On Wed, 22 Apr 2015 13:48:37 -0700 Dipesh wrote:
DG> With netsnmp stack 5.1 we are seeing that when getbulk is
DG> performed on 2 or more oids and if one one the OIDs is under
DG> "deny" list snmpd crashes.
5.1 is *really* old and no longer supported. Have you tried this
(+ [email protected])
On Wed, Apr 22, 2015 at 1:48 PM, Dipesh G wrote:
> Hello,
>
> With netsnmp stack 5.1 we are seeing that when getbulk is performed on
> 2 or more oids and if one one the OIDs is under "deny" list snmpd crashes.
>
> The stack
Hello,
With netsnmp stack 5.1 we are seeing that when getbulk is performed on
2 or more oids and if one one the OIDs is under "deny" list snmpd crashes.
The stack trace we see is pointing to _reorder_getbulk where a SEGV is
occuring
while accessing asp->bulkcache.
Is there a known
ly large amounts of data. I have noted that
>>>>> issuing a 'snmpbulkwalk' to Master Agent results in this broken down into
>>>>> GETNEXT requests to the Sub-Agent. This results in a lot of network
>>>>> traffic, as well as a tremendous number of co
broken down into
>>>> GETNEXT requests to the Sub-Agent. This results in a lot of network
>>>> traffic, as well as a tremendous number of context switches (everything
>>>> running on same host).
>>>>
>>>> Is there support in the Sub-Agent
emendous number of context switches (everything
>>> running on same host).
>>>
>>> Is there support in the Sub-Agent to accept GETBULK requests from the
>>> Master? I have looked at the code, their appears to be some infrastructure
>>> in place, but it is no
t; 'snmpbulkwalk' to Master Agent results in this broken down into GETNEXT
>> requests to the Sub-Agent. This results in a lot of network traffic, as
>> well as a tremendous number of context switches (everything running on same
>> host).
>>
>> Is there sup
GETNEXT
> requests to the Sub-Agent. This results in a lot of network traffic, as
> well as a tremendous number of context switches (everything running on same
> host).
>
> Is there support in the Sub-Agent to accept GETBULK requests from the
> Master? I have looked
raffic, as
well as a tremendous number of context switches (everything running on same
host).
Is there support in the Sub-Agent to accept GETBULK requests from the
Master? I have looked at the code, their appears to be some infrastructure
in place, but it is not clear to me if it is fully functional
ans some 5 secs. what is the reason.?
The coding structure using milliseconds. But the command line options
use seconds (and you can specify floats to get < 1 sec).
RKK> And with snmp v3 settings for getbulk req always time out is coming
RKK> even if agent is responding what could be the reas
what is the unit(sec's or millisec's) of timeout parameter when we are issuing
request. I set it as 500 but it is waiting for very less time means some 5
secs. what is the reason.?
And with snmp v3 settings for getbulk req always time out is coming even if
agent is responding what co
On 21 February 2011 14:10, Fulko Hew wrote:
> It doesn't appear that Net-SNMP uses AgentX getBulk PDUs.
I believe that's correct, yes.
Dave
--
The ultimate all-in-one performance toolkit: Intel(R) Paral
Just to confirm...
It doesn't appear that Net-SNMP uses AgentX getBulk PDUs.
Looking through the code, (and sniffing the wire) it looks like
the master agent maps an incoming SNMP getBulk onto a 'number of
AgentX getNext commands'... Implying that my AgentX sub-agent code
doesn&#x
Hi All,
I am using NET-SNMP version 5.4.2.1.
I have large data with multiple tables so I used mib2c.table_data.conf for
code generation.
When I tried snmpbulkwalk it is translated into consecutive GET requests.
1. Is table_data helper supports GETBULK or I need to change the
Hi Wes
(Adding Jochen, the debian maintainer of net-snmp)
On Tue, 4 Nov 2008 07:26:27 am Wes Hardaker wrote:
> > On Tue, 4 Nov 2008 02:02:54 +1100, Steffen Joeris
> > <[EMAIL PROTECTED]> said:
>
> SJ> This version of net-snmp was released with etch (our stable version)
> and SJ> unfortuna
> On Tue, 4 Nov 2008 02:02:54 +1100, Steffen Joeris <[EMAIL PROTECTED]>
> said:
SJ> Is there any chance you can provide a backported fix? It would be
SJ> much appreciated and I won't have time to look into it until
SJ> Thursday :(
Someone needs to make similar changes to the old code...
> On Tue, 4 Nov 2008 02:02:54 +1100, Steffen Joeris <[EMAIL PROTECTED]>
> said:
SJ> This version of net-snmp was released with etch (our stable version) and
SJ> unfortunately it is not possible to increase the upstream version there.
BTW, our packaging policy is that changes to the Z in
Hi Wes
Thanks for clarifying.
On Tue, 4 Nov 2008 01:49:06 am Wes Hardaker wrote:
> > On Sat, 1 Nov 2008 20:47:38 +1100, Steffen Joeris
> > <[EMAIL PROTECTED]> said:
>
> SJ> In Debian stable, we are shipping version 5.2.3 and I was wondering,
> SJ> if it is vulnerable as well.
>
> Yes, it i
> On Sat, 1 Nov 2008 20:47:38 +1100, Steffen Joeris <[EMAIL PROTECTED]>
> said:
SJ> In Debian stable, we are shipping version 5.2.3 and I was wondering,
SJ> if it is vulnerable as well.
Yes, it is vulnerable. Note that we released a 5.2.5.1 as well to
update that line of code. Any chan
Hi
(Please CC me on replies, since I am not subscribed)
I just saw the announcement[0] about the potential DoS attack and had a look
through the patches. In Debian stable, we are shipping version 5.2.3 and I
was wondering, if it is vulnerable as well. It seems that the vulnerable code
is not pr
@lists.sourceforge.net
Subject: Re: Getbulk maximum message size
>>>>> "RS" == Robert Story <[EMAIL PROTECTED]> writes:
DBGA> However, it seems that when the msgMaxSize of the originator is
DBGA> exceeded, Net-SNMP agent doesn't respond at all
RS> Ee
> "RS" == Robert Story <[EMAIL PROTECTED]> writes:
DBGA> However, it seems that when the msgMaxSize of the originator is
DBGA> exceeded, Net-SNMP agent doesn't respond at all
RS> Eeek. That's not good. Please file a bug report.
Unfortunately, I believe the fix is not easy at all. It'll requi
Blanco Gabriel-AGD019 <[EMAIL PROTECTED]>
Cc: [email protected]
Sent: Wednesday, July 4, 2007 11:18:47 PM
Subject: Re: Getbulk maximum message size
On Wed, 4 Jul 2007 11:24:06 -0400 Del wrote:
DBGA> I am using Net-SNMP 5.3.0.1, and I am having problems with the size
On Wed, 4 Jul 2007 11:24:06 -0400 Del wrote:
DBGA> I am using Net-SNMP 5.3.0.1, and I am having problems with the size of
the packets returned by getbulk.
DBGA> [...]
DBGA> However, it seems that when the msgMaxSize of the originator is exceeded,
Net-SNMP agent doesn't respon
I am using Net-SNMP 5.3.0.1, and I am having problems with the size of the
packets returned by getbulk.
According to RFC 3416 (section 4.2.3):
" (1) If the size of the message encapsulating the Response-PDU
containing the requested number of variable bindings wou
On Tue, 10 Apr 2007 05:46:03 -0400 Senthil wrote:
SN> I have posted lot of queries and got information regarding Get-Bulk request.
SN> for your information,I have to provide the info about my application
SN>
SN> I am using net-snmp 5.3.1 version.
SN> I have used mib2c in iterate config for code g
Dear Groups,
Dear all,
My Greetings to you.
I have posted lot of queries and got information regarding Get-Bulk request.
for your information,I have to provide the info about my application
I am using net-snmp 5.3.1 version.
I have used mib2c in iterate config for code generation.
The Snmp Age
rticular helper does not include direct support for
GetBulk requests, and relies on them being converted
into a sequence of GetNext requests.
If you were coding your MIB module directly (not using any
of the provided table helpers), then you could opt to handle
GetBulk requests yourself.
But
Dear all,
My Greetings to you.
I have posted lot of queries and got information regarding Get-Bulk request.
for your information,I have to provide the info about my application
I am using net-snmp 5.3.1 version.
I have used mib2c in iterate config for code generation.
During Get bulk query fire
net api.
> internally the snmp agent translates th eget bul request as consecutive get
> request only.
> I need to know the conversion from getBulk to get request.
> Is it possible to identify the request of Get bulk from netsnmp api level.
If you register a MIB handler using
>>>>> "SN" == Senthil Nathan <[EMAIL PROTECTED]> writes:
SN> Any one of you, help me to resolve this issue, so that during
SN> getbulk, it wont start call another table api.
that's how getbulk works within SNMP. From a legal SNMP standpoint,
it has to
ower layers.
In the get_first and get_next routines, i am using position coount as my
loop context.
Here with i have given the sample code for you reference.
The issue, i have faced, During getBulk query ,i have retrieved all the
rows from the table e.g 40 rows.
After that, it will call netsnmp
routines, i am using position coount as my loop
context.
Here with i have given the sample code for you reference.
The issue, i have faced, During getBulk query ,i have retrieved all the rows
from the table e.g 40 rows.
After that, it will call netsnmp next table handler api.
Any one of you, help
Hi,
I seek your help regarding my projectt.
Here, i am using netsnmp5.3.1 version.
I have generated c codes by using mib2c which create code skelton based on
mib2c iterator conf option .
My snmp agent get the values from other sub system not its own.
My issue is,
when i make get bulk query b
ubject: RE: Decrease execute times of get_first_data_point() when getBulk
I am stumped - I ran mib2c -c mib2c.iterate.conf -S cache=1
MIB_FILE::table_name
And got table_name.c/h with the expected container free/load routines.
The problem is the prototype and function declarati
: Decrease execute times of get_first_data_point() when getBulk
Thanks
I did this last night and it worked fine - simple cut and paste.
Along a similar line, is there a mechanism to hook into the processing such
that I get a call to a handler before ALL get/set processing but after the
OIDs are pa
going to be sent?
Adam
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Dave Shield
Sent: Friday, March 16, 2007 5:51 AM
To: Adam Bell
Cc: [email protected]
Subject: Re: Decrease execute times of get_first_data_point() when getBulk
On
On 15/03/07, Adam Bell <[EMAIL PROTECTED]> wrote:
> Is there a way to turn on the caching with mib2c.iterate_access.conf?
Probably not. The caching support was introduced completely
separately from the iterate_access config.
The best way forward is probably to merge the caching elements
into yo
@lists.sourceforge.net
Subject: Re: Decrease execute times of get_first_data_point() when getBulk
On 09/03/07, Zhang Chuan <[EMAIL PROTECTED]> wrote:
> Since I want to get the newest data from system when a GETBULK message
> come in. I have an init_table() function in get_first_data_poi
Thanks for useful advices. I have tried it but can not success in compile.
1.I use "mib2c -S cache=1 -c mib2c.iterate.conf portInfoTable" to generate
the framework.
2.After compared the new framework with the old one, I got there are just
minor difference between them.
3.I added the difference c
On 09/03/07, Zhang Chuan <[EMAIL PROTECTED]> wrote:
> Since I want to get the newest data from system when a GETBULK message come
> in. I have an init_table() function in get_first_data_point(). The
> init_table() function get the data from system by TCP and set them to a
> g
@lists.sourceforge.net
Subject: Decrease execute times of get_first_data_point() when getBulk
Hi,
Recently I have coded my own MIB module based on a stub file that was
originally auto-generated by mib2c using mib2c.iterate.conf. The version is
5.2.2. The module contains a table with 10 columns and
Hi,
Recently I have coded my own MIB module based on a stub file that was
originally auto-generated by mib2c using mib2c.iterate.conf. The version is
5.2.2. The module contains a table with 10 columns and currently 125 rows.
Since I want to get the newest data from system when a GETBULK message
On Wed, Jan 10, 2007 at 03:06:05PM +, Dave Shield wrote:
>
> On 09/01/07, Lilach Givaty <[EMAIL PROTECTED]> wrote:
> > You are right, it isn't in the snmpd project. Sorry.
> > But maybe you can help me, when I do getbulk with max-repetitions very
> > lar
]
On 09/01/07, Lilach Givaty <[EMAIL PROTECTED]> wrote:
> You are right, it isn't in the snmpd project. Sorry.
> But maybe you can help me, when I do getbulk with max-repetitions very
> large, I get a response from my device that is larger than a pdu (1500
> bytes). The pac
On 07/01/07, Lilach Givaty <[EMAIL PROTECTED]> wrote:
> I found one bug with get bulk - in table_array_ext_data.c
Errr where exactly is this file?
I've checked the source tree, and can't immediately spot it.
This code doesn't seem to apply to 'agent/helpers/table_data.c'
Dave
---
row.
Please post to http://www.net-snmp.org/patches so it won't get lost.
> Another bug we have is that when making a big request for getbulk, the
> snmp should fill the PDU and then stop. We get a full packet, in
> fregments with the entire request.
> Did anyone find a solotion for
row = find_next_row(tblreq_info, tad);
+row = find_next_row(tblreq_info, tad,request_group);
if (!row) {
/*
* no results found.
Another bug we have is that when making a big request for getbulk, the
snmp should fill the PDU and then st
On Tue, 5 Sep 2006 14:51:05 + Claus wrote:
CK> is there any patch available to fix the get-bulk problem?
Not that I know of. Did you ever file a bug-report for this problem? If not,
please do so, so we can get to it when someone gets properly motivated.
If you are looking at it yourself, net
On Wednesday 09 August 2006 17:40, Robert Story wrote:
> On Wed, 9 Aug 2006 15:11:52 + Claus wrote:
> CK> On Wednesday 09 August 2006 14:27, Robert Story wrote:
> CK> > On Wed, 9 Aug 2006 13:06:25 + Claus wrote:
> CK> > CK> IF-MIB::ifDescr[7] = STRING: br0
> CK> > CK> IF-MIB::ifDescr[501] =
]
On 04/09/06, Ritwik, Basu <[EMAIL PROTECTED]> wrote:
> The language I am using is C. In such a case how does the prototype
> differ from the ones for Get and Getnext requests?
It doesn't.
As I said:
>> The basic Net-SNMP library uses the same functions to send *all*
>&
On 04/09/06, Ritwik, Basu <[EMAIL PROTECTED]> wrote:
> Could you please tell me the function prototype for getbulk???
At what level are you working? (Which language? :-))
What function prototypes are you comparing GETBULK against?
The basic Net-SNMP library uses the same functions to
Hi,
Could you please tell me the function prototype for
getbulk???
How is the bulk data returned to the calling master agent? I
mean what is the nature of the pointer or structure used?
Thanks
Ritwik
On Wednesday 09 August 2006 17:40, Robert Story wrote:
> On Wed, 9 Aug 2006 15:11:52 + Claus wrote:
> CK> On Wednesday 09 August 2006 14:27, Robert Story wrote:
> CK> > On Wed, 9 Aug 2006 13:06:25 + Claus wrote:
> CK> > CK> IF-MIB::ifDescr[7] = STRING: br0
> CK> > CK> IF-MIB::ifDescr[501] =
On Wed, 9 Aug 2006 15:11:52 + Claus wrote:
CK> On Wednesday 09 August 2006 14:27, Robert Story wrote:
CK> > On Wed, 9 Aug 2006 13:06:25 + Claus wrote:
CK> > CK> IF-MIB::ifDescr[7] = STRING: br0
CK> > CK> IF-MIB::ifDescr[501] = STRING: nas0 <<
CK> > CK> FAULT positi
On Wednesday 09 August 2006 14:27, Robert Story wrote:
> On Wed, 9 Aug 2006 13:06:25 + Claus wrote:
> CK> [EMAIL PROTECTED]:~]# snmpbulkwalk localhost ifDescr
> CK> IF-MIB::ifDescr[1] = STRING: lo
> CK> IF-MIB::ifDescr[2] = STRING: eth0
> CK> IF-MIB::ifDescr[3] = STRING: eth1
> CK> IF-MIB::ifDe
On Wed, 9 Aug 2006 13:06:25 + Claus wrote:
CK> [EMAIL PROTECTED]:~]# snmpbulkwalk localhost ifDescr
CK> IF-MIB::ifDescr[1] = STRING: lo
CK> IF-MIB::ifDescr[2] = STRING: eth0
CK> IF-MIB::ifDescr[3] = STRING: eth1
CK> IF-MIB::ifDescr[4] = STRING: eth2
CK> IF-MIB::ifDescr[5] = STRING: dummy0
CK> I
Hallo,
this is our new iftable layout on AXR (embedded linux ppc system):
index 1-8, 501-503 from AgentxMaster: snmpd
other from AgentxClient: AXR nemo
Any suggestion?
claus
[EMAIL PROTECTED]:~]# snmpd -v
NET-SNMP version: 5.3.1
Web: http://www.net-snmp.
Ok - I got it to work. In 5.2.1, I had removed the init_snmp() call
because we handle our own config file. By adding this back in, as
init_snmp("") it works. I had noticed that a GET to the right community
returned error, but the wrong community didn't return anything at all,
so I investigated the
On Wed, 2006-03-01 at 15:08 +0200, Harold J. Ship wrote:
> I'm trying to compile, link and run in 5.3.0.1 the code generated and
> modified in 5.2.1 and I'm finding that nothing is happening:
> Before I get too caught up in this, is this legal? The code was
> generated in 5.2.1 using MFD; plus the
I'm trying to compile, link and run in 5.3.0.1 the code generated and
modified in 5.2.1 and I'm finding that nothing is happening: I receive
"No more variables left in this MIB View" for any GET or GETNEXT.
Before I get too caught up in this, is this legal? The code was
generated in 5.2.1 using MF
> On Sun, 5 Feb 2006 14:56:05 +0200, "Harold J. Ship" <[EMAIL PROTECTED]>
> said:
Harold> I've just tried a small example using 5.3.0.1 and GET-BULK with MFD
Harold> appears to be ok.
Harold> In order to port our agent over, it will take a few days of
Harold> effort. We may or may not de
I've just tried a small example using 5.3.0.1 and GET-BULK with MFD
appears to be ok.
In order to port our agent over, it will take a few days of effort. We
may or may not depending upon a number of factors.
Harold Ship
Giant Steps Networks
--
x27;t affect the iterator helper, or other
non-MfD modules.
> Is that a code generation problem or agent bug?
A code generation problem.
> I am using code base of 5.2.1 Is that a must to upgrade
> to 5.3 for GETBULK to work properly?
No. At least, I don't think so - suck it and see.
Hi all, will this problem affect other MIB2C template implementations like
mib2c.iterate.conf?
Is that a code generation problem or agent bug?
I am using code base of 5.2.1 Is that a must to upgrade to 5.3 for GETBULK
to work properly?
Thanks,
Shuki
-Original Message-
From: [EMAIL
On Sun, 22 Jan 2006 19:03:28 +0200 Harold wrote:
HJS> You are right, we are using 5.2.1. I'd like to avoid having to convert
HJS> all of our agent to 5.3 if I can.
Ok. What about 5.2.2? That's a smaller step (though I'm not 100% sure it will
fix the problem).
HJS> I tried the change you suggested
On Sun, 2006-01-22 at 00:16 +0530, [EMAIL PROTECTED] wrote:
> I have to send getbulk response for a table which has index
> MacAddress type. I am using net-snmp-5.2.1 in linux.
> I am putting the column name ( MIB-NAME::myColumn ) appended
> with index mac address into one
You are right, we are using 5.2.1. I'd like to avoid having to convert
all of our agent to 5.3 if I can.
I tried the change you suggested, changing ASN_PRIV_RETRY to
SNMP_NOSUCHINSTANCE in the _mfd_xTable_get_values() function both in the
current table and the following table, but that didn't help
On Sun, 22 Jan 2006 09:32:33 +0200 Harold wrote:
HJS> I have a problem with GETBULK on an agent developed using mib2c with
HJS> MFD.
You didn't say which release, but I'm guessing 5.2.x, right? If you can, jump
up to 5.3.0.1. There have been lots of improvements to the generated
Hello
I have a problem with GETBULK on an agent developed using mib2c with
MFD.
The agent was built by running mib2c with MFD on several tables, one at
a time. One of the subagent files was modified to run the "init"
function for each of the tables. Everything has been working fine
Title: Sending MacAddress type index for getBulk on table request
Hi,
I have to send getbulk response for a table which has index MacAddress type. I am using net-snmp-5.2.1 in linux.
I am putting the column name ( MIB-NAME::myColumn ) appended with index mac address into one
On Thu, 2005-12-15 at 22:20 +0100, Magnus Fromreide wrote:
> On Thu, Dec 15, 2005 at 10:38:05AM -0500, Robert Story wrote:
> > On Wed, 14 Dec 2005 04:29:54 -0600 Rajnish wrote:
Rajnish> What is the maximum size of PDU
Robert> 65k
Magnus> Is this limit mandated by standards
Yes - though it seems
On Thu, Dec 15, 2005 at 10:38:05AM -0500, Robert Story wrote:
> On Wed, 14 Dec 2005 04:29:54 -0600 Rajnish wrote:
> DR> 1. What is the maximum size of PDU in "snmpbulkwalk " response (In terms
> DR> of bytes)?
> DR> 2. What is the maximum size of any PDU message in "snmpget" response(in
> DR> terms
On Wed, 14 Dec 2005 04:29:54 -0600 Rajnish wrote:
DR> 1. What is the maximum size of PDU in "snmpbulkwalk " response (In terms
DR> of bytes)?
DR> 2. What is the maximum size of any PDU message in "snmpget" response(in
DR> terms of byte)?
65k, but you are unlikely to find implementations/networks t
Hi,
Help me for the
following queries,
1. What is the
maximum size of PDU in "snmpbulkwalk " response (In terms of
bytes)?
2. What is the
maximum size of any PDU message in "snmpget" response(in terms of
byte)?
Thanks,
Rajnish
Dey
known issue? Is there a workaround? I checked the bug database
and there did not seem to be any entries concerning table iterators and
GETBULK requests. I also didn't find anything in the coder's mailing list
archive.
This is on NET-Snmp 5.1.1. I also tried code on 5.2.1.2 and it a
> As the title says, how the bulk-get and bulk-walk
> works?
Well, both of these commands are actually just wrappers
round the SNMP GETBULK protocol operation, so there's
no inherent difference between them. It's just that
the "bulkwalk" will repeat the request and
Hello,
As the title says, how the bulk-get and bulk-walk
works?
More precisly, does the agent check all the access of
the requested oids , at one same time, or , is each
access on oid are checked one by one.
In fact I would like to know how the VACM is used in
case of bulk-request
Thanks,
Jeff
ombination of following
conditions
1. use agentX to get rows in table.
2. MG-snmp browser must enables getbulk feature.
3. polls more than 1 table at once.
If I do use getnext (instead of getbulk), no memory
leak.
If I do poll only 1 table at once, no memory
leak.
If those tables
> does GetNext and GetBulk functions column-wise in case of a table
> or row-wise..
They both walk the full length of each column (i.e. the value from
each row) before moving on to the next column.
Did you think of actually trying this with an existing agent,
to see what it does in pr
Hi!!,
Can anyone
tell me how GetNext and GetBulk functions. .I mean does GetNext and
GetBulk functions column-wise in case of a table or row-wise..
If possible if anyone can give a link which
explains the proper functionality of GETNEXT and GETBULK
functionality
81 matches
Mail list logo
