Dave Shield wrote:
This patch eliminates an incorrect logic test in ucd-snmp/disk.c


Hmmmm....
Sorry Mike, but I'm not convinced.

Just looking at the code, the check now seems to bear absolutely
no relation to the code it is allegedly protecting.
Can you remind me what the compile error was previously?
I'm guessing that it was complaining about 'vfs.f_favail'
and/or 'vfs.f_ffree'.

So the obvious answer would be to add a check for
STRUCT_STATVFS_HAS_F_FAVAIL (which is what is actually being used)
rather than STRUCT_STATFS_HAS_F_FAVAIL (which I suspect may well be
an error that's never got picked up before)

By dropping the check for STRUCT_STATVFS_HAS_F_FILES completely,
my suspicion is that this code block may well never be compiled on
*any* O/S.   (Even the ones that need it).


I realise that you need to move on to other things now, but if anyone else is able to investigate the effect of a new configure.in check on 'AC_CHECK_STRUCT_FOR(statvfs,f_favail)' (both on FreeBSD, and {Open,Net}BSD) that would be incredibly useful.

Dave




After staring at this last night, I realize that you had the right solution
from the beginning.  I'll try configure.in on FreeBSD, but I'll leave
{Open,Net}BSD to the professionals that have those platforms instantiated.

Thanks again, Dave!


By the way -- Has anyone seen an agent instantiate UCD-SNMP-MIB::dskTable lately ? On a system I'm using, /dev/da1 holds FreeBSD 5.2, /dev/da2 holds FreeBSD 4.9 Only partitions from da1 are instantiated on the agent running on FreeBSD 4.9 I've not seen dskTable on Linux, either. All tests done after applying the FreeBSD changes that I've been hawking.

-Mike Slifcak



-------------------------------------------------------
This SF.Net email is sponsored by: GNOME Foundation
Hackers Unite!  GUADEC: The world's #1 Open Source Desktop Event.
GNOME Users and Developers European Conference, 28-30th June in Norway
http://2004/guadec.org
_______________________________________________
Net-snmp-coders mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to