Cisco Switching/Routing :: High Memory And CPU With 3750 Stack Switch?
Dec 26, 2012
We are having 3750 Stack switch as our edge switch. Here we are having some issues as low memory & itermediate High CPU utilization.For memory when we checked its shwoing maximum memory is allocated to DHCPD receive & ARP Input as follows
Processor Pool Total: 70829380 Used: 56809532 Free: 14019848
I/O Pool Total: 12582912 Used: 8554920 Free: 4027992
Driver te Pool Total: 1048576 Used: 40 Free: 1048536
[code].....
i want to understand why this 2 proceses (DHCPD & ARP input) taking more memory?High CPU: We also facing intermediate high CPU utilization (above 95%). When CPU is high there were 2 processes consuming more CPU are 'Spanning Tree' (about 20%) & ARP Input (about 66%). we also checked the ARP entries, there were 1500 ARP entries & 2 incomplete ARP (out of 6000 ARP supported)
I have a new cisco 3750 stack comprised of 8 members (6 x 3750G-48PS and 2 x 3750V2-48PS) running 'c3750-ipservicesk9-mz.122-58.SE2.bin'. I am seeing high memory utilisation, it seems to sit fairly stable and I do not see an increase but this has only been running a few days. No errors or performance issues. Should this be a cause for concern?
We have a 3750 stack with a mix of 3750G and 3750X switches. The current active template on the switch is desktop default and we are running EIGRP on the switch with large routing table so we get the TCAM error that a specific prefix cannot be programmed in TCAM memory so it will be software forwarded.%PLATFORM_UCAST-4-PREFIX: One or more, more specific prefixes could not be programmed into TCAM and are being covered by a less specific prefix, and the packets may be software forwarded
This error is quite frequent on a daily basis. My question is that will this error create a high CPU situation and affect traffic processing on the stack causing ESX Hosts to become unresponsive?
WS-C3750G-24TS-S1U switch stack with high CPU utilization. 3 switches are stacked and using PBR on that. I am attaching the show cpu process as an attachment.The current template is "desktop routing" template.The selected template optimizes the resources in the switch to support this level of features for 8 routed interfaces and 1024 VLANs. [code]
I have a Switch WS-C3750E-48PD-SF with IOS 12.2(58)SE2 which reports more than 80% in utilization of memory I don't know what is causing the high memory utilization.
We added Site02 for redundancy and DR. Also added new SAN and ESX hosts. The CPU on device Site01Server01 is not very happy when files are being transferred between the sites. Since the setup is the same at both sites, why is the problem only at site01?
Site02 is newer and we have two 3750X-24 with IOS 12.2(55)SE3 on ipservices, for site02server01; no problem. Site01server01 is a stack of two 3750G-24PS with IOS 12.2(58)SE on ipservices.
I upgraded the newest IOS 15.0(1)SE for our SW C2960-24-TT-L, the process was succeseful. However, we found SW memory utilization raised to 82% this morning in normal situation. (Momery: Total 21341260, Used17471692) Did you guys meet the same problem after upgrading to 15.0(1)SE? I also would like to know what the normal memory utilization is? and what is the impact of memory utilization highly?
I upgrade the IOS of my switch 3560 from C3560-IPBASE-M), Version 12.2(25)SEB4, To C3560-IPBASEK9-M), Version 12.2(55)SE5 after that the utilizatio of the memory increase to 80% I attached the output of show memory statistics history
i am seeing very high utilization on a random basis on the stack. and the logs indicate me that there is a pattern where there is a stack power cable shows inserted (which was never unplugged) followed by sudden spike in the utilization.
001018: *Mar 6 16:21:22.138: %PLATFORM_STACKPOWER-6-CABLE_EVENT: Switch 4 stack power cable 1 inserted 001019: *Mar 6 18:18:37.982: %SYS-1-CPURISINGTHRESHOLD: Threshold: Total CPU
i want to remove an access layer 3750 48 POE master switch and replace with a new 3750 switch in the stack. i want to copy the same configuration to new switch since the old switch is having POE issue and I got replacement of the malfunctioned switch. we have VTP domain is configured in the network...
I configured 3750 A switch with vlan 20 and its IP address 192.168.20.41Its default gateway was 192.168.20.3Then i configured 3750 B switch with same default gateway and vlan 20 IP 192.168.20.43My question is now when we stack it becomes single switch and now vlan 20 ip address is 192.168.20.43 thats only IP i can see.So how does stack switch choose vlan 20 IP?Does it choose highest IP address between two switches if they have same vlan 20 as in my case?Also when i go to switch 3750 b by session command and do sh ip route it does not show ip default gateway .Also it shows vlan 20 as admin down .
We have a stack of 3 cisco swithes 3750s WS-C3750G-24TS 12.2(25)SEB4 .We plan to recover a switch from the stack as it is not needed any more The switch we plan to recover is the existing Master.If we change the priority (switch 1 priority) and make another switch the master using command line will we have any loss of connectivity to the servers conecting to the stack ?
We have a WS-C3750G-24PS switch running 12.2(50)SE , C3750-IPBASEK9-M in production & stand alone and we want to ask a new Cisco WS-C3750-24P running 12.2(55)SE3 , C3750-IPBASEK9-M into the stack with the 1st one.
Just want to know if it is possible and are there any problems with the IOS being different version.
I have a stack with a lot of stack-port changes, but on all the ports, how can I determine the faulty switch in the stack, cables are already verified. We did also a restart of the whole stack power off/on
running version
WS-C3750-48P 12.2(55)SE C3750-IPBASEK9-M
This is the output after 6weeks
Switch#/ Stack Neighbor Cable Link Link Sync # In Port# Port Length OK Active OK Changes Loopback Status To LinkOK -------- ------ -------- -------- ---- ------ ---- --------- -------- 1/1 OK 4 1
In the current set up I have a LAN stack of 5 3750 switches configured as an access stack. As per the requirement I need to partition the current 5 switch stack into two individual stacks of 3 & 2 switches respectively. Kindly let me know the step by step process withh all the necessary commands, reloads and reconnections. I purpose to do it by changing the priority of current member switch to 15 so that its elected as Master in the new partitioned switch stack. But not sure what all steps required and in which order.
In Cisco 3750 facing high CPU utilization for specific period. During this period traffic response times across the switch degrades. Steps to identify interface potentially generating traffic. I have attached 'sh proc cpu history' , ' sh ip traffic', 'sh proc cpu sorted' during troubled period .
I'm facing high CPU utilization issue due to a ssh process on 3750 switch. What is the use of SSH proces in the switch? Can I disable it for troubleshooting?
will there be any impact if I disable it?
#sh processes cpu sorted 5min
CPU utilization for five seconds: 99%/0%; one minute: 66%; five minutes: 46%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process 159 1896078 589986 3213 3.96% 10.76% 14.00% 0 Hulc LED Process 165 32871 591 55619 77.42% 33.40% 8.85% 3 SSH Process
One switch of our stack of 8 3750 have reboot with no reason. Nobody was in the room server when it appears et there was'nt any electrical problem at this moment.
I have four-switch Cat 3750 stack that has been working reliably for a couple years. Suddenly the other morning it reboots itself, and has been doing so every 24 hours since. take a look at the attached log? This should really be a Smartnet thing, but as it turns out the warranty we have on this is not Smartnet .
At a customer i've tried to add a new 3750 switch til a running stack.I used the following proccess:
1. connect stack cables to the new switch
2. bootup new switch.
Normally the new switch will become a member of the stack, but here, the switch says its the stack master. I can see that the stack port from the old switches connecting the new is up.The only way i could make it work was to provisioning the new switch as switch nr x, save, power off and then connect it and boot it up.The IOS i the same on all switches.
in our network we were using three 3750-48-s switches with stacking one of the switch due to some hardware failure power problem,
now i want to add a new switch 3750-48-s in the existing Stack two switches , the old two switches ios version is = 12.2(25r)SEC the new Catalyst 3750 io verion is = 12.2(35)SE5
how to add this switch in the existing two stack switches , with documentation
I have stack of 2 switches 3750?I config etherchannel between them.
here is result
2 Po2(SD) LACP Fa1/0/15(I) Fa2/0/15(I)
Both ports are up up but standalone Int port channel 2 is down down.Need to know if this is default behaviour when we config etherchannel between stack switches?
is there a document that explains the compatibility of 3750 series that can be stacked together? for example can we stack 3750-x with 3750-v2 without the power sharing? Can we stack a L3 3750 with L2 3750 without upgrading the L2 switch IOS to ip services?
We have a live stack with 4 Cat 3750 switches of mixed varieties. Switch 1 crashed the other day and didn't come back up. A power reset brought the switch online again, and switch 2 is now running as the stack master. We are going to replace switch 1 in a staged approach to minimise client downtime. Switch 1 is a ws-c3750-48ts and we will be replacing it with a ws-c3750g-48ts. I will be pre-provisioning and installing the new switch into the stack as switch 5 and configuing the ports to be the same as switch 1. The cables will then be moved over from switch 1 to switch 5 so the changeover time should be minimal. When all cutover, I'll power down switch 1, reconfigure the stack cables and remove switch 1 from the stack.
My question is to what occurs next. If I perform a 'no switch 1 provision ws-c3750-48ts' will I be left with switch 2 through to 5 (ie no switch 1)? I can live with this moving forward. I know I can renumber switch 5 to be switch 1 if I want, but my understanding is that I have to reload the member switch which is something I'd like to avoid at this stage. In my reading, I saw the statement - "Note: You cannot use the switch current-stack-member-number renumber new-stack-member-number global configuration command on a provisioned switch. If you do, the command is rejected." I'm a bit confused by this statement, I'm assuming that it is saying that if I do as above but don't issue the 'no switch 1 provision' command, I cannot renumber switch 5 to switch 1 but as long as I use the 'no switch 1 provision' command first, I can renumber switch 5 to switch 1? Will all of my switch 5 port configuration information and any associated references (eg port-channel configuration etc) be changed from gi5/0/x to gi1/0/x if I do renumber?
I have looked on notes to add a switch to an existing stack and haven’t identified answers on how the IOS will be affected on the new switch – will it downgrade to the current stack version or will the current stack upgrade to its version. At this moment in time I would prefer if the new switch IOS downgraded to the current stack IOS version.The new switch is a 3750V2 – will this affect how it joins the stack?The new switch has the image of IPBASEK9-M – again will this affect how it joins the stack?
I have a Cisco 3750G-48PS Running IOS 12.2(40) and was wondering how many physical ports are supported in an LACP Configuration. Is it 4 or 8 ports in a single switch.
If I have a dual 3750 stacked together, and I want to configure (8) port LACP, Can I take (4) ports from each switch in the stack and LACP together.
I have been looking at the Cisco Hardware docs and I can't find anything on the proper way to setup a 2 switch stack for Stackwise redundancy and Stackpower redundancy. Everything details 3 or more switches.
I have 4 switches in a 3750-X stack.Here is the output from the stack why are the switches out of order and I'm not able to bring up switch 1 port 1.
switch 4 port 2 shows not connect it is I've checked the cables.
Power stack name: Powerstack-4 Stack mode: Power sharing Switch 1: Power budget: 695 Low port priority value: 22 High port priority value: 13 Switch priority value: 4 Port 1 status: Shut Port 2 status: Connected Neighbor on port 1: 0000.0000.0000 Neighbor on port 2: c471.fe84.e100
Switch 2: Power budget: 713 Low port priority value: 21 High port priority value: 12 Switch priority value: 3 Port 1 status: Connected Port 2 status: Connected Neighbor on port 1: c471.fe85.3000 Neighbor on port 2: c471.fe80.2800
Switch 4: Power budget: 697 Low port priority value: 19 High port priority value: 10 Switch priority value: 1 Port 1 status: Connected Port 2 status: Not connected Neighbor on port 1: c471.fe80.2800 Neighbor on port 2: 0000.0000.0000
Switch 3: Power budget: 710 Low port priority value: 20 High port priority value: 11 Switch priority value: 2 Port 1 status: Connected Port 2 status: Connected Neighbor on port 1: c471.fe84.e100 Neighbor on port 2: c471.fe7f.eb00s
We have 2 3750 24 port fiber switches in a stack and we would like to change the member to become the master. Both are running the same IOS and they are alive on the network. I read that you can change the who becomes the master by setting the priority number higher on the member, which we did. However this requires a reboot of the member to take effect. Is there a way to force this without having to reboot? One other note, both switches in stack have a single gig connection that goes to the core (6500). One gig link is blocked due to spanning-tree and this is on the member that we want to become the master.
