Re: [NeoStats-Devel] Perl causing problems: http://www.neostats.net/boards/viewtopic.php?t=1663

Should be fixed with the commits I made today.

Most of the perl code is now complete as far as the next alpha is  
concerned. I've started working on the messaging interface (to allow  
"inter-network messages and secureserv/opsb updates) and that should  
only take a few days to complete - I've got a few days off work)
(I dont consider that SecureServ should have the perl stuff enabled  
to do the next alpha release, because it also requires me to update  
the secure.irc-chat.net site, which we should do as a seperate  
dependancy

Then we can start considering the next alpha release.

As far as some commits today, one thing I'm doing is around the  
secureserv username/password management. I've tentively called it  
NeoNet (anyone else got better suggestions) but basically its a LDAP  
server that will eventually hold all auth info for all our sites as  
well as for SecureServ/OPSB/<insert module that might benifit from it>


Also, on a side note, what is the outcome of our discussions about  
having the database closed after a period of time? DNB hit a problem  
today where NeoStats crashed cause his shell provider had set a limit  
on his memory consumption. so it would be great if we can address  
this before the next release (because I'm hoping that after this, we  
are into beta, and as quickly as possible, a full release)

Thanks

Justin
On 29 Sep 2005, at 5:49 AM, M wrote:

> With respect to the provided log file:
>
> It appears the startup problem is a non issue as discussed in the  
> thread.
>
> I have never seen the shutdown crash but Justin has, so I am going to
> suggest this is a perl issue for the following reasons. The log file
> suggests the segfault occurs at or around the unload_perlextension  
> call.
> Since this call makes no attempt to ensure valid pointers before  
> using them,
> I expect it to cause segfaults. Since I am not able to compile with  
> perl
> support on any platform, this would explain why I have never seen  
> the issue.
> For this same reason, I would prefer someone with perl support to  
> confirm
> this to be the cause of the crash before a fix is applied.
>
> Mark.
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: devel-unsubscribe-wool9L35kifE9wlyV4mCnKxOck334EZe@public.gmane.org
> For additional commands, e-mail: [email protected]
>
>