5.2.2.pre4 was uploaded moments ago to the SF servers and is available
for testing. Come Friday we'll move to rc1 after which point only
show stoppers will be addressed.
--
Wes Hardaker
Sparta, Inc.
---
This SF.Net email is sponsored by:
Po
> On Mon, 10 Oct 2005 18:26:06 +0200, Thomas Anders <[EMAIL PROTECTED]>
> said:
Thomas> is there a good reason why a few of the perl scripts we ship contain a
Thomas> "#!/usr/local/bin/perl" shebang line rather than the default
Thomas> "#!/usr/bin/perl" we use everywhere else?
Actually
> On Mon, 10 Oct 2005 11:41:19 -0700, Juan Betty-MGIA2720 <[EMAIL
> PROTECTED]> said:
Juan> One place that needs to be updated too is probably the
Juan> sysuptime.
Actually sysuptime was recently fixed using the times() function instead.
--
Wes Hardaker
Sparta, Inc.
Robert Story wrote:
Modified Files:
read_config.c
Log Message:
- break apart reading of config files and persistent config files (allows
proper handling of persistentDir token in config files)
- remove commented out code
Can you please elaborate? What problems did you fix this change?
+T
One place that needs to be updated too is probably the sysuptime.
-Original Message-
From: Magnus Fromreide [mailto:[EMAIL PROTECTED]
Sent: Friday, October 07, 2005 10:49 PM
To: [email protected]
Cc: Juan Betty-MGIA2720
Subject: Re: Patch for delaying of tasks attache
-Coders,
is there a good reason why a few of the perl scripts we ship contain a
"#!/usr/local/bin/perl" shebang line rather than the default
"#!/usr/bin/perl" we use everywhere else?
In particular, are there any objections against adjusting the path in
local/snmpcheck.def
perl/SNMP/t/bulkwal
On Sun, 2005-10-09 at 08:36 -0700, Ali Majdzadeh wrote:
> I found that with write_variable it is possible to SET
> variables.Is there any example explaining this routine?
Just a couple:
$ cd net-snmp/agent/mibgroup
$ find . -name \*.c | xargs grep ^write | wc -l
185
Did you actually
On Sun, 2005-10-09 at 14:31 +0200, Michael G wrote:
> 1. Is mib-2 support is mandatory?
> Are there any particular groups/objects that must be implemented?
It's not mandatory, in that no-one will be chasing after
you with a large axe if you fail to support it!
Fundamentally, the groups/objects t
On Fri, 2005-10-07 at 19:53 -0400, Robert Story wrote:
> On Fri, 07 Oct 2005 11:41:26 +0200 Thomas wrote:
> TA> Has anyone ever thought of replacing the ".index" handling with
> TA> something better?
>
> I'm sure it was done for simplicity. I wouldn't object to moving it,
> though it will be more
On Mon, 2005-10-10 at 13:06 +0300, Andrei Pisau wrote:
> > > What about the maximum length for the transport?
> > > Exceeding that can also trigger an SNMPERR_TOO_LONG error
> > > (see lines 4782ff of snmp_api.c)
> > >
> >
> > I have checked that lines also. How can it be possible
> > to exceed
Solved.
It was an inconsistency in my libraries.
I had installed libnetsnmp5 and 4 and also I had compiled
net-snmp5.2.2pre1. The libraries from the last were in /usr/local/lib so
I linked with some wrong ones.
Thanks for you help, and don't make the same mistake as I did.
Have a nice day :)
On Fr
Hello guys, I´ve sent an e-mail last week but I´ve got no answers, so I´m
reporting my problem again.
I´ve developed a agent with 2 tables(control table and result table). When I
query the agent with snmpget or snmpset everything goes fine. The problem is
when I use snmptable. If the control tabl
On Fri, 2005-10-07 at 17:39 -0400, Robert Story wrote:
> RS> I was surprised that this didn't include something to deal with
> RS> saving default values for columns, to be used during row creation.
> DS> I'm not at all sure what you mean by that.
> DS> Rather than responding to a (probably erron
Magnus Fromreide wrote:
I haven't seen exactly that case.
I do see that error in the subagent if I either
a) Tries to shut down a agentx subagent or
b) Tries to unregister a registration in a agentx subagent.
And no, they are not the same problem - a) hits even if there are no
registrations
On Fri, 2005-10-07 at 18:00 -0400, Robert Story wrote:
> On Fri, 07 Oct 2005 15:28:06 +0100 Dave wrote:
> DS> One other question has come to light:
> DS>When the current implementation sets up a "monitor" directive,
> DS> it doesn't appear to run the test immediately...
> DS> The new implementa
Hi net-snmp support,
Will the different compiler for net-snmp caused memory leak?
I couldn't reproduce the snmpd leak but my customer did.
I am not sure if the net-snmp version that caused this or the
compilation of net-snmp.
Could you please recommend me which net-snmp version or files to
downlo
On Fri, 2005-10-07 at 20:02 -0400, Robert Story wrote:
> DS> "agentSecName" is for backward compatibility with previous releases.
> DS> "iquerySecName" felt to be a more appropriate name.
>
> Then I strongly suggest that some error checking be introduced to check for
> both being used, and issue a
On Fri, 2005-10-07 at 19:54 -0400, Robert Story wrote:
> On Fri, 07 Oct 2005 10:33:11 +0200 Thomas wrote:
TA> > counter64 support to snmp_add_var (so they can be sent w/snmptrap)
TA>
TA> Can this be applied to the patches branches (5.[12].x) as well?
RS> I have no objections, if you can find anot
On Fri, 2005-10-07 at 20:04 -0400, Robert Story wrote:
> On Fri, 7 Oct 2005 23:02:19 +0100 Dave wrote:
> DS> Perhaps when the chaos of the 5.2.2 and 5.3 releases has
> DS> died down, we could return to this again.
>
> Yeah, right! If I hand a nickel for every issue we've tabled,
> never to be seen
On Fri, 2005-10-07 at 19:51 -0400, Robert Story wrote:
> DS> netsnmp_arch_interface_get_if_speed
>
> That last one, get_if_speed, is linux specific and can be omitted.
> Probably should have named it differently..
OK - Done.
(Feel free to choose a more appropriate name if you prefer).
Dave
-
--- Begin Message ---
On Fri, 2005-10-07 at 16:21 +0100, Dave Shield wrote:
> On Fri, 2005-10-07 at 18:01 +0300, Andrei Pisau wrote:
> > After I do snmp_send I have the snmp_errno = -5 which means
> > SNMPERR_TOO_LONG. But the session field sndMsgMaxSize is 0, so the
> > conditions for snmp_send t
On Sun, Oct 09, 2005 at 10:54:08PM +0200, Thomas Anders wrote:
> -Coders,
>
> CVS MAIN currently produces a
>
> *** glibc detected *** double free or corruption (out): 0x40825838 ***
>
> for me when shutting down the agent (Linux 2.6, x86, non-root). AFAICS
> this only happens if I have the (au
22 matches
Mail list logo
