Over the last few hours, I uninstalled the MySql RPM from the server and installed the latest binary distro. I sat down with each individual perl module and made sure it installed correctly. Then I recompiled and installed slash.
All went fine, except for the following:
Bundle::libnet -- Net::Telnet did not install or test properly... (failed test 3) all else seems fine. Hopefully slash doesn't need to telnet out places to run >> grins
Data::ShowTable -- Massive Errors... what a nightmare. The CPAN package simply will not compile at all, but returns all values as if it installs correctly. I tried installing the RPM file to get around this, since every tar for this module did the same thing. See last journal entry for more beitching about this...
Template Toolkit -- Failed Three Tests under t/dbi... test 59, 61, 63. I can't seem to figure out what these tests actually are, or if they are important (they probably are) but database access seems to work properly... perhaps these tests relate to Data::ShowTable (?)
MailTools -- Couldn't figure out the CPAN Identifier for this package, had to download and install by hand. Compiled and installed ok, except it had an error making a man page during compile.
TimeDate and Time-Modules -- Also couldn't figure out the CPAN Identifier, but installed fine.
Then, upon installing Bundle::Slash, new forced installs of LWP, Msql and Mailtools are run. Hoorah.
Then slash installed, configured and started up the website fine.
==================
These steps resolved the problem with a huge number of mysqld processes spawning, but did not resolve the erros getting dumped into the error log...
[Sat Jan 26 21:50:52 2002] [error]
[Sat Jan 26 21:50:52 2002] [error] Which was called by:Slash::Display::Plugin:/usr/local/lib/perl5/si
I'm really not sure what to do next, but when I figure something out i'll report it.
In process of installing each CPAN bundle required by slash individually so I can see that there are no problems with any of them.
The Data::ShowTable bundle is giving me hell... The CPAN module will not make or install properly, but even though it is full of errors it is returning an OK runlevel, so everything THINKS it has installed properly. Full make and make test output is listed at the end of this post.
I tried to manually make this module from several CPAN mirrors, so it doesn't seem to be a corrupt file. There doesn't seem to be any previous versions of this module available, and the contact information listed in the readme no longer exists. No help or updates there...
I grabbed the binary RPM package for this from redhat.com, the package is for perl 5.005, not perl 5.6.0. I forced an install, then moved the package files from
but hay at least as always, the CPAN shell still gives a reassuring Data::ShowTable is up to date...
== BEGIN TEXT OF SHOW-TABLE COMPILE AND TEST ERROR OUTPUT ==
CPAN.pm: Going to build A/AK/AKSTE/Data-ShowTable-3.3.tar.gz
Checking if your kit is complete...
Looks good
Writing Makefile for Data::ShowTable
cp ShowTable.pm blib/lib/Data/ShowTable.pm
cp showtable blib/script/showtable
Manifying blib/man3/Data::ShowTable.3
*** ERROR: unterminated I at line 724 in file ShowTable.pm
*** ERROR: unterminated I at line 724 in file ShowTable.pm
Manifying blib/man1/showtable.1
Running make test
PERL_DL_NONLAZY=1
t/box-calls.........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/box-calls.........ok
t/box-scaling.......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/box-scaling.......ok
t/box-widths........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/box-widths........ok
t/box-wrap..........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/box-wrap..........ok
t/html-calls........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/html-calls........ok
t/html-fmt..........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/html-fmt..........ok
t/html-scaling......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/html-scaling......ok
t/html-widths.......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/html-widths.......ok
t/html-wrap.........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/html-wrap.........ok
t/list-calls........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/list-calls........ok
t/list-scaling......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/list-scaling......ok
t/list-widths.......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/list-widths.......ok
t/list-wrap.........Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/list-wrap.........ok
t/simple-calls......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/simple-calls......ok
t/simple-scaling....Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/simple-scaling....ok
t/simple-widths.....Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/simple-widths.....ok
t/simple-wrap.......Prototype mismatch: sub main::start_tests ($) vs none at t/Test-Setup.pl line 27.
Prototype mismatch: sub main::run_test ($&) vs none at t/Test-Setup.pl line 102.
t/simple-wrap.......ok
All tests successful.
Files=17, Tests=128, 4 wallclock secs ( 3.03 cusr + 1.37 csys = 4.40 CPU)
Running make install
Installing
Installing
Installing
Writing
Appending installation info to
I have been working for a few days now on getting slash up and running for our DDR fan site, www.teamleapfrog.com. Since I have ran into lots of issues that are probably mostly due to my own inexperience, I will document what I am going through here so in the future anybody with the same issues might find their answers easier than I have.
As of now, I have gotten the site completly up and running on the production server. After some difficulties here and there, everything seems to run fine... except...
On occasion, about 10% of the time after the site has been running for a few minutes, some pages will simply spit out "Internal Server Error" rather than displaying properly.
Upon further investigation, when this is happening, the error log shows that DBIx::Password is spitting out the following error:
unable to connect to MySQL: : Too many connections
DBI->connect(database=slash;host=localhost) failed: Too many connections at
Checking processes shows that the mysqld has spawned nearly as many mysql processes as apache!! None of the other DB apps I have written in perl have done this with msyql, but that still doesn't really tell me if this is normal or not. I'm betting no it's not normal... so next step I am un-installing the mysql server RPM that was installed on the server several months ago, and installing the latest binary dist of mysqld.