>>> On 2/8/2008 at 1:44 PM, in message
<[EMAIL PROTECTED]>, "Bernard Li"
<[EMAIL PROTECTED]> wrote:
> Hi guys:
> 
> I was just alerted of a bug in the 3.0.x branch frontend where the
> host graph's (now x.xx) number is always 0.00.
> 
> Jesse Becker has already submitted a patch against trunk and Carlo has
> already checked it in, the patch also cleanly applies to the branch.
> 
> In my books -- this is quite a bad bug and I think we need to do a
> release (either as 3.0.6.1 or 3.0.7).
> 
> Thoughts?
> 
> BTW, given the fact that 3.1.0 will break compatibility with 3.0.x, I
> have a feeling that a lot of folks probably won't want to upgrade
> (unless we can work out the upgrade path for past data).  So, perhaps
> we might need to maintain the 3.0.x branch after all...
> 

We should release a 3.0.7 rather than introduce another level of revision 
number.  The layout that we have currently for releases is similar to the same 
release layout as Apache.  This will allow us to continue to support the 3.0.x 
branch for as long as we need to.  I would guess that we will need to support 
the branch at least until after 3.1.x is released and migration from 3.0.x to 
3.1.x starts to happen.  But the 3.0.x branch should be just maintenance and 
releases should only happen in cases such as this.

Nothing has changed in the RRD file format, so there shouldn't be any reason 
why a simple upgrade of gmond and gmetad to 3.1.0 wouldn't work.  The things 
that changed that are not backward compatible where the way that gmonds talk to 
each other and the format of the XML produced by gmond and gmetad.  In fact the 
XML format is most likely backward compatible anyway.  There were just some new 
tags added that would be ignored by an older gmetad as well as frontend.  The 
upgrade path would really be:

1. Upgrade all gmond nodes within a cluster
2. Progressively upgrade all clusters
3. Progressively upgrade all gmetad nodes
4. Upgrade the frontend

We will have to test the above path to make sure that there aren't any 
glitches, but it should work.

Brad




-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ganglia-developers mailing list
Ganglia-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-developers

Reply via email to