Re: 5.7.2.pre3 published for testing

2012-08-06 Thread Bart Van Assche
On 08/05/12 21:32, Magnus Fromreide wrote: On Sun, 2012-08-05 at 10:41 +, Bart Van Assche wrote: [ ... ] The above patch looks ok, so +1 for it. Thanks ! [PATCH] Make get_exec_output() work on big endian systems. I have a hard time understanding why cachebytes and out_size have to be

Re: 5.7.2.pre3 published for testing

2012-08-06 Thread Dave Shield
On 5 August 2012 11:41, Bart Van Assche bvanass...@acm.org wrote: --- a/agent/mibgroup/utilities/execute.h +++ b/agent/mibgroup/utilities/execute.h @@ -4,8 +4,8 @@ config_belongs_in(agent_module) int run_shell_command(char *command, char *input, - char *output,

Re: 5.7.2.pre3 published for testing

2012-08-06 Thread Bart Van Assche
On 08/06/12 07:18, Dave Shield wrote: On 5 August 2012 11:41, Bart Van Assche bvanass...@acm.org wrote: --- a/agent/mibgroup/utilities/execute.h +++ b/agent/mibgroup/utilities/execute.h @@ -4,8 +4,8 @@ config_belongs_in(agent_module) int run_shell_command(char *command, char *input, -

Re: 5.7.2.pre3 published for testing

2012-08-06 Thread Dave Shield
On 6 August 2012 08:34, Bart Van Assche bvanass...@acm.org wrote: Doesn't this introduce a change in the API ? That change restores the API to what it has always been before July 30, 2012 (see also commit 7374b84fe4c2ef8497fde3dae80a69aa89eba19c). So I'm not breaking the API but restoring it

win32: errno instead of WSAGetLastError to handle WinSock errors?

2012-08-06 Thread Schmoll Walter
I'm working on an application using the net-snmp library (net-snmp-5.6-2). And being curious, I took a look at the source code :-). Doing so I've mentioned the use of errno to get information about failed calls to WinSock. For example in snmpUDPBaseDomain.c: ... rc = sendto(t-sock,

5.7.2.pre3 and BSD

2012-08-06 Thread Niels Baggesen
The update to agent/kernel.[ch] introduced just before pushing pre3 have differing prototypes for klookup, and cannot compile. I suggest the following change --- a/agent/kernel.c +++ b/agent/kernel.c @@ -201,7 +201,7 @@ klread(char *buf, int buflen) * @return gives 1 on success and 0 on

procTable and regex

2012-08-06 Thread Niels Baggesen
The configure test for pcre requires that you explicitly configure --without-pcre if you are on a system without pcre support. Isn't this a bit overreacting? Actually, are there any reason to go for pcre for this, adding a dependency that is not standard on BSD or Solaris (and probably AIX and

Re: procTable and regex

2012-08-06 Thread Magnus Fromreide
On Mon, Aug 06, 2012 at 07:49:54PM +0200, Niels Baggesen wrote: The configure test for pcre requires that you explicitly configure --without-pcre if you are on a system without pcre support. Isn't this a bit overreacting? I have noticed that as well but not gotten around to fix it.