External ids buffer

[supalik]

ChangeLog: ovs_stas plugin:These changes will enhance the feature of receiving the data in buffer.

By including "ExternalIds true" as a New Config parameter all the external ids will be populated in the buffer
O/P: ovs_stats="vmId=3979a0b0-bddc-4cdd-ae87-0a4c2071d1f9,ifaceId=7e0607f6-9179-4852-aa36-b56279c3310,mac=fa:16:3e:18:a1:ff",instance=""}

which can provide the visibility of external ids if exists.

[supalik]

Can I get some comments on these changes

[mrunge]

So, for me to get this straight: I asked you to separate the change for #3924 and this enhancement. You agreed to that.
This enhancement now also contains your change for #3924.

[mrunge]

as agreed in the other PR, let's separate bug fixes and enhancements.

[supalik]

Hi,
I have removed the #3924 changes, but now test_plugin_capabilities is getting FAILED .
Could you please assist here.

[supalik]

Hi,
I have executed CI Test cases in collectd/ci:trusty_amd64 on main branch, without my changes, but "test_plugin_capabilities" is failing there as well.

[gyanranjan]

@Supali-Kash the issue seems to be due to capability plugin

Success.
==28070== Invalid free() / delete / delete[] / realloc()
==28070== at 0x4C2BDEC: free (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28070== by 0x55CEE2B: __libc_freeres (in /lib/x86_64-linux-gnu/libc-2.19.so)
==28070== by 0x4A256BC: _vgnU_freeres (in /usr/lib/valgrind/vgpreload_core-amd64-linux.so)
==28070== by 0x54A216A: __run_exit_handlers (exit.c:97)
==28070== by 0x54A21F4: exit (exit.c:104)
==28070== by 0x4021DF: main (capabilities_test.c:282)
==28070== Address 0x58283c0 is 0 bytes inside data symbol "noai6ai_cached"
==28070==
{
<insert_a_suppression_name_here>
Memcheck:Free
fun:free
fun:__libc_freeres
fun:_vgnU_freeres
fun:__run_exit_handlers
fun:exit
fun:main
}

[supalik]

Hi, @paul/@mrunge /@bernd ,

I would be really grateful if you could assist me handling this PR.
This PR CI is failing but I could not get the exact reason of the ci:trusty_amd64 failure

[supalik]

@mrunge suggested change is done and present in last commit

[elfiesmelfie]

Hi, @paul/@mrunge /@bernd ,

I would be really grateful if you could assist me handling this PR. This PR CI is failing but I could not get the exact reason of the ci:trusty_amd64 failure

The failure here looks like it is due to test_capabilities failing when make check is run.
I'm investigating these failures, and it might be because the trusty container has not been updated in a while.
The CI is undergoing some updating at the moment, please bear with us while that happens.

This failure is probably nothing to worry about, I'll try to verify it manually today, to see if I can reproduce the error.

[gyanranjan]

Hi Emma
You are right!
Valgrind error seems to come probably due to old glibc

cap_config -> cf_util_get_port_number -> service_name_to_port_number (probably getaddrinfo)

root@e185e03cf263:/collectd# ldd --version
ldd (Ubuntu EGLIBC 2.19-0ubuntu6.15) 2.19
Copyright (C) 2014 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Written by Roland McGrath and Ulrich Drepper.

https://bugs.gentoo.org/472544

Unfortunately, Ubuntu 14.04 is stuck with 2.19

Confirmed it is glib Issue

#include <sys/types.h>
#include <sys/socket.h>
#include <netdb.h>
#include <stdlib.h>
#include <string.h>
#include <stdio.h>

// Trimmed from libaudit.
static void _resolve_addr(char buf[], const char *host)
{
	struct addrinfo *ai;
	struct addrinfo hints;
	int e;

	memset(&hints, 0, sizeof(hints));
	hints.ai_flags = AI_ADDRCONFIG;
	hints.ai_socktype = SOCK_STREAM;

	e = getaddrinfo(host, NULL, &hints, &ai);
	if (e != 0) {
		printf("error\n");
		return;
	}
	freeaddrinfo(ai);
	printf("resolve one\n", buf);
}

int main() {
	char buf[INET6_ADDRSTRLEN];
	int i;
	for (i = 0; i < 3; i++) {
		_resolve_addr(buf, "localhost");
	}
}

root@e185e03cf263:/collectd#
root@e185e03cf263:/collectd# valgrind ./a.out
==30519== Memcheck, a memory error detector
==30519== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==30519== Using Valgrind-3.10.1 and LibVEX; rerun with -h for copyright info
==30519== Command: ./a.out
==30519==
resolve one
resolve one
resolve one
==30519== Invalid free() / delete / delete[] / realloc()
==30519== at 0x4C2BDEC: free (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==30519== by 0x4F9FE2B: __libc_freeres (in /lib/x86_64-linux-gnu/libc-2.19.so)
==30519== by 0x4A256BC: _vgnU_freeres (in /usr/lib/valgrind/vgpreload_core-amd64-linux.so)
==30519== by 0x4E7316A: __run_exit_handlers (exit.c:97)
==30519== by 0x4E731F4: exit (exit.c:104)
==30519== by 0x4E58F4B: (below main) (libc-start.c:321)
==30519== Address 0x51f93c0 is 0 bytes inside data symbol "noai6ai_cached"
==30519==
==30519==
==30519== HEAP SUMMARY:
==30519== in use at exit: 0 bytes in 0 blocks
==30519== total heap usage: 53 allocs, 54 frees, 8,217 bytes allocated
==30519==
==30519== All heap blocks were freed -- no leaks are possible
==30519==
==30519== For counts of detected and suppressed errors, rerun with: -v
==30519== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)

[elfiesmelfie]

Hi Emma You are right! Valgrind error seems to come probably due to old glibc

cap_config -> cf_util_get_port_number -> service_name_to_port_number (probably getaddrinfo)

root@e185e03cf263:/collectd# ldd --version ldd (Ubuntu EGLIBC 2.19-0ubuntu6.15) 2.19 Copyright (C) 2014 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Written by Roland McGrath and Ulrich Drepper.

https://bugs.gentoo.org/472544

Unfortunately, Ubuntu 14.04 is stuck with 2.19

Confirmed it is glib Issue

#include <sys/types.h>
#include <sys/socket.h>
#include <netdb.h>
#include <stdlib.h>
#include <string.h>
#include <stdio.h>

// Trimmed from libaudit.
static void _resolve_addr(char buf[], const char *host)
{
	struct addrinfo *ai;
	struct addrinfo hints;
	int e;

	memset(&hints, 0, sizeof(hints));
	hints.ai_flags = AI_ADDRCONFIG;
	hints.ai_socktype = SOCK_STREAM;

	e = getaddrinfo(host, NULL, &hints, &ai);
	if (e != 0) {
		printf("error\n");
		return;
	}
	freeaddrinfo(ai);
	printf("resolve one\n", buf);
}

int main() {
	char buf[INET6_ADDRSTRLEN];
	int i;
	for (i = 0; i < 3; i++) {
		_resolve_addr(buf, "localhost");
	}
}

root@e185e03cf263:/collectd# root@e185e03cf263:/collectd# valgrind ./a.out ==30519== Memcheck, a memory error detector ==30519== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al. ==30519== Using Valgrind-3.10.1 and LibVEX; rerun with -h for copyright info ==30519== Command: ./a.out ==30519== resolve one resolve one resolve one ==30519== Invalid free() / delete / delete[] / realloc() ==30519== at 0x4C2BDEC: free (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30519== by 0x4F9FE2B: __libc_freeres (in /lib/x86_64-linux-gnu/libc-2.19.so) ==30519== by 0x4A256BC: _vgnU_freeres (in /usr/lib/valgrind/vgpreload_core-amd64-linux.so) ==30519== by 0x4E7316A: __run_exit_handlers (exit.c:97) ==30519== by 0x4E731F4: exit (exit.c:104) ==30519== by 0x4E58F4B: (below main) (libc-start.c:321) ==30519== Address 0x51f93c0 is 0 bytes inside data symbol "noai6ai_cached" ==30519== ==30519== ==30519== HEAP SUMMARY: ==30519== in use at exit: 0 bytes in 0 blocks ==30519== total heap usage: 53 allocs, 54 frees, 8,217 bytes allocated ==30519== ==30519== All heap blocks were freed -- no leaks are possible ==30519== ==30519== For counts of detected and suppressed errors, rerun with: -v ==30519== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)

Since trusty is out of support, it's reasonable to not use it in the CI, so I'll remove it.
#3927 should address the failing CI

Thanks for triaging the issue!

[supalik]

thank you @elfiesmelfie for the response.
Could you please help me in getting review comments for the PR.

[mrunge]

Thank you for your PR! You could help us by reviewing other PRs. That would lower the amount of work and would help you in getting a review for this PR.