cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
4093
Views
5
Helpful
5
Replies

ISE root disk filesystem

registry
Level 1
Level 1

We are using the Cisco ISE and one of the appliance servers is not calculating used disk space correctly...it states 75%, but it should be more like 70%. We have the threshold snmp trap set to 25 ( =75% used ) and the server started to send alerts reaching that level. After removing some patches and corefile dumps - "show disk" still says 75% used. Now, if I poll the server with snmpwalk, the used disk is 70%. But it is still trapping...

 

show disk:

 

Internal filesystems:
/ : 75% used ( 10532540 of 14987616)
/dev : 0% used ( 0 of 32842248)
/dev/shm : 0% used ( 0 of 32851936)
/run : 1% used ( 1092 of 32851936)
/sys/fs/cgroup : 0% used ( 0 of 32851936)
/storedconfig : 2% used ( 1588 of 95054)
/boot : 21% used ( 94882 of 487634)
/boot/efi : 4% used ( 8792 of 276312)
/tmp : 1% used ( 7000 of 1983056)
/opt : 26% used ( 276781400 of 1125348968)
/run/user/440 : 0% used ( 0 of 6570388)
/run/user/301 : 0% used ( 0 of 6570388)
/run/user/321 : 0% used ( 0 of 6570388)
/run/user/0 : 0% used ( 0 of 6570388)
/run/user/304 : 0% used ( 0 of 6570388)
/run/user/303 : 0% used ( 0 of 6570388)
/run/user/322 : 0% used ( 0 of 6570388)
/run/user/308 : 0% used ( 0 of 6570388)
all internal filesystems have sufficient free space

 

System info:

Cisco Application Deployment Engine OS Release: 3.0
ADE-OS Build Version: 3.0.3.030
ADE-OS System Architecture: x86_64

 

The second question I have is: What is taking up all space? There are no big files in the root system, but still 70% is used. I have tried to reload the system.

 

Regards / Fred

 

1 Accepted Solution

Accepted Solutions

Jason Kunst
Cisco Employee
Cisco Employee
I would checking out the info under http://cs.co/ise-help how to get help in the community, we are not the TAC, you should open a case with them to debug

View solution in original post

5 Replies 5

Jason Kunst
Cisco Employee
Cisco Employee
I would checking out the info under http://cs.co/ise-help how to get help in the community, we are not the TAC, you should open a case with them to debug

hslai
Cisco Employee
Cisco Employee

As Jason said, the root file system can't be cleaned up without TAC assistance. What we are able to clean usually is only the /localdisk, which uses the file system mounted in /opt. The "show disks" gives the outputs from Linux command "df" and is more accurate than the data obtained from a SNMP query, which is also unclear from which OID you received the data.

Maybe it's time to create a cron job for ISE systems that does a daily disk cleanup (around midnight or whatever) - these used to be part of every unix system ever since I can remember.  I am sure the TAC have gathered enough best practice experience of problems that occur most frequently that this can be automated.  This should not require human intervention.

ISE does have such cron jobs in cleaning files. The original post did not provide enough data and would be best investigated by TAC.

packetplumber9
Level 1
Level 1

I've seen the "show disk" output take some time to update after deleting files to free up space.  This could at least temporarily explain the difference between the command output and SNMP polling results

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: