pcp
[Top] [All Lists]

Re: [pcp] PCP Updates: Implement fche's Threading Algorithm for Active P

To: Dave Brolley <brolley@xxxxxxxxxx>
Subject: Re: [pcp] PCP Updates: Implement fche's Threading Algorithm for Active Probing
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Wed, 4 Jun 2014 00:35:23 -0400 (EDT)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <421058401.19861494.1401854272445.JavaMail.zimbra@xxxxxxxxxx>
References: <538DF862.603@xxxxxxxxxx> <538E28A5.2000104@xxxxxxxxxx> <421058401.19861494.1401854272445.JavaMail.zimbra@xxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: ij/0QJyY2vYeSXDrgtys+3mJEVZRIKLh5e6s
Thread-topic: PCP Updates: Implement fche's Threading Algorithm for Active Probing

----- Original Message -----
> Hi Dave,
> 
> Lookin' good - I guess HAVE_SEM_T can go the way of the dodo now?
> 

Also, we seem to always HAVE_SERVICE_DISCOVERY now (probe.c is
always built in libpcp), which is giving some build fallout...

<mgoodwin> nathans, I'm seeing a build error in libpcp, /usr/bin/ld: 
libpcp.so.3: hidden symbol `__pmAddDiscoveredService' isn't defined
<mgoodwin> http://fpaste.org/107014/40185551/
<nathans> mgoodwin, this should get ya going - i'm not convinced its the ideal 
fix though, i need to chat to dave a bit
<nathans> http://www.fpaste.org/107015/


I think we should probably ditch that macro from the configure-y
also, right?  (in libpcp/src/config.c SERVICE_DISCOVERY_ENABLED
would become simply "enabled" only, I guess)

cheers.

--
Nathan

<Prev in Thread] Current Thread [Next in Thread>