I use net-snmp 5.4 to send out enterprise-specific traps.
The trap-sending code is like agent/mibgroup/examples/notification.c
When sending trap very fast, we can see that snmpd consumes more and
more memory(up to 500M).
Is there any memory leak problem in net-snmp 5.4?
>From http://net-snmp.sour
I have modified snmptrapd codes, removed codes relate to sysytem service and
some other codes and options that I think I don't need, and changed main
function of snmptrapd.c into a regular function which is a child thread entry
function from a main process which in turn started by another Window
On Thu, 19 Jul 2007 20:31:52 +0100 Dave wrote:
DS> It will probably slip very slightly, but ought to be out within a week.
Or two...
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Stud
Dear net-snmp-coders,
There is another major bug I discovered when working on
nlmLogVariableTable.
netsnmp_table_data_add_row() fails to add rows if a lower oid row does
not exist.
For example, nlmLogVariableTable has the following objects:
nlmLogVariableID
nlmLogVariableValueType
nlmLogVari
On 19/07/07, Sukwoo Kang <[EMAIL PROTECTED]> wrote:
> We plan to release our subagent software which is based on Net-SNMP shared
> libraries on AIX. According to your release schedule, the final of 5.4.1
> will be released tomorrow. Is it going to be released as planned?
It will probably slip very
Hello,
We plan to release our subagent software which is based on Net-SNMP shared
libraries on AIX. According to your release schedule, the final of
5.4.1will be released tomorrow. Is it going to be released as planned?
Then we
can guide customers to use 5.4.1 on AIX without no problem. Please le
Jayson Robinson wrote:
> We found the issue. I thought you might like to know. After continuous
> polling of the agent we saw it die in the TCP mib. I was able to
> correlate that to a weblogics server on the box that isn't cleaning up
> its connections and leaving idle connections lying around.
>I offer the following (two-patch) patch for consideration.
>
>The first part adds checks for 'chown()' and 'localtime_r()'functions
to the configure mechanism.
>The second part uses the results of these checks to handle the two code
fragments that triggered the problem.
Does that require a fresh
Dave,We found the issue. I thought you might like to know. After continuous
polling of the agent we saw it die in the TCP mib. I was able to correlate
that to a weblogics server on the box that isn't cleaning up its connections
and leaving idle connections lying around.When the net-snmp
On 19/07/07, Thomas Anders <[EMAIL PROTECTED]> wrote:
> >I'm not sure what the purpose of win32/include/net-snmp-config.h.in
> > is anyway. Does anything actaully use this?
>
> win32/Configure
Noted.
> And given that changes to include/net-snmp/net-snmp-config.h too often
> haven't been mi
Dave Shield wrote:
> On 19/07/07, Thomas Anders <[EMAIL PROTECTED]> wrote:
>> One of us must be confused by 'win32' vs. 'include', I guess.
>
> Doh!
> My apologies. Hadn't spotted you were talking about that one.
>
>
> Strictly speaking, yes - the win32 header files should probably
> include t
On 19/07/07, Thomas Anders <[EMAIL PROTECTED]> wrote:
> One of us must be confused by 'win32' vs. 'include', I guess.
Doh!
My apologies. Hadn't spotted you were talking about that one.
Strictly speaking, yes - the win32 header files should probably
include the same definitions. But since the
Dave Shield wrote:
> On 19/07/07, Thomas Anders <[EMAIL PROTECTED]> wrote:
>> Don't we also need a third part (or an addition to the first) that
>> changes win32/net-snmp/net-snmp-config.h{,in}?
>
> Have you looked at the file "mingw-config.patch" ?
It may come as a surprise, but I have. Even my
That clears it up . thanks a lot.
Robert Story <[EMAIL PROTECTED]> wrote: On Wed, 18 Jul 2007 13:51:30 -0700
(PDT) Need wrote:
NH> I just do not understand how these SNMP requests are related to port 705?
I mean how does the master agent get the SNMP request if the master agent is
only l
On 19/07/07, Thomas Anders <[EMAIL PROTECTED]> wrote:
> Don't we also need a third part (or an addition to the first) that
> changes win32/net-snmp/net-snmp-config.h{,in}?
Have you looked at the file "mingw-config.patch" ?
Dave
Dave Shield wrote:
> I offer the following (two-patch) patch for consideration.
> The first part adds checks for 'chown()' and 'localtime_r()'
> functions to the configure mechanism.
> The second part uses the results of these checks to
> handle the two code fragments that triggered the problem
On 17/07/07, Wes Hardaker <[EMAIL PROTECTED]> wrote:
Well, we're in various states of partial agreement. So I punted and
released rc3 as is.
Wimp!
I don't mind doing an immediate rc4 if we get a
patch everyone agrees on.
I offer the following (two-patch) patch for
17 matches
Mail list logo
