0)= 30
rt_sigaction(SIGUSR1, {0x809f294, [], SA_INTERRUPT|0x400}, {SIG_IGN}, 8) = 0
alarm(0)= 0
munmap(0x40015000, 566) = 0
rt_sigaction(SIGALRM, {0x809def4, [], SA_RESTART|0x400}, {0x809def4, [],
SA_RESTART|0x400}, 8) = 0
accept(16, {sin_family=AF_INET, sin_port=htons(32846),
sin_addr=inet_addr("127.0.0.1")}}, [16]) = 3
rt_sigaction(SIGUSR1, {SIG_IGN}, {0x809f294, [], SA_INTERRUPT|0x400}, 8) = 0
getsockname(3, {sin_family=AF_INET, sin_port=htons(8529),
sin_addr=inet_addr("127.0.0.1")}}, [16]) = 0
setsockopt(3, IPPROTO_TCP1, [1], 4) = 0
alarm(300) = 0
read(3, "GET /test.html HTTP/1.0\r\nHost: l"..., 4096) = 79
rt_sigaction(SIGUSR1, {SIG_IGN}, {SIG_IGN}, 8) = 0
time(NULL) = 984080384
alarm(300) = 300
alarm(0)= 300
rt_sigaction(SIGALRM, NULL, {0x809def4, [], SA_RESTART|0x400}, 8) = 0
dup2(15, 2) = 2
rt_sigprocmask(SIG_BLOCK, NULL, [], 8) = 0
row locking problems with AutoCommit, Apache::Session, and Oracle
Hi,
I've been experiencing row locking problems using
Oracle 8.1.6 and Apache::Session with AutoCommit on.
It seems that when AutoCommit is on, doing a "select
for update" locks the row until you execute a
subsequent update command or issue an explicit commit
statement. (You can replicate this using sqlplus in
two terminal windows.)
The problem with this is that the materialize method
in Apache::Session::Store::Oracle issues a "select for
update" command during the TIEHASH call, even if no
subsequent update ever occurs. So if you're just
reading from the session object, the row gets locked
but never unlocked, and subsequent session calls hang.
A quick fix is to always issue the
$self-{dbh}-commit call in the DESTROY method of
Apache::Session::Store::Oracle, regardless of whether
or not $self-{commit} is set to 0 or 1. This will
generate a warning when AutoCommit is on, but seems to
solve the problem.
Dan
__
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/
Re: Fwd: Different code in different children
On Thu, 8 Mar 2001, Kelly Dodge wrote: I have a site that is up and running and serving several customers using Mod_Perl, Apache, and MySQL on a Freebsd 4 machine. I am trying to add enhancements to my code. I created a new directory called dev and copied all of my scripts into that directory to work on them. Several of my scripts 'require' a package I made called init.pr. As I do my testing on the dev directory, and the live people are in my live directory, it seems that some Apache children have the dev init.pr loaded, and some have the live init.pr loaded. This is true even if I give the whole path to the require file. (yes, I am restarting the server...) Is this a common problem? Do I HAVE to have an entirely different server for development? Any help will be appreciated... It's very confusing to reply to an email which already has been quoted as replied :( Please RTFM: http://perl.apache.org/guide/porting.html#Name_collisions_with_Modules_and http://perl.apache.org/guide/modules.html#Apache_PerlVINC_Allows_Module http://perl.apache.org/guide/multiuser.html#Virtual_Hosts_in_the_guide _ Stas Bekman JAm_pH -- Just Another mod_perl Hacker http://stason.org/ mod_perl Guide http://perl.apache.org/guide mailto:[EMAIL PROTECTED] http://apachetoday.com http://logilune.com/ http://singlesheaven.com http://perl.apache.org http://perlmonth.com/
RE: Make test hangs on httpdconf
On Thu, 8 Mar 2001, John Walker wrote:
-Original Message-
From: Stas Bekman [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, March 06, 2001 10:08 AM
To: John Walker
Cc: '[EMAIL PROTECTED]'
Subject: RE: Make test hangs on httpdconf
Actually, it seems I've screwed things up worse than that.
[...]
try to run the test under strace
% strace perl t/TEST ...
[...]
Alternatively, you can attach to the hanging process
fails on the same test still and I've reaquired the source, and done the
simplest of build from you guide. (And there aren't any errors in the
build.)
I've attached a txt file containing the output of strace -p. It doesn't mean
much to me, but perhaps you will see something meaningful. One line did
strike me as odd, but I got it again even after downloading a new tarball.
(No the file is not there...why do we test for it?)
open("/usr/src/mod_perl-1.25/t/.htaccess", O_RDONLY) = -1 ENOENT (No such
file or directory)
That's absolutely normal. See:
http://perl.apache.org/guide/performance.html#Reducing_the_Number_of_stat_Ca
In the mean time, I have done a few other things to try to resolve the
problem. I am using Debian linux for the first time, and since I am familiar
with CPAN module shell, I uninstalled all the .deb perl files (except the
required base package) and then reinstalled everything through CPAN. CPAN
tells me (cpan r) that I should upgrade Test::Harness, but when I try this
also fails. I don't know if this is related, but the name coincidence
strikes me as odd.
What could underly both of these and cause my problems? (I think my gcc
envirnment is pretty good... I was able to build a new 2.4 kernel with it
anyway...)
Hmm, I've compared the trace generated on my machine during the:
% strace perl t/TEST modules/httpdconf
and it's very different. Are you running the same test?
What perl/apache/linux kernel/gcc/ versions are you using? I'm sure thing
didn't try it on kernel 2.4 yet. I prefer to wait a little before all the
bugs will be gleaned out.
_
Stas Bekman JAm_pH -- Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide http://perl.apache.org/guide
mailto:[EMAIL PROTECTED] http://apachetoday.com http://logilune.com/
http://singlesheaven.com http://perl.apache.org http://perlmonth.com/
[OT] UNIX timestamp hits 1,000,000,000 this year!
Just as a FYI about something that caught my attention recently. This year on Saturday September 8, 2001, the unix time stamp flips to 1 billion and gets another digit going from 9 to 10 digits. Not sure if anyone else but me is using the timestamp in ways that were set to 9 digits, such as DB column int length or in strings of fixed lengths, but these possibilities should be looked at just in case they could fail when the time changes to the longer int. Just a heads up, hope it helps someone. Bill -- Bill Desjardins - [EMAIL PROTECTED] - (USA) 305.205.8644 Unix/Network Consulting - perl/mod_perl/SQL development http://www.CarRacing.com - Powered by FreeBSD/mod_perl http://www.FreeBSD.org - The Best OS Money can't buy f u cn rd ths, u cn gt a gd jb n cmptr prgrmmng...
