High Latency on T1 Line

[Litanel]

Hi All ,

Its my first post here, Please help me on this as its giving me nightmares and a lot of pressure on my neck.

i have  4  T1 lines with 2mbps bandwidth going from HQ to 4 remote offices. Ping latency between HQ to one of the remote office is quite high and causing huge service disruption.

ISP have ran a test on line 2ce which they said its clean and there is no issue on their side.

When i ping the serial interface from the router,

ping latency result  Success rate is 60 percent (3/5), round-trip min/avg/max = 1084/1157/1240 ms

But when i did a loopback ping test. latency looks normal.

i might be wrong but this tells me there is a lot more traffic trying to leave the interface than it can handle. i just don't know where to look at

 i have changed wic cards, changed cable on both ends but no luck.

However, there are some errors on the Serial interfaces. 

Below is show interfaces <interface-id> ouput

Serial0/2/1 is up, line protocol is up
Hardware is WIC MBRD Serial
Internet address is 10.x.x.x./24
MTU 1500 bytes, BW 1984 Kbit/sec, DLY 20000 usec,
reliability 255/255, txload 250/255, rxload 88/255
Encapsulation HDLC, loopback not set
Keepalive set (10 sec)
Last input 00:00:00, output 00:00:02, output hang never
Last clearing of "show interface" counters 3d20h
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 2129836
Queueing strategy: fifo
Output queue: 33/40 (size/max)
5 minute input rate 690000 bits/sec, 202 packets/sec
5 minute output rate 1947000 bits/sec, 209 packets/sec
17201870 packets input, 2048560447 bytes, 0 no buffer
Received 36490 broadcasts (0 IP multicasts)
107 runts, 0 giants, 0 throttles
244 input errors, 101 CRC, 0 frame, 0 overrun, 0 ignored, 36 abort
24929107 packets output, 1180313150 bytes, 0 underruns
0 output errors, 0 collisions, 838 interface resets
0 unknown protocol drops
0 output buffer failures, 0 output buffers swapped out
7 carrier transitions
DCD=up DSR=up DTR=up RTS=up CTS=up

Please help guys. 

[Georg Pauwen]

Hello,

first of all, a T1 (usually 1,5Mbps) is extremely low capacity nowadays, so your line probably suffers from massive congestion. You have 300+ interface resets in less than 3 days, which are likely caused by missed keepalives due to congestion, or a bad line. I would suggest to get with your provider and check the status and the health of the line, and possibly think of a line upgrade...

[Litanel]

Thank You Georg for your response.

Yes, i think you are right cos other 3 T1 lines have same interface configurations. I will request an line integrity test from ISP and see what they find. I will keep you posted on things

thanks a lot 

[syrehan]

Hello Litanel,

T1 line maxs out at 1.5 Mbps and by the looks of it 838 times interface resets means there is a lot of congestion probably due to one of the facts that a lot of traffic is trying to leave the interface but is being choked due to the lack of bandwidth. Your ping results also show 40% packet loss due to the same reason. I would recommend upgrading the WAN connection to see if it helps. 

[Litanel]

@syrehan thanks for your response.

ISP said we cannot upgrade WAN connection further than what we currently have due to infrastructural constraint in the area where remote office is located. 

i am waiting on ISP to carry out integrity test on line and hoping that will alleviate issues.

But while i await them please is there any other tweaking i could on router to  improve performance?

[syrehan · Accepted solution]

Since the ISP can't upgrade the WAN connection. In that case I would recommend you to check and see how many users are accessing that particular connection? Limiting the bandwidth and implementing load balancing may provide temporary solution.

[Joseph W. Doherty]

policy-may Try_This
class class-default
fair-queue

interface Serial0/2/1
service-policy output Try_This

[Litanel]

@Joseph W. Doherty @syrehan @Georg Pauwen. Hi guys thanks all for your contribution. i have acted on all you suggested and there seem to be an improvement on performance now.

However ISP also called later in the evening yesterday  me after i have applied suggested actions in afternoon that they found some problems on their COMS which affected the data link. They will sort out issue today and hopefully there will much better performance.

Now i got pressure off my neck :-) .. thanks to you all once again

[syrehan]

You're welcome! Glad to be of assistance. You can mark the query as solved.