IP Office 500 ICMP latency & call quality

[M1strals]

Hi,

We have an IPoffice 500 system that intermittently the call quality is very choppy, and almosy inaudible, which is becoming a serious issue. We have a ping running to the IPoffice which shows latency, when the call quality becomes very bad ping times are in the region of 500-1000ms and we do get timeouts. It seems that the ping times steadily rise and then bottom out which is when we get very bad call issues and then will steadily drop back to normal (normal being 3-30ms :-( ) I am also running a ping to another device within the same Vlan and as expected ping times are <=1ms.

I have spanned the Ip office port to a wireshark but cannot see any obvious issues. We do have 2 SIP providers coming into this ip office.


Is it anyway possible to view the IPoffice CPU load, memory load or anything that might point to the IP office being 'busy' and unable to reply to the pings , which is when we get dire voice quality. Any help would be gratefully received.

Cheers,

M1strals






Reply from 10.0.1.9: bytes=32 time=3ms TTL=98
Reply from 10.0.1.9: bytes=32 time=4ms TTL=98
Reply from 10.0.1.9: bytes=32 time=5ms TTL=98<
Reply from 10.0.1.9: bytes=32 time=11ms TTL=98
Reply from 10.0.1.9: bytes=32 time=6ms TTL=98
Reply from 10.0.1.9: bytes=32 time=6ms TTL=98
Reply from 10.0.1.9: bytes=32 time=3ms TTL=98
Reply from 10.0.1.9: bytes=32 time=5ms TTL=98
Reply from 10.0.1.9: bytes=32 time=7ms TTL=98
Reply from 10.0.1.9: bytes=32 time=7ms TTL=98
Reply from 10.0.1.9: bytes=32 time=4ms TTL=98
Reply from 10.0.1.9: bytes=32 time=203ms TTL=9
Reply from 10.0.1.9: bytes=32 time=285ms TTL=9
Reply from 10.0.1.9: bytes=32 time=310ms TTL=9
Reply from 10.0.1.9: bytes=32 time=226ms TTL=9
Reply from 10.0.1.9: bytes=32 time=2ms TTL=98
Reply from 10.0.1.9: bytes=32 time=7ms TTL=98
Reply from 10.0.1.9: bytes=32 time=26ms TTL=98
Reply from 10.0.1.9: bytes=32 time=11ms TTL=98

 

[tlpeter]

Monitor can show CPU and memory stats.
Open monitor and go to filters -> system and select development tracing.
Then go to status and select performance or memory status.
This will help you a lot i think.


BAZINGA!

I'm not insane, my mother had me tested!

 

[amriddle01]

99% of such issues are network related, it is the network/network equipment carrying this traffic after all

 

[M1strals]

Hi Many thanks. We do have 2 other IP offices within other Vlans which behave normally, no voice call problems. We have checked all aspects of the LAN and there are no issues there and also other devices do not show this kind of latency. The LAN is setup correctly. We are not seeing any errors on the switchport, relating to NIC issues/duplex etc.

Cheers,

 

[amriddle01]

Swap the LAN port to LAN2 (or vice versa) and reprogramme, have seen dodgy LAN ports before but it's usually more pronounced issues than this

 

[M1strals]

Hi,

I have now got'performance' monitor on. CPU currently is 3-10% however, icmp times are low @ 1-3ms. Is there anything else to look at that might see where potential issues are???

Cheers,

M1strals

 

[tlpeter]

When you ping the IPO do you also ping the default gateway of that IPO?
Does it show the same ping times?
What is the port settings of the switch where the IPO is connected to?


BAZINGA!

I'm not insane, my mother had me tested!

 

[M1strals]

Hi, yes, ping times to the Gateway are =<1ms, see below.


Reply from 10.0.1.254: bytes=32 time<1ms TTL=64
Reply from 10.0.1.254: bytes=32 time<1ms TTL=64
Reply from 10.0.1.254: bytes=32 time<1ms TTL=64
Reply from 10.0.1.254: bytes=32 time<1ms TTL=64
Reply from 10.0.1.254: bytes=32 time<1ms TTL=64
Reply from 10.0.1.254: bytes=32 time<1ms TTL=64


The switchport settings/stats are as follows:


GigabitEthernet3/46 is up, line protocol is up (connected)
Hardware is C6k 1000Mb 802.3, address is 000f.2320.9de7 (bia 000f.2320.9de7)
Description: IP Office 10.0.1.9/24
MTU 1500 bytes, BW 100000 Kbit, DLY 10 usec,
reliability 255/255, txload 3/255, rxload 3/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 100Mb/s
input flow-control is off, output flow-control is off
Clock mode is auto
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output 00:00:21, output hang never
Last clearing of "show interface" counters 11w3d
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 1418000 bits/sec, 823 packets/sec
5 minute output rate 1393000 bits/sec, 813 packets/sec
1753935064 packets input, 371170987955 bytes, 0 no buffer
Received 1019509 broadcasts (0 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 0 multicast, 0 pause input
0 input packets with dribble condition detected
1762790192 packets output, 368055841516 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 PAUSE output
0 output buffer failures, 0 output buffers swapped out

 

[tlpeter]

Is the port fixed to full duplex?
Try auto as the IPO has it set to auto and cannot be changed.


BAZINGA!

I'm not insane, my mother had me tested!

 

[M1strals]

Hi, no its set to auto/auto and has negotiated a-100 a-full

 

[tlpeter]

Clear, then i suggest (like Amriddle also did) is to try the second lan port (called Wan on the IPO)
I have seen broken ports too myself and it will give odd issues.


BAZINGA!

I'm not insane, my mother had me tested!

 

[M1strals]

Hi Many thanks, do you think that as we are running a small communinity network with 2 other IP offices conected. 2 Trunks come into the main IP office, all calls therefore route through the main IP office ,would this be a potentail bottleneck resulting in the latency we are seeing.?

M1strals

 

[amriddle01]

If the latency coincides with high call volume, then yes potentially