[ic] Re: ALERT: bad pipe signal received for /page.html

[Music]

> I've been seeing this too, on my Apache 2 and latest Link.pm. I also  had 
> to use PERL_SIGNALS=unsafe and so I get quite a lot of these.
>
> The visible effect on the browser is that the page or image (which 
> Link.pm apparently still has some part in delivering) does not load.  I 
> get them myself when browsing and testing my websites, and I have  never 
> stopped loading a page or had any other problems on non-IC  sites I host.
>
> I was told the problem stems from either the browser and a stop  button or 
> some other network fault. I may go back to Apache 1.3 to  get around this.

I have no idea if this is related in anyway however just in case there is 
any connection:

I find that IC 5.4.1 on Apache 2.0.46-61.ent.centos3 (using secondary 
unthreaded perl 5.8.0 and suexec) and the standard CGI link gets into a 
'faulty' state that only occurs after certain catalogs have recently had 
'Apply Changes' applied.

The result is an intermittent error where a bunch of IC tags are not parsed 
and dumped to the screen (can happen on any page)

Depending on the catalog the webpage output looks something like a broken 
page with something similar displayed:

[control-set search_box_small[/component] [/control-set] [calc 
$Variable->{MV_DHTML_BROWSER} ||= 
'MSIE.*[5-9]\..*Windows|Mozilla.*Gecko|Opera.*[7-9]\.'; 
$Scratch->{dhtml_browser} = $Session->{browser} =~ 
m{$Variable->{MV_DHTML_BROWSER}}; if($Scratch->{members_only} and ! 
$Session->{logged_in}) { $Scratch->{mv_successpage} = $Tag->var('MV_PAGE', 
1); $Tag->deliver({ location => $Tag->area('login')}); } return; [/calc]

To reproduce the error I just need to hit apply changes with a new STANDARD 
store, and then keep refreshing a page. (Hit F5 over and over again.)
The pages will render beautifully however eventually the broken page will 
appear. Another refresh and the problem disappears. (Very intermittent but 
consistent)

The bad thing is that the faulty condition affects all catalogues on the 
server and similar garble will intermittently appear on other catalog pages.

To fix the problem I need to restart Interchange.

I did some more experimenting today inspired by this thread however the 
problem persists.

- I have tried setting all catalogs envrionments to use mod_perl
- I have disabled all mod_perl
- I have restarting both with and without PERL_SIGNALS=unsafe in command 
line.
- I have only tried this using RPC mode with variations of:

ifdef TRAFFIC =~ /rpc/i
Message RPC traffic settings.
PreFork             Yes
StartServers        5
MaxServers          0
MaxRequestsPerChild 100
HouseKeeping        2
PIDcheck            120
ChildLife 30 minutes
endif

I also tried disabling ChildLife and PreFork and increasing start servers.

All with no joy.

'Applying Changes' to one catalogue puts the Interchange daemon into faulty 
state.

No errors are displayed in IC error.log, catalog error.log nor apache error 
logs.

Just thought I would document this on the list to see if any feedback etc.