Few of our branch office switches are running with high memory utilization, we are receiving alerts on our monitoring tool for the same. Monitoring tool shows Memory 1 is running with 90% (above threshold limit) and other memory is fine. These are 3500 series switches used in branch offices.
My concern is....
Is this really a issue or a bug?
What is Memory 1 and Memory 2?
How to identify the issue, any commands other than Show process memory?
IOS (tm) C3500XL Software (C3500XL-C3H2S-M), Version 12.0(5.1)XP,
SWITCH03>sh pro mem
Total: 12415464, Used: 10696620, Free: 1718844
Hoooooooooooooooo boy! I've never seen this IOS version before. This is probably better than the Interim Software (XU) version. Is this a 3524XL-EN, 3548XL-EN or 3508XL-EN?
High memory utilization on the 2900XL/3500XL is commonly attributed by high bandwidth utilization and/or spanning-tree loop.
May I ask if you can post your config?
Please Help me to resolve this issue.
These are WS-C3548-XL-EN switches.
Below is the another switch details with the same model number.
IOS (tm) C3500XL Software (C3500XL-C3H2S-M), Version 12.0(5.1)XP, MAINTENANCE IN
ROM: Bootstrap program is C3500XL boot loader
INB00switch1 uptime is 30 weeks, 20 hours, 37 minutes
System returned to ROM by power-on
System restarted at 07:42:59 EDT Thu Jun 4 2009
System image file is "flash:c3500XL-c3h2s-mz-120.5.1-XP.bin"
cisco WS-C3548-XL (PowerPC403) processor (revision 0x01) with 16384K/1024K bytes
Model number: WS-C3548-XL-EN
Sorry, I cant provide you the configure it against the company policy. But i can tell you what are all we have configured.
We have configured following on switches
Let me know if you require any other details.
About the only thing you will probably be able to do with those is see if you can get ahold of the last IOS version they made for that switch and load it up. Those have been out of support for over 2 years now. a lot of switches if they have been up a long time will eventually get a low memory condition due to a slow memory leak where it does not release memory that is not being used like it should . Usually a reboot will fix this in most cases. You also have to make sure it does not have another cause like an infected machine generating a ton of traffic etc... a look thru this doc might help.
Could you please check from the (sh processes memory) output , what exactly consumes memory from the Allocated portion?
check also the CPU status? Post back the output from both.
As A side note, your configuration are simple and straight forward and shouldnt consume such memory.
Can I know more information on *Dead* Process which always has PID "0" (Zero) and Memory details for the same.
I could see the holding memory with high values, does it affect the memory utilization?
Can *Dead* process be killed and restarted without rebooting the switch?
Hi, Ive run show process memory on individual switches, could anyone advise how I could get this information from over 500+ switches within DNA? I want to be able to compare results over a period of time. Thanks
High Memory Utilization
You can check the memory by "sh proc mem" command . You can enable VTP pruning in the network so that unnecessary broadcast will not pass through the switch