Configuring contextEngineId and Context Name
Hello, Do we need to configure "context Name" and "context Engine" somewhere in snmpd.conf or it is automatically done for snmpv3 by Net-SNMP I have added following lines in snmpd.conf for authentication : defSecurityName abc defAuthTypeMD5 defSecurityLevel authPriv defAuthPassphrase mysecretpass defPrivType DES defPrivPassphrasemypassphrase defVersion3 and it works fine through command line when i give following : snmpwalk -v3 -u abc -l authPriv -a MD5 -A mysecretpass -x DES -X mypassphrase localhost but through a MIB Browser(GUI) it ask for "context Name" and "Context Engine" ; -- if i keep it empty i get a response : User Authentication Failed -- If i enter something i get a response : Timeout I had searched a lot on this .. but was unable to find it Could u please help in this Br Gaurav -- My life has changed. What about yours? Log on to the new Indiatimes Mail and Live out of the Inbox! - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Passing context name and Engine ID in Request
Hello, Do we need to configure "context Name" and "context Engine" somewhere in snmpd.conf or it is automatically done for snmpv3 by Net-SNMP I have added following lines in snmpd.conf for authentication : defSecurityName abc defAuthTypeMD5 defSecurityLevel authPriv defAuthPassphrase mysecretpass defPrivType DES defPrivPassphrasemypassphrase defVersion3 defContextName context and it works fine through command line when i give following : snmpwalk -v3 -u abc -l authPriv -a MD5 -A mysecretpass -x DES -X mypassphrase localhost but through a MIB Browser(Mibbel GUI) it ask for "context Name" and "Context Engine" ; -- if i keep it empty i get a response -- If i enter something(context) i get a response : Timeout do anything to be done in access specifiers in snmpd.conf I had searched a lot on this .. but was unable to find it Could u please help in this Br Gaurav -- Guess 1st innings score. Win 1 Lakh* Rupees. SMS RUN to . - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ ___ Net-snmp-users mailing list [EMAIL PROTECTED] Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Failed to connect to agentx master agent
Hello ,
I am running the agent as a Agentx configuration
snmpd--- Master Agent
myAgent --- subAgent
But each time i try to run it i get message "Failed to connect to agentX master
agent(localhost:706)
Following are the configuration related to agentx in
/usr/local/share/snmp/snmpd.conf
#
masteragentx
AgentXSocket tcp:localhost:706
#
I am running the snmpd(master) as
/usr/local/sbin/snmpd -c /usr/local/share/snmpsnmpd.conf -X tcp:localhost:706
In the subagent main() method i have added
if (agentx_subagent)
{
netsnmp_ds_set_boolean(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_ROLE, 1);
/*
* AGENTX_ADDRESS is initialized to default port for AgentX.
*/
netsnmp_ds_set_string(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_X_SOCKET, "localhost:706");
syslog(LOG_INFO, "Agent starting as AgentX Subagent.");
}
But when i run the subagent it gives me error as:
Warning: Failed to connect to agentX master agent(localhost:706):
Warning: Failed to connect to agentX master agent(localhost:706):
Please could u provide me some hint whats the problem in this case.
Thanks
Gaurav
--
Guess 1st innings score. Win 1 Lakh* Rupees.
SMS RUN to .
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-users mailing list
[EMAIL PROTECTED]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
RE: Failed to connect to agentx master agent
Hi,
That was mistakenly written but i actually used,
/usr/local/sbin/snmpd -c /usr/local/share/snmpd.conf -X tcp:localhost:706
But anyhow , i was successfull in running it ,
the problem was that in the sub agent code i changed the agentx port from
"localhost:706" to "tcp:localhost:706" and it worked fine
if (agentx_subagent)
{
netsnmp_ds_set_boolean(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_ROLE, 1);
/*
* AGENTX_ADDRESS is initialized to default port for
AgentX.
*/
netsnmp_ds_set_string(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_X_SOCKET,
"tcp:localhost:706");
Br
Gaurav
- Original Message -
From: Jayaprakasha Guddenahalli Naganna <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED], [EMAIL PROTECTED], net-snmp-coders@lists.sourceforge.net
Sent: Fri, 31 Aug 2007 21:04:58 +0530 (IST)
Subject: RE: Failed to connect to agentx master agent
>>/usr/local/sbin/snmpd -c /usr/local/share/snmpsnmpd.conf -X
>>tcp:localhost:706
I think you have given wrong configuration file name "snmpsnmpd.conf".
Please check the file name.
Thanks
-JP
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]
Sent: Wednesday, August 29, 2007 10:50 AM
To: [EMAIL PROTECTED];
net-snmp-coders@lists.sourceforge.net
Subject: Failed to connect to agentx master agent
Hello ,
I am running the agent as a Agentx configuration
snmpd--- Master Agent
myAgent --- subAgent
But each time i try to run it i get message "Failed to connect to agentX
master agent(localhost:706)
Following are the configuration related to agentx in
/usr/local/share/snmp/snmpd.conf
#
masteragentx
AgentXSocket tcp:localhost:706
#
I am running the snmpd(master) as
/usr/local/sbin/snmpd -c /usr/local/share/snmpsnmpd.conf -X
tcp:localhost:706
In the subagent main() method i have added
if (agentx_subagent)
{
netsnmp_ds_set_boolean(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_ROLE, 1);
/*
* AGENTX_ADDRESS is initialized to default port for
AgentX.
*/
netsnmp_ds_set_string(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_X_SOCKET,
"localhost:706");
syslog(LOG_INFO, "Agent starting as AgentX Subagent.");
}
But when i run the subagent it gives me error as:
Warning: Failed to connect to agentX master agent(localhost:706):
Warning: Failed to connect to agentX master agent(localhost:706):
Please could u provide me some hint whats the problem in this case.
Thanks
Gaurav
--
Guess 1st innings score. Win 1 Lakh* Rupees.
SMS RUN to .
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-users mailing list
[EMAIL PROTECTED]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
--
Guess 1st innings score. Win 1 Lakh* Rupees.
SMS RUN to .
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Able to send v2 traps but Issue in sending v3 traps
Hello, I am facing problem sending SNMPv3 traps to the trap listener (snmptrad , AdventNet Trap Listener) from the agent which i have compiled as Master - subagent configuration(Agentx) Following is the configuration i did in /usr/local/share/snmp/snmpd.conf com2sec v3_userlocalhost v3_private group v3_group usm v3_user rwuser v3_user auth viewall included .1 80 accessv3_group"" any noauth exact allall none master agentx AgentXSockettcp:localhost:706 trapsess -e 0x0102030405 -v 3 -u v3_user -a MD5 -A mysecretpass -l authNoPriv localhost:162 snmpsess -e 0x0102030405 -v 3 -u v3_user -a MD5 -A mysecretpass -l authNoPriv localhost:162 = Following is the configuration for /usr/local/share/snmp/snmptrapd.conf == logoption o ignoreauthfailure no authCommunity log v3_private == Following lines were added in /var/net-snmp/snmpd.conf & /var/net-snmp/snmptrapd.conf === snmpd.conf > createUser v3_user MD5 mysecretpass DES mypassphrase snmptrapd.conf > createUser 0x0102030405 v3_user MD5 mysecretpass DES mypassphrase === Problem is that when i run "snmptrapd -Dusm -P" and the run my agent from where i send trap using "send_v2trap()" it gives me following error and no traps are recieved USM processing begun... usm: usm: match on user v3_user match on user v3_user usm: no match on engineID (80 00 1F 88 04 53 54 52 49 4E 47 usm: no match on engineID (80 00 1F 88 04 53 54 52 49 4E 47 ) ) usm: usm: match on user v3_user match on user v3_user usm: usm: Verification succeeded. Verification succeeded. usm: usm: USM processing completed. USM processing completed. Could u help me in this case as when i use "trap2sink localhost v3_private" directive in snmpd.conf then i am able to recieve the traps but these are V2 trap.How to get V3 traps Thanks Gaurav -- My life has changed. What about yours? Log on to the new Indiatimes Mail and Live out of the Inbox! - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Possibility of HardCoding engine-ID for each USM user??
Hello ,
Could it be possible that i can hardcode the engine id for each user i create
i.e
In /var/net-snmp/snmpd.conf
createUser -e 0x0102030405 v3_user MD5 mysecretpass DES mypassphrase
This lines in snmpd.conf is replaced by an encrypted key , but what i suppose
that the entry for this user does not get created in USM table
as when i hit snmpget request as:
snmpget -e 0x0102030405 -v v3_user -l authNoPriv -a MD5 -A mysecretpass
localhost sysUptime.0
RESPONSE = "Timeout : No Response from localhost"
snmpget -v v3_user -l authNoPriv -a MD5 -A mysecretpass localhost sysUptime.0
RESPONSE = snmpget : Unkwon user Name
Whereas for TRAPS:
Though i read that we need to hardcode the engineID for sending trap in
snmd.conf as the engineID of the engine *sending* the trap PDU and the trap
reciever would need to be configured with appropriate users for each trap
generator with the hardcoded engineID as send by the agent
{ [ -- FROM EMAIL ARCHIVE NET-SNMP FORUM -- ] If you are sending an SNMPv3 trap
PDU, then the engineID should be that of the engine *sending* the trap (and the
trap receiver would need to be configured with appropriate users for each trap
generator. If you are sending an SNMPv3 inform PDU, then the engine ID should
be that of the engine *receiving* the trap.}
so i hardcoded the engine-id at agent in /usr/local/share/snmp/snmpd.conf as :
trapsess -e 0x0102030405 -v 3 -u v3_user -a MD5 -A mysecretpass -l authNoPriv
I was able to recieve the traps
Could u explain why so and can we hardcode the engineID for the Users while
creating
Br
Gaurav
--
My life has changed. What about yours?
Log on to the new Indiatimes Mail and Live out of the Inbox!
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
addding new fd in select() for recieving request
Hello,
I have a requirement that i also need to listen for alarms coming from another
application on socket and forward them to manager apart from GET/SET request
from Manager
Now in order to listen on socket for alarms i need to accept the connection but
this will make it blocking. So i have to add it in select() to hear whether it
is ready to be read or not and then accept connection.
I am thinking of doing it in following manner.
Please comment on the following if it is wrong
This code will be main
main()
{
fd_set master;
fd_set read_fds;
int listener;
int fdmax;
int newfd;
FD_ZERO(&master);
FD_ZERO(&read_fds);
listener = socket(AF_INET, SOCK_STREAM, 0);
bind(listener, (struct sockaddr *)&myaddr, sizeof(myaddr));
listen(listener, 10)
// add the listener to the master set
FD_SET(listener, &master);
fdmax = listener;
while()
{
read_fds = master;
snmp_select_info(&numfds, &read_fds, NULL, 0);
select(numfds,&read_fds,0,0,NULL)
if (FD_ISSET(listener , &read_fds))
{
newfd = accept(listener, (struct sockaddr
*)&remoteaddr,&addrlen)
//function1 --->reads the data add it to queue and forward to
manager as a TRAP
}
}
}
But this does'nt seem to work , do i need to register my function1.Please help
me on this
Thanks
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Logging of subagent to different file
Hello, Could it be possible for creating a different log file for subagent and manage the log level and type of log messages to be logged. What i understood was that all the logs are written to /var/log/snmpd.log , but is it right that 2 different process master agent and subagen write to this same file Thanks - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Timeout on GET request from second Manager
Hello, Do NEtSnmp support request from multiple managers I developed SNMPv3 master-subagent. I started a first MibBrowser(Acting as first Manager) and sent GET rtequest , it was successful Then i started second MibBrowser(Acting as second Manager) and sent GET request , it TimedOut. Do i need to do some configuration or netsnmp not at all support this requirement I also need to sent taps to these managers How to do it. My snmd.conf is ### rwuser sbssEmsUser auth com2sec sbssEmsUser 192.168.150.5 sbssEmsCommPrivate view sbssEmsAll included .1 80 access sbssEmsUserGroup "" usm auth exact sbssEmsAll sbssEmsAll sbssEmsAll master agentx AgentXSockettcp:localhost:706 trap2sink 192.168.150.5 sbssEmsCommPrivate 162 informsink 192.168.150.5 sbssEmsCommPrivate 162 trapcommunity sbssEmsCommPrivate - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Logging of subagent to different file
OK i am using Agentx Configuration
as a read somewhere in the forum that :
###
If your subagent understands the standard command-line arguments,
then you can use the -Lf option to specify the appropriate log file to use.
Otherwise, you could either code your subagent to use the
snmp_log_options() routine (to handle command-line options),
or extract the relevant code from this routine.
You basically need to set up a NETSNMP_LOGHANDLER_FILE
style log handler, and specify the name of the file to log to
(logh->token)
i tried using this as
logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
if (logh)
{
logh->pri_max = pri_max;
logh->token = strdup("/root/myAgentx.log");
netsnmp_enable_filelog(logh,
netsnmp_ds_get_boolean(NETSNMP_DS_LIBRARY_ID,
NETSNMP_DS_LIB_APPEND_LOGFILES));
}
but when i used API log file was created but logs for DEBUGMSG("Started") were
found in the file
then i also added
debug_register_tokens("snmpV3Agent,currentAlarmTable");
snmp_set_do_debugging(1);
Then the logs start coming but they were also shown on stdout(console) which i
don,t won't so could it be possible that my subagent(Agentx) logs come in
different file and master agent in different file
Thanks
- Original Message -
From: Magnus Fromreide <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: net-snmp-coders@lists.sourceforge.net
Sent: Fri, 13 Jun 2008 11:44:56 +0530 (IST)
Subject: Re: Logging of subagent to different file
On Tue, 2008-06-10 at 23:28 +0530, [EMAIL PROTECTED] wrote:
> Hello,
>
> Could it be possible for creating a different log file for subagent
> and manage the log level and type of log messages to be logged.
Eh, it already is?
> What i understood was that all the logs are written
> to /var/log/snmpd.log , but is it right that 2 different process
> master agent and subagen write to this same file
Lets assume that you have the programs
snmpd (generated from ./configure ; make ; make install) and subagent
(generated from net-snmp-config --compile-subagent subagent subagent1.c
subagent2.c)
then
snmpd -Lf /tmp/my.snmpd.log.file
will send all logs from snmpd to /tmp/my.snmpd.log.file
and
subagent -Lf /tmp/my.subagent.log.file
will send all logs from subagent to /tmp/my.subagent.log.file.
/MF
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Logging of subagent to different file
i tried using this as
netsnmp_log_handler *logh;
logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
if (logh)
{
logh->pri_max = pri_max;
logh->token = strdup("/root/myAgentx.log");
netsnmp_enable_filelog(logh,
netsnmp_ds_get_boolean(NETSNMP_DS_LIBRARY_ID,
NETSNMP_DS_LIB_APPEND_LOGFILES));
}
but when i used API log file was created but logs for DEBUGMSG("Started") were
found in the file
then i also added
debug_register_tokens("snmpV3Agent,currentAlarmTable");
snmp_set_do_debugging(1);
Then the logs start coming but they were also shown on stdout(console) which i
don,t won't so could it be possible that my subagent(Agentx) logs come in
different file and master agent in different file
Thanks
- Original Message -
From: Dave Shield <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: net-snmp-coders@lists.sourceforge.net, [EMAIL PROTECTED]
Sent: Wed, 11 Jun 2008 14:15:34 +0530 (IST)
Subject: Re: Logging of subagent to different file
2008/6/10 <[EMAIL PROTECTED]>:
> Could it be possible for creating a different log file for subagent
> and manage the log level and type of log messages to be logged.
If your subagent understands the standard command-line arguments,
then you can use the -Lf option to specify the appropriate log file to use.
Otherwise, you could either code your subagent to use the
snmp_log_options() routine (to handle command-line options),
or extract the relevant code from this routine.
You basically need to set up a NETSNMP_LOGHANDLER_FILE
style log handler, and specify the name of the file to log to
(logh->token)
Dave
PS: Can you please post requests to *one* of the mailing lists,
not both. Thanks
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Timeout on GET request from second Manager
Hello These managers are configured for V3 request I sent same GET request to retrieve the value of alarmID from the alarmtable and the User was "sbssEmsUser" My snmd.conf: -- trapsess -e 0x80001f888017454900b6f44848 -v 3 -u sbssEmsUser -a MD5 -A sbssEmsSecretpass -l authNoPriv 192.168.150.5:162 rwuser sbssEmsUser auth com2sec sbssEmsUser 192.168.150.5 sbssEmsCommPrivate group sbssEmsUserGroup usm sbssEmsUser view sbssEmsAll included .1 80 access sbssEmsUserGroup "" usm auth exact sbssEmsAll sbssEmsAll sbssEmsAll master agentx AgentXSockettcp:localhost:706 That is i should be able to recieve request from multiple managers One more question related to configuration were in agent we configure from which IP to listen for request or we just create user , it's group and view access Thanks a lot - Original Message - From: Dave Shield <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: net-snmp-coders@lists.sourceforge.net Sent: Fri, 13 Jun 2008 14:21:42 +0530 (IST) Subject: Re: Timeout on GET request from second Manager 2008/6/13 <[EMAIL PROTECTED]>: > Do NEtSnmp support request from multiple managers Yes. > I developed SNMPv3 master-subagent. > I started a first MibBrowser(Acting as first Manager) and sent GET rtequest , > it was successful > Then i started second MibBrowser(Acting as second Manager) and sent GET > request , it TimedOut. How are these two browsers configured? What version of SNMP and what community/user are they using? What were the two GET requests? > My snmd.conf is > rwuser sbssEmsUser auth > com2sec sbssEmsUser 192.168.150.5 sbssEmsCommPrivate No. Do not use the same security name for an SNMPv3 user and an SNMPv1/2c community. It *might* work, but I'm not convinced. And it's just asking for confusion. Keep different access control settings distinct > view sbssEmsAll included .1 80 > access sbssEmsUserGroup "" usm auth exact sbssEmsAll sbssEmsAll sbssEmsAll You haven't defined 'sbssEmsUserGroup' (and I assume that you've created the SNMPv3 user 'sbssEmsUser' in the /var/net-snmp/snmpd.conf file?) Why not simply use rocommunity sbssEmsCommPrivate ? That's much simpler! > trap2sink 192.168.150.5 sbssEmsCommPrivate 162 > informsink 192.168.150.5 sbssEmsCommPrivate 162 Do you really want to send two copies of every trap to the same management station? > trapcommunity sbssEmsCommPrivate The trapcommunity needs to come *before* the sink lines. Dave - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Logging of subagent to different file
The priority was set to LOG_DEBUG
sorry for the mistake
but when i used API log file was created but logs for DEBUGMSG("Started") were
"NOT" found in the file
The problem is that if i don't add following lines in code the logs for
DEBUGMSG("Started") and DEBUGMSGTL("snmpV3Agent","Started") do not come but my
logfile is created
debug_register_tokens("snmpV3Agent,currentAlarmTable");
snmp_set_do_debugging(1);
but on adding these above lines log start coming in my logfile but also come on
console
code piece:
-
debug_register_tokens("snmpV3Agent,currentAlarmTable");
snmp_set_do_debugging(1);
/* Logging to a different file */
logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
if (logh)
{
logh->pri_max = pri_max;
logh->token = strdup("/root/AGENT/Agent.log");
netsnmp_enable_filelog(logh,
netsnmp_ds_get_boolean(NETSNMP_DS_LIBRARY_ID,
NETSNMP_DS_LIB_APPEND_LOGFILES));
}
if (agentx_subagent)
{
DEBUGMSG(("Using as Agentx SubAgent\n"));
DEBUGMSGTL(("snmpV3Agent", "Using as Agentx SubAgent\n"));
netsnmp_ds_set_boolean(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_ROLE, 1);
netsnmp_ds_set_string(NETSNMP_DS_APPLICATION_ID,
NETSNMP_DS_AGENT_X_SOCKET, AGENTX_ADDRESS);
DEBUGMSGTL(("snmpV3Agent","Agent starting as AgentX
Subagent.\n"));
}
--
I am attaching my main file.I actually wanted that logs of mysubagent(AgentX)
come in mylogfile and master agent in /var/logs/snmpd.log
Thanks a Lot
- Original Message -
From: Dave Shield <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: net-snmp-coders@lists.sourceforge.net
Sent: Fri, 13 Jun 2008 18:46:56 +0530 (IST)
Subject: Re: Logging of subagent to different file
2008/6/13 <[EMAIL PROTECTED]>:
> i tried using this as
>
> netsnmp_log_handler *logh;
> logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
"priority" ?
What value does this have?
> but when i used API log file was created but logs for DEBUGMSG("Started")
> were found in the file
Isn't that what you wanted?
For messages to be logged in this file?
> then i also added
>debug_register_tokens("snmpV3Agent,currentAlarmTable");
>snmp_set_do_debugging(1);
>
> Then the logs start coming but they were also shown on stdout(console) which
> i don,t won't
Where in your subagent code did you add these lines?
It's almost impossible to comment on what might be happening,
given such an isolated fragment of code.
> so could it be possible that my subagent(Agentx) logs come in
> different file and master agent in different file
Yes - both Magnus and I have already told you that this
is possible.
But without a better description of what you are actually doing,
there's no way that we can help you find out what's going wrong.
Dave
snmpV3Agent.c
Description: Binary data
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Logging of subagent to different file
Thanks it worked fine setting the log handler *before* activating the debug
code.
Thanks
- Original Message -
From: Dave Shield <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: net-snmp-coders@lists.sourceforge.net
Sent: Sat, 14 Jun 2008 18:31:28 +0530 (IST)
Subject: Re: Logging of subagent to different file
2008/6/14 <[EMAIL PROTECTED]>:
> The problem is that if i don't add following lines in code the logs for
> DEBUGMSG("Started")
> and DEBUGMSGTL("snmpV3Agent","Started") do not come but my logfile is created
>
>debug_register_tokens("snmpV3Agent,currentAlarmTable");
>snmp_set_do_debugging(1);
>
> but on adding these above lines log start coming in my logfile but also come
> on console
>
> code piece:
> -
>debug_register_tokens("snmpV3Agent,currentAlarmTable");
>snmp_set_do_debugging(1);
>
>/* Logging to a different file */
>logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
Try setting up the log handler *before* activating the debug code.
If you look at the output currently produced, my guess is that
it probably starts with:
No log handling enabled - turning on stderr logging
Dave
-
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Timeout on GET request from second Manager
When the request from AdventNet Mib Browser configured for V3 request , then request from second Mibbrowser timed out But from command line both were successfull >From Machine 1: === snmpget -v 3 -u sbssEmsUser -l authPriv -a MD5 -A sbssEmsSecretPass -x DES -X sbssEmsPassphrase localhost HNS-MSV-3GSBSS-EMS-OSS-MIB::currentAlarmID.0 RESPONSE : SUCCESS >From Machine 2: === snmpget -v 3 -u sbssEmsUser -l authPriv -a MD5 -A sbssEmsSecretPass -x DES -X sbssEmsPassphrase 192.168.150.6 HNS-MSV-3GSBSS-EMS-OSS-MIB::currentAlarmID.0 RESPONSE : SUCCESS Thanks - Original Message - From: Dave Shield <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: net-snmp-coders@lists.sourceforge.net Sent: Fri, 13 Jun 2008 18:49:29 +0530 (IST) Subject: Re: Timeout on GET request from second Manager 2008/6/13 <[EMAIL PROTECTED]>: > These managers are configured for V3 request > I sent same GET request to retrieve the value of alarmID from the alarmtable > and the User was "sbssEmsUser" So you sent the exact same request from two client systems to the same SNMP agent. One worked, and one timed out. Is that correct? Can you post the *exact* commands that you used in each case? Dave - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Multithreaded : IS IT Safe to create new thread for updating CONTAINERS
Hello, I read NET_SNMP is single threaded and does not claim to be thread safe either. Now I have requirement to periodically fetch all the alarms related data from DB and update the NETSNMP CONTAINER from which GET/GET NEXT/GET BULK request will return data. I have implemented this in following way. I created a new thread and make that thread sleep for some time and after it resumes it fetches the data from DB (this happens continously in while(1)). Now in this thread only,i updated the NetSNMP CONTAINER using CONTAINER.INSERT() and took a mutex before updating it. But the cacth here is while updation is going on and mutex lock is there and some GET/GET NEXT/GET BULK request comes so it won't be able to access CONTAINER and will timeout/ error will be thrown. So how to handle this. Is my approach right Thanks - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Multithreaded : IS IT Safe to create new thread for updating CONTAINERS
Hello, Yes you are right that lock should be taken on all areas where the shared CONTAINER is being used/accessed I have a query related to this only I have alarmTable mib module The alarmTable CONTAINER will be accessed by 2 different threads 1) netsnmp thread for GET/SET operations 2) my subagent(master-subagent configuration) thread which will continously updating this container with new alarms recieved As this is a table so using mib2c.array.conf it gave me folowing methods: currentAlarmTable_cmp() currentAlarmTable_get() init_currentAlarmTable() currentAlarmTable_row_copy() currentAlarmTable_extract_index() currentAlarmTable_can_delete() currentAlarmTable_duplicate_row() currentAlarmTable_delete_row() currentAlarmTable_set_reserve1() currentAlarmTable_set_reserve2() currentAlarmTable_set_action() currentAlarmTable_set_commit() currentAlarmTable_set_free() currentAlarmTable_set_undo() initialize_table_currentAlarmTable() currentAlarmTable_get_value() currentAlarmTable_get_by_idx() updateCurrentAlarmTable_MyOwnMethod()--->In this method i do CONTAINER.INSERT() So in order to have syncronization i am using mutex and condition variable. But i am not sure where all to use mutexs. I am using mutexes in following places : updateCurrentAlarmTable_MyOwnMethod() initialize_table_currentAlarmTable() Do i also need to take lock on following methods provided by netsnmp currentAlarmTable_set_reserve1() currentAlarmTable_set_reserve2() currentAlarmTable_set_action() currentAlarmTable_set_commit() currentAlarmTable_set_free() currentAlarmTable_set_undo() currentAlarmTable_get_value() as they also access CONTAINER - Original Message - From: Dave Shield <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: net-snmp-coders@lists.sourceforge.net Sent: Fri, 20 Jun 2008 14:30:34 +0530 (IST) Subject: Re: Multithreaded : IS IT Safe to create new thread for updating CONTAINERS 2008/6/19 <[EMAIL PROTECTED]>: > I created a new thread and... in this thread only,i updated the NetSNMP > CONTAINER using > CONTAINER.INSERT() and took a mutex before updating it. That's half of the task, yes. But as I understand threading (and I am *not* an expert in this field), you will also need to have a corresponding lock around the code that *uses* this container. I.e. the code for your MIB module (and any helpers that it might rely on). You will need to look closely at the code that implements your MIB module, to see exactly where the container is referenced, and hence the region you need to protect with the lock. Dave - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Re: Multithreaded : IS IT Safe to create new thread for updating CONTAINERS
I agree to your point that we need to take lock on every place of my mib module (whose code is generated by mib2c) which access the CONTAINER (*_set_reserve1() , *_set_reserve2()., *_set_free(), *_set_undo(),*_get_value()...) But could u just let me know that do in any case locking to be done in netsnmp code and api's. Br Gaurav - Original Message - From: Dave Shield <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: net-snmp-coders@lists.sourceforge.net Sent: Mon, 30 Jun 2008 13:38:44 +0530 (IST) Subject: Re: Multithreaded : IS IT Safe to create new thread for updating CONTAINERS 2008/6/30 <[EMAIL PROTECTED]>: > . But i am not sure where all to use mutexs. > I am using mutexes in following places : > updateCurrentAlarmTable_MyOwnMethod() > initialize_table_currentAlarmTable() > > Do i also need to take lock on following methods provided by netsnmp > currentAlarmTable_set_reserve1() > currentAlarmTable_set_reserve2() > currentAlarmTable_set_action() > currentAlarmTable_set_commit() > currentAlarmTable_set_free() > currentAlarmTable_set_undo() > currentAlarmTable_get_value() > as they also access CONTAINER Bear in mind that I am not an expert in multi-threaded programming. But I would expect that you would need a mutex lock around *every* bit of code that works with row data obtained from the table container, That's the code in your MIB module *and* any similar code within the relevant handler routines. I am not familiar with the module framework you are using, but I would expect that most of the places you mention would indeed need to be protected. The question to bear in mind throughout is: "would it matter if updateCurrentAlarmTable_MyOwnMethod() cleared all the rows from the table at this precise moment?" If the table being emptied would break a given section of code, then that code needs to be protected. Or at least, that's my understanding. Someone who actually knows about such things might be able to advise you differently. Dave - Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Control Logging Priorities(LOG_DEBUG , LOG_CRIT) for different log file
Hello,
I want to do logging of my subagent(Master[snmpd] - subagent[myAgent]
configuration) to a different file for this i have created a new handler and
registered my tokens :
:
=
netsnmp_log_handler *logh;
int priority = LOG_DEBUG;
int pri_max = LOG_EMERG;
/* Logging to a different file */
logh = netsnmp_register_loghandler(NETSNMP_LOGHANDLER_FILE, priority);
if (logh)
{
logh->pri_max = pri_max;
logh->token = strdup("mySubAgent.log");
netsnmp_enable_filelog(logh,
netsnmp_ds_get_boolean(NETSNMP_DS_LIBRARY_ID,
NETSNMP_DS_LIB_APPEND_LOGFILES));
}
debug_register_tokens("snmpV3Agent,currentAlarmTable");
snmp_set_do_debugging(1);
=
:
and then use DEBUGMSGTL() to enter logs, e.g :
:
=
DEBUGMSGTL(("snmpV3Agent", "Using as Agentx SubAgent\n"));
=
:
But i want to do logging of my subagent to a different file for various
priorities and control that priority through config file i.e
LOG_DEBUG , LOG_INFO , LOG_WARNING , LOG_CRIT
i.e While debugging just enable log messages for LOG_DEBUG and in production
enable log messages for LOG_CRIT and LOG_WARNING.
Which api to use for this ( shall i use snmp_log or DEBUGMSGTL api) and how to
control logging through configuration
Thanks
-
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Updating existing rows in CONTAINER
Hello,
I have a table which have index ranging from 1-1000 , once the index reaches
1000 it is rounded off back to 1.
Now in such case the existing rows with that index need to be updated.
1) Can i safely go ahead with finding a row with the index(rounded off back to
1) and simpy updating it.
i.e
CONTAINER_FIND(cb.container,index);
if(found)
{
simply update the row
}
2) Or I first need to delete that particular row from table with the
index(rounded off back to 1) , allocate new memory and then insert the row with
same index
i.e
CONTAINER_FIND(cb.container,index);
if(found)
{
delete row
SNMP_MALLOC_TYPEDEF() //allocate memory
CONTAINER_INSERT() //insert the row with same index
}
As i read in forum that:
"modifying pointers after they've been inserted, without removing them is NOT
allowed options for a container."
Thanks
Gaurav
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Issues while handling data in CONTAINER
Hello, 1)When ever i insert a row in CONTAINER i get error --- "error on subcontainer '' insert (-1)" What could be the reason for this. This happens always when i insert a row 2) When deleting a row from CONTAINER(array-user) , can i simply call CONTAINER_REMOVE(cb , ctx) or i need to call _delete_row() method generated by mib2c after calling CONTAINER_REMOVE(cb , ctx) - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Regarding error on subcontainer '' insert (-1)
Hello, We are getting error in log file for every insert, delete on NetSnmp container as --> "error on subcontainer '' insert (-1)" it is flling up the log file unneccesary. I went through net-snmp mailing list and came to know that: 1) It is likely one of default network-related agent/mib with net-snmp which is causing the issue 2) ipAddressTable could not deal with the interfaces with the same IP(The indexes for this table are ipAddressAddrType and ipAddressAddr. That means that an ipAddress on multiple interfaces will not have a unique index.) , but our interfaces are not having same IP 3) i even tried to disable this mib by sing: snmpd -I-ip... but still the error is coming in log file Could anyone help me out in this. Or is there any patch available.We are using net-snmp release 5.4.1 Thanks Gaurav -- Check out the Mahindra Xylo inside out with reviews, pictures, video and more.Exclusively on ZigWheels.com http://www.zigwheels.com/b2cam/mahindra-xylo.html -- Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
Issue with sending Traps as fast rate
Hello,
We are using netsnmp release 5.4.1. We are facing issue with trap sending
We are sending trap at a fast rate 1000 Alarms/sec. The Agent hangs. Adding
some debug statement , we found that the code hangs when we give call to
send_v2trap(var_list);
At our trap receiver we received 500 Alarms and then we saw no traps (as agent
which was forwarding traps was hanged) . Could you suggest what is the issue.
Following is the configuration in snmpd.conf for sending trap:
trapsess -e -v 3 -u -a MD5 -A -l authNoPriv :
Following is the code used for sending trap:
<>
int
send_alarmChanged_trap(struct alarmdata trapData)
{
netsnmp_variable_list *var_list = NULL;
oid alarmChanged_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 3, 6 };
oid currentAlarmIndex_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 1
/* insert index here */ };
oid currentAlarmNotificationID_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 2
/* insert index here */ };
oid currentAlarmID_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 3
/* insert index here */ };
oid currentAlarmEventType_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 4
/* insert index here */ };
oid currentAlarmSeverity_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 5
/* insert index here */ };
oid currentAlarmReportingElement_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 6
/* insert index here */ };
oid currentAlarmFaultElement_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 7
/* insert index here */ };
oid currentAlarmName_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 8
/* insert index here */ };
oid currentAlarmProbableCause_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 9
/* insert index here */ };
oid currentAlarmAdditionalInformation_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 10
/* insert index here */ };
oid currentAlarmChangedTime_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 12
/* insert index here */ };
oid ipAddress_oid[] = {1, 3, 6, 1, 6, 3, 18, 1, 3 };
char ip_address_buffer[128];
struct sockaddr_in my_addr;
OSSAgentConfiguration::getAgentAddress(ip_address_buffer);
inet_aton(ip_address_buffer, &(my_addr.sin_addr));
/*
* Set the snmpTrapOid.0 value
*/
snmp_varlist_add_variable(&var_list,
snmptrap_oid, OID_LENGTH(snmptrap_oid),
ASN_OBJECT_ID,
(const u_char *)&alarmChanged_oid,
sizeof(alarmChanged_oid));
/*
* Add any objects from the trap definition
*/
snmp_varlist_add_variable(&var_list,
currentAlarmIndex_oid,
OID_LENGTH(currentAlarmIndex_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmIndex,
sizeof(trapData.alarmIndex));
snmp_varlist_add_variable(&var_list,
currentAlarmNotificationID_oid,
OID_LENGTH(currentAlarmNotificationID_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmNotificationId,
sizeof(trapData.alarmNotificationId));
snmp_varlist_add_variable(&var_list,
currentAlarmID_oid,
OID_LENGTH(currentAlarmID_oid), ASN_INTEGER,
(const u_char *)&trapData.alarmId,
sizeof(trapData.alarmId));
snmp_varlist_add_variable(&var_list,
currentAlarmEventType_oid,
OID_LENGTH(currentAlarmEventType_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmEventType,
sizeof(trapData.alarmEventType));
snmp_varlist_add_variable(&var_list,
currentAlarmSeverity_oid,
OID_LENGTH(currentAlarmSeverity_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmSeverity,
sizeof(trapData.alarmSeverity));
snmp_varlist_add_variable(&var_list,
currentAlarmReportingElement_oid,
OID_LENGTH(currentAlarmReportingElement_oid),
ASN_OCTET_STR,
(const u_char *)alarmReportingElement,
strlen(alarmReportingElement));
snmp_varlist_add_variable(&var_list,
currentAlarmFaultElement_oid,
OID_LENGTH(currentAlarmFaul
Fw: Issue with sending Traps as fast rate
Hello,
I am wondering wheter there is something related to adding up varbid in varlist
using api
snmp_varlist_add_variable() or issue with send_v2trap(). Because when traps are
send at fast rate 1000 alarm/sec then the code hangs somewhere in send_v2trap()
api.
Could any body guide me on this.
Thanks and Regards
Gaurav
- Original Message -
From: gauravmeh2...@indiatimes.com
To: D T Shield , dave shield
Cc: net-snmp-coders@lists.sourceforge.net
Sent: Wed, 9 Sep 2009 11:12:25 +0530 (IST)
Subject: Issue with sending Traps as fast rate
Hello,
We are using netsnmp release 5.4.1. We are facing issue with trap sending
We are sending trap at a fast rate 1000 Alarms/sec. The Agent hangs. Adding
some debug statement , we found that the code hangs when we give call to
send_v2trap(var_list);
At our trap receiver we received 500 Alarms and then we saw no traps (as agent
which was forwarding traps was hanged) . Could you suggest what is the issue.
Following is the configuration in snmpd.conf for sending trap:
trapsess -e -v 3 -u -a MD5 -A -l authNoPriv :
Following is the code used for sending trap:
<>
int
send_alarmChanged_trap(struct alarmdata trapData)
{
netsnmp_variable_list *var_list = NULL;
oid alarmChanged_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 3, 6 };
oid currentAlarmIndex_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 1
/* insert index here */ };
oid currentAlarmNotificationID_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 2
/* insert index here */ };
oid currentAlarmID_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 3
/* insert index here */ };
oid currentAlarmEventType_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 4
/* insert index here */ };
oid currentAlarmSeverity_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 5
/* insert index here */ };
oid currentAlarmReportingElement_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 6
/* insert index here */ };
oid currentAlarmFaultElement_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 7
/* insert index here */ };
oid currentAlarmName_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 8
/* insert index here */ };
oid currentAlarmProbableCause_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 9
/* insert index here */ };
oid currentAlarmAdditionalInformation_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 10
/* insert index here */ };
oid currentAlarmChangedTime_oid[] =
{ 1, 3, 6, 1, 4, 1, 303, 3, 3, 9, 36, 3, 3, 2, 2, 1, 12
/* insert index here */ };
oid ipAddress_oid[] = {1, 3, 6, 1, 6, 3, 18, 1, 3 };
char ip_address_buffer[128];
struct sockaddr_in my_addr;
OSSAgentConfiguration::getAgentAddress(ip_address_buffer);
inet_aton(ip_address_buffer, &(my_addr.sin_addr));
/*
* Set the snmpTrapOid.0 value
*/
snmp_varlist_add_variable(&var_list,
snmptrap_oid, OID_LENGTH(snmptrap_oid),
ASN_OBJECT_ID,
(const u_char *)&alarmChanged_oid,
sizeof(alarmChanged_oid));
/*
* Add any objects from the trap definition
*/
snmp_varlist_add_variable(&var_list,
currentAlarmIndex_oid,
OID_LENGTH(currentAlarmIndex_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmIndex,
sizeof(trapData.alarmIndex));
snmp_varlist_add_variable(&var_list,
currentAlarmNotificationID_oid,
OID_LENGTH(currentAlarmNotificationID_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmNotificationId,
sizeof(trapData.alarmNotificationId));
snmp_varlist_add_variable(&var_list,
currentAlarmID_oid,
OID_LENGTH(currentAlarmID_oid), ASN_INTEGER,
(const u_char *)&trapData.alarmId,
sizeof(trapData.alarmId));
snmp_varlist_add_variable(&var_list,
currentAlarmEventType_oid,
OID_LENGTH(currentAlarmEventType_oid),
ASN_INTEGER,
(const u_char *)&trapData.alarmEventType,
sizeof(trapData.alarmEventType));
snmp_varlist_add_variable(&var_list,
currentAlarmSeverity_oid,
OID_LENGTH(currentAlarmSeverity_oid),
ASN_INTEGER,
(c
Issue sending traps (Is it Master-subAgent Configuration usingAgentx ?? )
Hello, We are using Using net-snmp release 5.4.1 Facing issue with sending traps at high rate. As sending traps at high rate snmpv2_trap() api hangs somewhere , i am doubting that it is somewhere when sending data to master agent on port 706 on tcp. I am running master agent using following command: /usr/local/sbin/snmpd -D -Lsd -Lf /root/GAURAV_OSS_AGENT/TEST_ENV/snmpd.log -a -c /usr/local/share/snmp/snmpd.conf -x tcp:localhost:706 In log file i also saw "AgentX master agent failed to respond to ping. Attempting to re-register." thrice So i am doubting that tcp queue is filling up while sending data at fast rate. Is there some issue with using Agentx configuration using master-subagent Can i configure my agent without agentx configuration. So could it be possible that if i run my agent without agentx confgiuration , it might solve the issue Following is the configuration in snmpd.conf for sending v3 traps: trapcommunity sbssEmsOssTrapCommPrivate trapsess -e 0x80001f8804534253535f454d535f4f53535f31 -v 3 -u sbssEmsOss -a MD5 -A sbssEmsOssSecretpass -l authNoPriv 192.168.150.5:9162 Can anyone help me on it, or provide me some direction where can i debug. Thanks and Regards -- Click for exclusive coverage on the New Bajaj Pulsar 220 the fastest Indian bike http://www.zigwheels.com/Features/Bajaj-Pulsar-220-DTSi-Special-Coverage/Pulsar_20090623-1-1 -- Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf ___ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
