Ping times to switches very high but not to other servers or devices
Hello all,
So im having a weird issue here and perhaps its not even an issue.
ICMP / ping traffic to my switches on my LAN is very confusing. I am not a novice network technician but im no expert either and I cant quite explain it.
I ping any device on my network and the highest ping time I get is 2ms (for when it has to run across to the other end of my warehouse generally) however pinging any of my switches varies in ping times for example:
ITMGR2:~ username$ ping 192.168.1.13
PING 192.168.1.13 (192.168.1.13): 56 data bytes
64 bytes from 192.168.1.13: icmp_seq=0 ttl=64 time=6.769 ms
64 bytes from 192.168.1.13: icmp_seq=1 ttl=64 time=5.961 ms
64 bytes from 192.168.1.13: icmp_seq=2 ttl=64 time=7.013 ms
64 bytes from 192.168.1.13: icmp_seq=3 ttl=64 time=8.050 ms
64 bytes from 192.168.1.13: icmp_seq=4 ttl=64 time=9.456 ms
64 bytes from 192.168.1.13: icmp_seq=5 ttl=64 time=10.468 ms
64 bytes from 192.168.1.13: icmp_seq=6 ttl=64 time=10.134 ms
64 bytes from 192.168.1.13: icmp_seq=7 ttl=64 time=3.384 ms
64 bytes from 192.168.1.13: icmp_seq=8 ttl=64 time=5.241 ms
64 bytes from 192.168.1.13: icmp_seq=9 ttl=64 time=6.240 ms
64 bytes from 192.168.1.13: icmp_seq=10 ttl=64 time=7.274 ms
64 bytes from 192.168.1.13: icmp_seq=11 ttl=64 time=8.276 ms
64 bytes from 192.168.1.13: icmp_seq=12 ttl=64 time=10.027 ms
64 bytes from 192.168.1.13: icmp_seq=13 ttl=64 time=11.822 ms
64 bytes from 192.168.1.13: icmp_seq=14 ttl=64 time=3.276 ms
64 bytes from 192.168.1.13: icmp_seq=15 ttl=64 time=4.531 ms
64 bytes from 192.168.1.13: icmp_seq=16 ttl=64 time=6.061 ms
64 bytes from 192.168.1.13: icmp_seq=17 ttl=64 time=7.084 ms
64 bytes from 192.168.1.13: icmp_seq=18 ttl=64 time=8.407 ms
64 bytes from 192.168.1.13: icmp_seq=19 ttl=64 time=10.335 ms
64 bytes from 192.168.1.13: icmp_seq=20 ttl=64 time=11.338 ms
64 bytes from 192.168.1.13: icmp_seq=21 ttl=64 time=12.406 ms
64 bytes from 192.168.1.13: icmp_seq=22 ttl=64 time=3.844 ms
^C
--- 192.168.1.13 ping statistics ---
23 packets transmitted, 23 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 3.276/7.713/12.406/2.648 ms
ITMGR2:~ username$
Switching environment is DLINK DGS 1500-28's and 52's with Fibre GBIC's for the swtiches that run to my warehouse.
Thoughts?
So im having a weird issue here and perhaps its not even an issue.
ICMP / ping traffic to my switches on my LAN is very confusing. I am not a novice network technician but im no expert either and I cant quite explain it.
I ping any device on my network and the highest ping time I get is 2ms (for when it has to run across to the other end of my warehouse generally) however pinging any of my switches varies in ping times for example:
ITMGR2:~ username$ ping 192.168.1.13
PING 192.168.1.13 (192.168.1.13): 56 data bytes
64 bytes from 192.168.1.13: icmp_seq=0 ttl=64 time=6.769 ms
64 bytes from 192.168.1.13: icmp_seq=1 ttl=64 time=5.961 ms
64 bytes from 192.168.1.13: icmp_seq=2 ttl=64 time=7.013 ms
64 bytes from 192.168.1.13: icmp_seq=3 ttl=64 time=8.050 ms
64 bytes from 192.168.1.13: icmp_seq=4 ttl=64 time=9.456 ms
64 bytes from 192.168.1.13: icmp_seq=5 ttl=64 time=10.468 ms
64 bytes from 192.168.1.13: icmp_seq=6 ttl=64 time=10.134 ms
64 bytes from 192.168.1.13: icmp_seq=7 ttl=64 time=3.384 ms
64 bytes from 192.168.1.13: icmp_seq=8 ttl=64 time=5.241 ms
64 bytes from 192.168.1.13: icmp_seq=9 ttl=64 time=6.240 ms
64 bytes from 192.168.1.13: icmp_seq=10 ttl=64 time=7.274 ms
64 bytes from 192.168.1.13: icmp_seq=11 ttl=64 time=8.276 ms
64 bytes from 192.168.1.13: icmp_seq=12 ttl=64 time=10.027 ms
64 bytes from 192.168.1.13: icmp_seq=13 ttl=64 time=11.822 ms
64 bytes from 192.168.1.13: icmp_seq=14 ttl=64 time=3.276 ms
64 bytes from 192.168.1.13: icmp_seq=15 ttl=64 time=4.531 ms
64 bytes from 192.168.1.13: icmp_seq=16 ttl=64 time=6.061 ms
64 bytes from 192.168.1.13: icmp_seq=17 ttl=64 time=7.084 ms
64 bytes from 192.168.1.13: icmp_seq=18 ttl=64 time=8.407 ms
64 bytes from 192.168.1.13: icmp_seq=19 ttl=64 time=10.335 ms
64 bytes from 192.168.1.13: icmp_seq=20 ttl=64 time=11.338 ms
64 bytes from 192.168.1.13: icmp_seq=21 ttl=64 time=12.406 ms
64 bytes from 192.168.1.13: icmp_seq=22 ttl=64 time=3.844 ms
^C
--- 192.168.1.13 ping statistics ---
23 packets transmitted, 23 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 3.276/7.713/12.406/2.648 ms
ITMGR2:~ username$
Switching environment is DLINK DGS 1500-28's and 52's with Fibre GBIC's for the swtiches that run to my warehouse.
Thoughts?
Andy M
Not sure on the cause but you are not alone - I've seen this happen with managed Netgear switches in multiple environments - internal pings between devices getting 1-2ms, switches vary from 4-10ms. As we don't have any reported or noticed problems with access or network performance we never really looked into it closely.
From my understanding ( I don't have experience with your switches though), it is because of how the icmp packet is being processed by the switches. Switches are optimized to switch packets coming in on one interface and forwarded out another interface. So processing is streamlined and minimized to provide a consistent low latency to communication per switching hop. Better hardware can do it at near line rate due to lookup efficiencies
When you ping the device though, it isn't doing a lookup in a table to effectively just forward it on. It is pushing it up the processing ladder in the device and the switch now has to "think" about what to do with it. This added processing increases the latency a little bit. How much depends on how busy the switch is and how much resources it has at its disposal. This kind of explains why its not totally stable either. Because the cpu load on the switch can vary depending on what it needs to do to support its other functions so those icmp packets queue and wait their turn for processing.
At least that's how it was explained to me. Since it made since, I've never looked further into why so much latency is added unless it starts impacting something.
When you ping the device though, it isn't doing a lookup in a table to effectively just forward it on. It is pushing it up the processing ladder in the device and the switch now has to "think" about what to do with it. This added processing increases the latency a little bit. How much depends on how busy the switch is and how much resources it has at its disposal. This kind of explains why its not totally stable either. Because the cpu load on the switch can vary depending on what it needs to do to support its other functions so those icmp packets queue and wait their turn for processing.
At least that's how it was explained to me. Since it made since, I've never looked further into why so much latency is added unless it starts impacting something.
When you are ping devices on your network, you have to figured the switch is just switching the traffic at Layer 2 based on it's MAC address table. Now when you ping it's own management address, it now has to deal with Layer 3, so it now has to strip the frame off to read the packet. Then re-encapsulate the packet and send it back. That is my take on it.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.