> "TA" == Thomas Anders <[EMAIL PROTECTED]> writes:
>> Give that Thomas tested w/out issues: +1
TA> Is there any reason why this hasn't been applied yet even though
TA> we're at +3?
Only because it hasn't been that long business day wise and I'm at a
very busy IETF. I'll go apply it now.
--
Robert Story wrote:
> Give that Thomas tested w/out issues: +1
Is there any reason why this hasn't been applied yet even though we're at +3?
+Thomas
--
Thomas Anders (thomas.anders at blue-cable.de)
-
Using Tomcat but nee
> "RS" == Robert Story <[EMAIL PROTECTED]> writes:
RS> Agreed. And on the basis of that sliver of hope: +1
I think that this major of a change is far too late in our publication
process to make me feel safe. But then, letting Thomas stay grumpy
makes me feel unsafe too. +1 Apply it please.
Wes Hardaker wrote:
>> "RS" == Robert Story <[EMAIL PROTECTED]> writes:
>
> RS> [...] it seems like fixing those
> RS> modules would be cleaner than including a bunch of code that isn't
> RS> going to do anything.
Still, klookup involves system calls (lseek and read on /dev/kmem, usually) tha
On 06/11/06, Thomas Anders <[EMAIL PROTECTED]> wrote:
> Dave Shield wrote:
> > Revised patch, taking into account Robert's suggestions.
>
> +1, provided you:
>
> - will also adjust the MODULE-IDENTITY part of UCD-SNMP-MIB
> (LAST-UPDATED, REVISION/DESCRIPTION)
Good point.
> - will add a simi
On Mon, 06 Nov 2006 10:01:53 -0800 Wes wrote:
WH> The one thing that I
WH> think it might be worth doing for is so that you can at least compile
WH> everything and you may find some of the objects in the code file do
WH> work even if they all don't because KLOOKUP is used for some of the
WH> data.
Dave Shield wrote:
> Revised patch, taking into account Robert's suggestions.
+1, provided you:
- will also adjust the MODULE-IDENTITY part of UCD-SNMP-MIB (LAST-UPDATED,
REVISION/DESCRIPTION)
- will add a similar patch (MIB changes, code for new MIB objects) to the
5.[23].x branches
+Thomas
-
> "DS" == Dave Shield <[EMAIL PROTECTED]> writes:
DS> But Wes has been nagging me for *years* to commit changes little-and-often.
DS> I do occasionally listen to him :-)
Gasp.
/me runs off to check of "get Dave to listen to me" on his my todo list
--
Wes Hardaker
Sparta, Inc.
-
> "RS" == Robert Story <[EMAIL PROTECTED]> writes:
RS> The other is, is there really any benefit here? While it allows
RS> one to remove --enable-mini-agent, it will just enable a bunch of
RS> code that isn't going to do anything, right? Why does a broken
RS> nlist/klookup require --enable-min
Robert Story wrote:
> I have a few concerns with this patch.
>
> One is side-effects. In particular, all these breaks and returns are changing
> code paths.
Only if klookup() fails. I've added proper error handling for exactly
two reasons:
1) It was missing before (which hardly feels good).
2) I
> a number of systems (including OSX, AIX, IRIX and Tru64) have known problems
> when using nlist/klookup (/dev/kmem access) for a number of reasons. A minimal
> stop-gap fix is to turn off nlist/klookup usage completely, most conveniently
> done by "configure --without-kmem-usage" (which prevents
On Mon, 6 Nov 2006 16:33:25 + Dave wrote:
DS> Revised patch, taking into account Robert's suggestions.
+1
-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integra
Revised patch, taking into account Robert's suggestions.
Dave
Index: agent/mibgroup/ucd-snmp/memory.c
===
RCS file: /cvsroot/net-snmp/net-snmp/agent/mibgroup/ucd-snmp/memory.c,v
retrieving revision 5.17
diff -u -r5.17 memory.c
--- ag
Robert Story wrote:
> diff -u -p -r5.6 kernel.h
> --- agent/kernel.h31 Oct 2006 20:35:58 - 5.6
> +++ agent/kernel.h5 Nov 2006 22:13:22 -
> @@ -3,6 +3,8 @@
> extern void init_kmem(const char *);
> extern int klookup(unsigned long, char *, int);
> #define NETSNMP_KLOO
On 06/11/06, Robert Story <[EMAIL PROTECTED]> wrote:
> DS> I therefore offer the following patch which:
> DS> - reports the amount of memory used rather than free,
> DS> - introduces new MIB objects with consistent names/descriptions
> DS> - deprecates (but retains) the inconsistent
On Mon, 06 Nov 2006 13:13:51 +0100 Thomas wrote:
TA> my former "rfv: build fix for --without-kmem-usage"
TA>
TA> http://sourceforge.net/mailarchive/message.php?msg_id=37193353
TA>
TA> raised objections wrt. klookup error handling in certain mib modules, so
TA> I'm hereby proposing a modified pa
On Mon, 6 Nov 2006 11:12:35 + Dave wrote:
DS> > > Way back in February Dave Shield wrote:
DS> > > >>> b) Should 'memAvailSwapTXT' and 'memAvailRealTXT' report the
DS> > > >>> amount of memory free, or the amount used?
DS> > > >>> The MIB object name implies one, the description t
To the net-snmp patch tracker admin,
By mistake, I submitted twice the snmpusm application patch, do
you mind deleting/rejecting one of them? Preferable the one that was
submitted as nobody.
Thanks, and sorry about that.
Pablo
---
-Coders,
my former "rfv: build fix for --without-kmem-usage"
http://sourceforge.net/mailarchive/message.php?msg_id=37193353
raised objections wrt. klookup error handling in certain mib modules, so
I'm hereby proposing a modified patch that allows for
--without-kmem-usage *and* adds proper erro
> Way back in February Dave Shield wrote:
> >>> b) Should 'memAvailSwapTXT' and 'memAvailRealTXT' report the
> >>> amount of memory free, or the amount used?
> >>> The MIB object name implies one, the description the other!
The latest HAL-based memory module followed the name of t
On 06/11/06, Thomas Anders <[EMAIL PROTECTED]> wrote:
> Unless/until someone can propose an immediate fix for the
> proxy code, I'd propose to document this as a missing feature and adjust
> the proxy tests to skip on TCP. Objections?
Sounds good to me.
Dave
---
Magnus Fromreide wrote:
> On lör, 2006-11-04 at 15:09 +0100, Thomas Anders wrote:
>> (i.e. proxy over TCP) will fail. UDP works fine, as well as starting the
>> proxy
>> destination agent before the agent itself.
>
> Hmm. To me it looks like the problem is that we aren't considering
> connection
22 matches
Mail list logo
