I would test and figure out what your max ASIC load is along with your oversubscription rate. Then connect your host and devices accordingly. For example I have 3750Gs deployed still for several things and can only have two ports of a group of 4 connected to a specific type of device. Anymore and I start seeing the same issue as you.
Really the only solution is more hardware to even the load or better preforming switches. I would jump next to a Cisco 4948 if you want something similar to the 3750. Ryan Booth Blog.moveingonesandzeros.net<http://blog.moveingonesandzeros.net/> From: Michael Sprouffske [mailto:[email protected]] Sent: Thursday, April 18, 2013 4:15 PM To: Booth, Ryan Subject: Re: [c-nsp] cisco buffer misses Yeah from what I'm hearing, these switches are bad for our VM environment. Is there anything that can be done to help this issue without replacing equipment? ________________________________ From: "Booth, Ryan" <[email protected]<mailto:[email protected]>> To: Michael Sprouffske <[email protected]<mailto:[email protected]>>; "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Sent: Thursday, April 18, 2013 2:06 PM Subject: RE: [c-nsp] cisco buffer misses No QoS will just degrade your performance. 3750s are notorious for having horrible buffer sizes and crappy ASIC performance. QoS will just carve this up into smaller chunks for each group of traffic making the situation worse My initial guess is you are overloading and ASIC or the switch all together with either micro-burst or high levels of traffic. Check your ASICs to see if you are getting drops. Im not familiar with the ASIC performance and port-layout of the X series so you might need to move some heavy loaded links away from each other to even out the load across ASICS. TAC can also give you specifics on ASIC performance and help. Ryan Booth Blog.moveingonesandzeros.net<http://blog.moveingonesandzeros.net/> -----Original Message----- From: cisco-nsp [mailto:[email protected]<mailto:[email protected]>] On Behalf Of Michael Sprouffske Sent: Thursday, April 18, 2013 12:47 PM To: [email protected]<mailto:[email protected]> Subject: Re: [c-nsp] cisco buffer misses I'm thinking that some switch qos needs to be put in place to resolve this issue. What does everyone think? We currently don't have qos running in the switched network. We only have qos running on the routers for the uplinks. ________________________________ From: Michael Sprouffske <[email protected]<mailto:[email protected]>> To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Sent: Thursday, April 18, 2013 9:40 AM Subject: cisco buffer misses Could someone give me some insight as to what is causing the misses? I'm currently researching this on the inter webs. I also notice an interface with several drops as well. model: WS-C3750X-24T-S Buffer elements: 1061 in free list (500 max allowed) 3479036431 hits, 0 misses, 1024 created Public buffer pools: Small buffers, 104 bytes (total 50, permanent 50, peak 119 @ 7w0d): 49 in free list (20 min, 150 max allowed) 1712506149 hits, 23 misses, 69 trims, 69 created 0 failures (0 no memory) Middle buffers, 600 bytes (total 25, permanent 25, peak 85 @ 7w0d): 23 in free list (10 min, 150 max allowed) 167702 hits, 58 misses, 174 trims, 174 created 0 failures (0 no memory) Big buffers, 1536 bytes (total 50, permanent 50, peak 119 @ 7w0d): 50 in free list (5 min, 150 max allowed) 244703400 hits, 39 misses, 117 trims, 117 created 0 failures (0 no memory) VeryBig buffers, 4520 bytes (total 16, permanent 10, peak 16 @ 7w0d): 0 in free list (0 min, 100 max allowed) 59 hits, 3 misses, 1429 trims, 1435 created 0 failures (0 no memory) Large buffers, 5024 bytes (total 0, permanent 0): 0 in free list (0 min, 10 max allowed) 0 hits, 0 misses, 0 trims, 0 created 0 failures (0 no memory) Huge buffers, 18024 bytes (total 4, permanent 0, peak 7 @ 7w0d): 4 in free list (0 min, 4 max allowed) 145363512 hits, 412449 misses, 822729 trims, 822733 created 0 failures (0 no memory) GigabitEthernet1/0/11 is up, line protocol is up (connected) Hardware is Gigabit Ethernet, address is 649e.f3e2.7b8b (bia 649e.f3e2.7b8b) Description: server ports MTU 9000 bytes, BW 100000 Kbit, DLY 100 usec, reliability 255/255, txload 1/255, rxload 1/255 Encapsulation ARPA, loopback not set Keepalive set (10 sec) Full-duplex, 100Mb/s, media type is 10/100/1000BaseTX input flow-control is off, output flow-control is unsupported ARP type: ARPA, ARP Timeout 04:00:00 Last input 00:00:16, output 00:00:01, output hang never Last clearing of "show interface" counters never Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 54011 Queueing strategy: fifo Output queue: 0/40 (size/max) 5 minute input rate 0 bits/sec, 0 packets/sec 5 minute output rate 49000 bits/sec, 28 packets/sec 837676 packets input, 118543994 bytes, 0 no buffer Received 492811 broadcasts (481207 multicasts) 0 runts, 0 giants, 0 throttles 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored 0 watchdog, 481207 multicast, 0 pause input 0 input packets with dribble condition detected 670450255 packets output, 115399962776 bytes, 0 underruns 0 output errors, 0 collisions, 1 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 _______________________________________________ cisco-nsp mailing list [email protected]<mailto:[email protected]> https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
