Re: [Freeipmi-users] in-band failure on Suns

[Al Chu]

Hey Dave,

I think I got the fix in.  Could you please check it out?

http://ftp.zresearch.com/pub/freeipmi/qa-release/freeipmi-0.7.10.beta1.tar.gz

and the important part of the ChangeLog

        * common/src/toolcommon/tool-common.c (ipmi_open),
        ipmimonitoring/src/ipmimonitoring/: Fix in-band communication
bug
        when workaround flags for non-inband are specified.


Al

On Thu, 2009-05-28 at 22:25 +0100, Dave Love wrote:
> I've figured out what's going on to give the difference between my
> simple build of the new version and the rpm.  First of all, they are
> both dynamically linked -- I'd forgotten how libtool builds work, and
> had just seen ldd say `not a dynamic executable' without thinking in a
> hurry.
> 
> The difference is that the packaged version was picking up my existing
> /etc/freeipmi.conf (unlike the one destined for /usr/local).  That has
> 
>   workaround-flags authcap
> 
> appropriate for the majority of our nodes out-of-band.  Commenting that
> out makes the new bmc-info work in-band with the kernel drivers loaded.
> 
> Is that a bug or expected?  (The business of these workarounds is rather
> a nightmare.)
> 
-- 
Albert Chu
address@hidden
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory