No shortage of APIs

2005-05-25 Thread Vance Shipley
Folks, Having just read through April's table_data helper thread I'm back to where I was last week wondering exactly which of the many APIs I should be using. I have an existing application for which I would like to embed a subagent. There are two tables to implement for which there exist data s

Changing the timeout once a session is open

2005-05-25 Thread Toth, Gregory S
We are using the single session api to communicate with SNMP agent. Once we have the session open, can we change the timeouts "on the fly"? struct snmp_session session, *sptr; session.peername = ip; session.community = (u_char*)community; session.community_len = strlen(community);

Row creation with table_data helper

2005-05-25 Thread Vance Shipley
I'm implementing an agent using the table_data helper. One area where I have yet to see the light is in row creation. When a SET arrives for an existing row I can extract my row data using: netsnmp_extract_table_row_data(request); I take it though that this will return NULL when the r

Re: net-snmp on Mac OS X

2005-05-25 Thread OpenMacNews
hi all, i'm building 521 on OSX 10.4.1 as well. the changes above: # defineTCPTV_MIN( 1*PR_SLOWHZ)/* minimum allowable value */ # defineTCPTV_REXMTMAX( 64*PR_SLOWHZ)/* max allowable REXMT # value */ applied to a config: SNMP Versions Supported:1 2

Re: net-snmp on Mac OS X

2005-05-25 Thread Richard J. Zak
I did as you recommended. I had to add the following two lines to tcp.c under agent/mibgroup/ mibii/ #defineTCPTV_MIN( 1*PR_SLOWHZ)/* minimum allowable value */ #defineTCPTV_REXMTMAX( 64*PR_SLOWHZ)/* max allowable REXMT value */ which were from tcp_timer.h in

Re: integrate SuSE 9.3 patches?

2005-05-25 Thread Thomas Anders
Wes Hardaker wrote: On Tue, 24 May 2005 15:43:53 +0200, Thomas Anders <[EMAIL PROTECTED]> said: Thomas> ...Mike has done similar things with the Red Hat... They gave us direct permission to use those patches under a BSD license. If the patches are GPLed in any way (you'll need to ask them) th

Re: print_value(vars->name,vars->name_length,vars)

2005-05-25 Thread Dave Shield
On Wed, 2005-05-25 at 12:39, Alaios wrote: > Thx really a lot. Where i can find the val structure ? Try looking in the header files! (esp. snmp_api.h) > how i can check the oid i want what type is? Look in the relevant MIB file. Dave ---

Re: print_value(vars->name,vars->name_length,vars)

2005-05-25 Thread Alaios
Thx really a lot. Where i can find the val structure ? and how i can check the oid i want what type is? Have a nice day --- Dave Shield <[EMAIL PROTECTED]> wrote: > On Wed, 2005-05-25 at 11:33, Alaios wrote: > > Hi .. i want to know how i can store in to a > variable > > the single value that i get

Re: print_value(vars->name,vars->name_length,vars)

2005-05-25 Thread Dave Shield
On Wed, 2005-05-25 at 11:33, Alaios wrote: > Hi .. i want to know how i can store in to a variable > the single value that i get from this loop > > for(vars=response->variable; vars; > vars=vars->next_variable) long intvalue = *vars->val.integer; > I want an integer variable to hold t

print_value(vars->name,vars->name_length,vars)

2005-05-25 Thread Alaios
Hi .. i want to know how i can store in to a variable the single value that i get from this loop for(vars=response->variable; vars; vars=vars->next_variable) print_value(vars->name,vars->name_length,vars) I want an integer variable to hold the value e.x 1,2 or 3 e.t.c

Re: getting rid of strtok: leftover in perl/SNMP/SNMP.xs [patch]

2005-05-25 Thread Dave Shield
> Thomas> Fixed what? AFAICS, perl/SNMP/SNMP.xs still is *not* fixed in > Thomas> 5.3.dev, unless cvs is playing tricks on me. > > Wes> I suspect it is: In which case it's playing tricks on me as well: $ cd SNMP/net-snmp/perl/SNMP $ cvs -q update $ grep strtok *.xs

Re: snmpv3 Decryption error with net-snmp 5.2.1

2005-05-25 Thread Dave Shield
On Wed, 2005-05-25 at 00:47, Wes Hardaker wrote: > Dave> is there any reason why the command options... > Dave> are the [wrong] way round...? > > No, I caught that a while ago too but far too late. I think I must > have been brain dead and reversed it in my head when I wrote it. Fair enough - I