Skip to Content.
Sympa Menu

perfsonar-user - Re: [perfsonar-user] bwctl OSX giving strange results

Subject: perfSONAR User Q&A and Other Discussion

List archive

Re: [perfsonar-user] bwctl OSX giving strange results


Chronological Thread 
  • From: Aaron Brown <>
  • To: Brian Candler <>
  • Cc: Brian Tierney <>, "" <>
  • Subject: Re: [perfsonar-user] bwctl OSX giving strange results
  • Date: Fri, 19 Dec 2014 12:47:43 +0000
  • Accept-language: en-US
  • Authentication-results: pobox.com; dkim=none (message not signed) header.d=none;

Hey Brian,

On Dec 19, 2014, at 4:51 AM, Brian Candler <> wrote:

On 18/12/2014 21:26, Aaron Brown wrote:

It converts iperf3 from a library into an exec call.
I may be able to try this another day (v. busy right now), but I'll point out something else.

Yesterday I put the perfsonar node and the Mac Mini into the production network (with the Mac Mini on the end of a GPON link), and I've enabled only throughput test and ping test (no latency/loss).

The graphs I see are:

* IPv4: reverse throughput and reverse ping only
* IPv6: reverse throughput only

I presume the ping test is using bwping -T ping? However if I do the tests from the command line, they all work fine (see below: all combinations of -c and -s, IPv6 and IPv4)

I'm therefore leaning to the view that it's the core of bwctl itself which is the problem, rather than the external testing tools it invokes. As I say, I don't have time right now to continue investigation and fully understand the code paths. One useful approach might be to run bwctld under valgrind.

If bwping is running fine on its own, then it may be an issue higher up in the stack. Either the daemon that reads the bwping results, and stuffs them into the DB, or the DB itself. Were these tests configured by hand, or via the mesh configuration?

That should be unrelated to the “iperf3 empty results” issue that you’ve seen. The thing that’s weird with “iperf3 empty results” thing is that, based on the time, the server is brought up a few seconds before the client tries to connect, and then it bombs out. That’s one reason why I’ve wanted to swap to running iperf3 as a separate binary, so we can make sure it’s a bwctl issue and not some weird interaction between bwctl and iperf3 libraries, or bwctl using the iperf3 library wrong. At least with exec’ing the binary, we can guarantee that it matches reality. I’ve run bwctl in valgrind before, but it’s not warned of anything weird, nor have I been able to reproduce that problem locally.

Cheers,
Aaron


Regards,

Brian.

[example@cov-perf-1 ~]$ bwping -c cov-client-1 -T ping
bwping: Using tool: ping
bwping: 21 seconds until test results available

SENDER START
PING 2001:db8:0:64::252(2001:db8:0:64::252) from 2001:db8:0:2::254 : 56 data bytes
64 bytes from 2001:db8:0:64::252: icmp_seq=1 ttl=63 time=0.532 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=2 ttl=63 time=0.451 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=3 ttl=63 time=0.486 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=4 ttl=63 time=0.470 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=5 ttl=63 time=0.460 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=6 ttl=63 time=0.454 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=7 ttl=63 time=0.447 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=8 ttl=63 time=0.490 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=9 ttl=63 time=0.450 ms
64 bytes from 2001:db8:0:64::252: icmp_seq=10 ttl=63 time=0.450 ms

--- 2001:db8:0:64::252 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9000ms
rtt min/avg/max/mdev = 0.447/0.469/0.532/0.025 ms

SENDER END
[example@cov-perf-1 ~]$ bwping -c 192.0.2.252 -T ping
bwping: Using tool: ping
bwping: 21 seconds until test results available

SENDER START
PING 192.0.2.252 (192.0.2.252) from 192.0.2.254 : 56(84) bytes of data.
64 bytes from 192.0.2.252: icmp_seq=1 ttl=63 time=0.451 ms
64 bytes from 192.0.2.252: icmp_seq=2 ttl=63 time=0.497 ms
64 bytes from 192.0.2.252: icmp_seq=3 ttl=63 time=0.436 ms
64 bytes from 192.0.2.252: icmp_seq=4 ttl=63 time=0.434 ms
64 bytes from 192.0.2.252: icmp_seq=5 ttl=63 time=0.411 ms
64 bytes from 192.0.2.252: icmp_seq=6 ttl=63 time=0.419 ms
64 bytes from 192.0.2.252: icmp_seq=7 ttl=63 time=0.443 ms
64 bytes from 192.0.2.252: icmp_seq=8 ttl=63 time=0.442 ms
64 bytes from 192.0.2.252: icmp_seq=9 ttl=63 time=0.441 ms
64 bytes from 192.0.2.252: icmp_seq=10 ttl=63 time=0.428 ms

--- 192.0.2.252 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9000ms
rtt min/avg/max/mdev = 0.411/0.440/0.497/0.025 ms

SENDER END
[example@cov-perf-1 ~]$ bwping -s cov-client-1 -T ping
bwping: Using tool: ping
bwping: 21 seconds until test results available

SENDER START
PING6(56=40+8+8 bytes) 2001:db8:0:64::252 --> 2001:db8:0:2::254
16 bytes from 2001:db8:0:2::254, icmp_seq=0 hlim=63 time=0.327 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=1 hlim=63 time=0.306 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=2 hlim=63 time=0.364 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=3 hlim=63 time=0.347 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=4 hlim=63 time=0.319 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=5 hlim=63 time=0.312 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=6 hlim=63 time=0.405 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=7 hlim=63 time=0.330 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=8 hlim=63 time=0.321 ms
16 bytes from 2001:db8:0:2::254, icmp_seq=9 hlim=63 time=0.381 ms

--- 2001:db8:0:2::254 ping6 statistics ---
10 packets transmitted, 10 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 0.306/0.341/0.405/0.031 ms

SENDER END
[example@cov-perf-1 ~]$ bwping -s 192.0.2.252 -T ping
bwping: Using tool: ping
bwping: 21 seconds until test results available

SENDER START
PING 192.0.2.254 (192.0.2.254) from 192.0.2.252: 56 data bytes
64 bytes from 192.0.2.254: icmp_seq=0 ttl=63 time=0.286 ms
64 bytes from 192.0.2.254: icmp_seq=1 ttl=63 time=0.413 ms
64 bytes from 192.0.2.254: icmp_seq=2 ttl=63 time=0.439 ms
64 bytes from 192.0.2.254: icmp_seq=3 ttl=63 time=0.297 ms
64 bytes from 192.0.2.254: icmp_seq=4 ttl=63 time=0.298 ms
64 bytes from 192.0.2.254: icmp_seq=5 ttl=63 time=0.324 ms
64 bytes from 192.0.2.254: icmp_seq=6 ttl=63 time=0.372 ms
64 bytes from 192.0.2.254: icmp_seq=7 ttl=63 time=0.362 ms
64 bytes from 192.0.2.254: icmp_seq=8 ttl=63 time=0.353 ms
64 bytes from 192.0.2.254: icmp_seq=9 ttl=63 time=0.298 ms

--- 192.0.2.254 ping statistics ---
10 packets transmitted, 10 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 0.286/0.344/0.439/0.050 ms

SENDER END
[example@cov-perf-1 ~]$





Archive powered by MHonArc 2.6.16.

Top of Page