On Sat, 11 Jun 2005 00:56:48 +0200 Thomas wrote:
TA> Robert Story wrote:
TA> > snmpd -c +/home/bob/my.conf
TA> >
TA> > would cause my.conf to be processed before the default config files.
TA>
TA> At a first glance, I find that counterintuitive. For mib dirs, "+" means
TA> "append" which I'd tr
Robert Story wrote:
snmpd -c +/home/bob/my.conf
would cause my.conf to be processed before the default config files.
At a first glance, I find that counterintuitive. For mib dirs, "+" means
"append" which I'd translate to "be processed *after* the default config
files" here. But that
On Sat, 11 Jun 2005 00:08:31 +0200 Thomas wrote:
TA> Robert Story wrote:
TA> Further down the road, we could also offer nightly snapshots (source or
TA> binary or both) for those who find CVS access (source) and manual
TA> compilation/packaging (binary) too difficult/impossible, but still (we)
T
On Thu, 09 Jun 2005 17:01:56 +0200 Thomas wrote:
TA> I managed to compile net-snmp 5.2.1 on Tru64 Unix 5.1B (Alpha EV5.6)
TA> using Compaq C 6.5. FYI, here's what needed to be done:
TA>
TA> - apply my build fixes (Patch #1217475)
applied.
TA> - ./configure --with-out-mib-modules="mibII/tcpTable
Dave Shield wrote:
I've just commit this to the CVS code base. Can you confirm
whether this seems to fix your problems?
Sure, thanks. My buildbot is back in business.
+Thomas
--
Thomas Anders (thomas.anders at blue-cable.de)
---
This SF.
Robert Story wrote:
On Fri, 10 Jun 2005 11:19:24 +0200 Thomas wrote:
TA> is there any interest in setting up automated build and testing on
TA> multiple platforms to better detect regressions in the code?
It sort of already exists. The dist directory contains several net-snmp-build
(nsb-*) scr
While applying a patch to support detecting gigabit interfaces on linux, I
noticed a bug in the ifSpeed implementation. The ifSpeed object description
states:
An estimate of the interface's current bandwidth in bits
per second.
So, it seems to me that an interface that i
While investigating the engineID problem caused by using -C with -c, i
discovered that optional config files are processed last, and processes PREMIB
and NORMAL config at the same time. It seems to me that this will cause
problems for tokens which might expect that all PREMIB tokens will be dealt
w
Bruce Shaw wrote:
Could you post it as a patch? I'd like to have a look at it. My
code works on my E450's, V880's and 280r's.
Here's what I came up with as a patch (large) to lmSensors.c based on
the test code I posted yesterday. It works on a V1280, but I haven't
tested it further than
On Fri, Jun 10, 2005 at 11:39:19AM -0400, Vance Shipley wrote:
}
} Should this result in the OID of each index being registered?
Ahh ... yes, it does. Never mind. :)
-Vance
---
This SF.Net email is sponsored by: NEC IT Guy Games.
On Fri, 10 Jun 2005 11:39:18 +0200 Thomas wrote:
TA> > Didn't we change code to delay SNMPv3 engine probe recently, Robert?
TA> > May be related.
TA>
TA> s/May be/Is/
TA>
TA> Backing out Robert's change in snmplib/snmp_api.c between revisions 5.73
TA> and 5.74 brings "make test" back to "49 / 49
On Fri, 10 Jun 2005 11:19:24 +0200 Thomas wrote:
TA> is there any interest in setting up automated build and testing on
TA> multiple platforms to better detect regressions in the code?
It sort of already exists. The dist directory contains several net-snmp-build
(nsb-*) scripts that I hacked toge
On Fri, 10 Jun 2005 12:51:29 +0200 Eirik wrote:
EN> The patch solved our problem, thank you.
EN>
EN> > -Original Message-
EN> >
EN> > On Wed, 8 Jun 2005 11:19:07 +0200 Eirik wrote:
EN> > EN> We are running 5.2.1 on a Linux kernel without IPv6. No IPv6 is
EN> > EN> enabled in net-snmp. E
On Fri, 10 Jun 2005 12:37:45 +0100 Dave wrote:
DS> Robert> 1) aren't all new routine supposed to be prefixed with netsnmp_?
DS> Wes> I'm pretty sure we came to a good agreement that netsnmp_
DS> Wes> prefixes were to be used.
DS>
DS> Fair enough.
DS> I wasn't particularly happy with some of the na
I'm using a table_data helper in my agent. I don't seem to be getting
my rows indexed as I get "No Such Instance currently exists at this OID".
In my instrumentation code I register the table:
netsnmp_register_table_data()
Then I add each of the rows:
netsnmp_create_table_data_row()
Hi ,
I am Gagiga NarasimhaChary from Wipro Technologies. I just
want some information regarding the Net – SNMP V3
How net-SNMP supports registering of the TRAPS and sends
response.
How the notification service and SNMP module maintain the
data on who has requested which kind
On Fri, 2005-06-10 at 14:28, Thomas Anders wrote:
> >> Dave> Patch #1109719: Remove (apparently) redundant quoting of
> >> Dave> configure options.
> >
> Wes> I think the problem was that some OSes needed it and some didn't
> Wes> I think that may break other things now. I guess we'll see.
>
Dave Shield wrote:
On Wed, 2005-05-04 at 19:18, Wes Hardaker wrote:
Dave> Patch #1109719: Remove (apparently) redundant quoting of
Dave> configure options.
err... I think the problem was that some OSes needed it and some
didn't...
Possibly.
The individual configure options are (single) qu
On Fri, 2005-06-10 at 11:58, Thorsten Freimann wrote:
> this is what I searched! But it doesnt work :-(
Oh, yes it does!
> I write down this code:
[snip]
I've just copied that code into a file, added the missing
header lines, and got the following:
$ cc z.c `net-snmp-config --libs`
$ ./a.
On Wed, 2005-06-08 at 17:14, Wes Hardaker wrote:
Robert> 1) aren't all new routine supposed to be prefixed with netsnmp_?
Wes> I'm pretty sure we came to a good agreement that netsnmp_
Wes> prefixes were to be used.
Fair enough.
I wasn't particularly happy with some of the names anyway.
Changing
this is what I searched! But it doesnt work :-(
I write down this code:
>>
void do_nothing(const char *word, char *cptr)
{
gnRun++;
printf("run");
return;
}
int main(void)
{
netsnmp_session session, *ss;
struct config_line cl;
netsnmp_ds_s
The patch solved our problem, thank you.
> -Original Message-
> From: Robert Story [mailto:[EMAIL PROTECTED]
> Sent: 8. juni 2005 17:44
> To: Eirik Nordbrøden
> Cc: [email protected]
> Subject: Re: Linux kernel without IPv6 gives log entries
>
> On Wed, 8 Jun 2005 11
Correct.
1) solves our intial problem, since we're "linux" and not "sunV3".
2) came out while I was looking for 1). If you look at the code (or
the diff) you'll notice that the only thing you ever really do with
if_unit is to "string_append_int()" or "strncat()" it to if_name
before the result is
Thomas Anders wrote:
Dave Shield wrote:
On closer examination, the agent seems to be falling over as soon
as it receives the initial SNMPv3 engine probe. It quits with
a SegV fault in 'usm_add_user_to_list', while trying to construct
the REPORT response.
Ring any bells with anyone?
Didn't w
On Fri, 2005-06-10 at 09:35, Gunnar Lindberg wrote:
> Here we go. Two "diff -c" below. They fix:
>
> 1)'if (strcmp(saveName, "ip") == 0)'
> Now put back into the original "#ifdef sunV3" (and only that).
I've applied this to the main CVS code base and the 5.2.x line.
I welcome co
-Coders,
is there any interest in setting up automated build and testing on
multiple platforms to better detect regressions in the code?
It doesn't need to be as ambitious as e.g. the Samba Build Farm
(http://build.samba.org/about.html) (yet), but AFAICS a basic automated
"cvs update && conf
Here we go. Two "diff -c" below. They fix:
1) 'if (strcmp(saveName, "ip") == 0)'
Now put back into the original "#ifdef sunV3" (and only that).
2) The whole notion of (ifnet *)->if_unit is removed.
The entire interface name is now in (ifnet *)->if_name.
As a side
On Thu, 2005-06-09 at 07:39, David T. Perkins wrote:
> > On Wed, 25 May 2005 09:18:35 +0100 Dave wrote:
> > DS> How likely is it that someone would be using an authentication
> > DS> password of "MD5" or "SHA", or an encryption password of "DES"?
> > DS> (suitably extended to cover the full range o
On Thu, 2005-06-09 at 11:55, Thorsten Freimann wrote:
> I want to write a simple application to send traps (similar to the
> snmptrap app from net-snmp). The idea is, that the application
> takes all trapsinks from the snmpd.conf file and send the trap
> to all defined trapsinks.
>
> Is there
On Thu, 2005-06-09 at 19:21, Fong Tsui wrote:
> On Wed, 08 Jun 2005 09:36:29 +0100 Dave wrote:
> DS>> You might be able to short-circuit this by tweaking the
> DS>> var_XXX() code, so that it updates the 'name' parameter
> DS>> to refer to the *last* column in the table. But I'm not
> DS>> 100% su
30 matches
Mail list logo
