perl-embperl mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kee Hinckley <>
Subject Re: @ISA corrupted
Date Mon, 11 Nov 2002 22:01:42 GMT
At 8:33 PM +0100 11/10/02, Gerald Richter wrote:
>First print out the content of @ISA, see what it contains when the page is
>served ok and when it fails. Maybe some other module youare using is

I put a debug statement in both the template, and in the page that it 
is executing.

1036987605      69622   Template.html   HTML::Embperl::Req 
1036987605      69622   DAlfonso.html    HTML::Embperl::DOC::_4
1036987609      69631   Template.html   HTML::Embperl::Req 
1036987609      69631   DAlfonso.html    HTML::Embperl::DOC::_9
1036987612      69631   Template.html   HTML::Embperl::Req 
1036987612      69631   DAlfonso.html    HTML::Embperl::DOC::_9
1036987616      69620   Template.html   HTML::Embperl::Req 
1036987616      69620   DAlfonso.html    HTML::Embperl::DOC::_8

When I get the failure, it refers to a different document than most 
of the other, but since that's the object name, it doesn't tell me 
which document it was.  The way it's behaving it's pretty clear that 
one of the Apache processes has something corrupt in the structure, 
and whenever it serves the page things die.

The question is, what sequence of hits caused it to get screwed up. 
I can't run http -X, I don't have access to the server on this 
machine.  All I can do is restart it.  The error doesn't occur 
elsewhere, and it doesn't occur immediately on this machine.

I'm going to go through the log file and see if I can pinpoint the 
sequence of hits and duplicate it, however it appears that I may have 
found a work around.

I was doing an Execute('*') of the document (with various other 
arguments).  But it occured to me that in this case the documents 
contain no Embperl code, so I added the flag to tell it to just 
process as Text (thanks for that!) and so far we've run for almost 24 
hours with no more errors.

Not a fix, but at least I'm not in a panic.

If I can come up with a reasonable set of test conditions to cause 
the error I'll temporarily set it back and see if I can duplicate it.

Kee Hinckley - Somewhere.Com, LLC

I'm not sure which upsets me more; that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message