Script Error Clearance Please

[Carlton Patterson]

Hello Community,

I understand why I'm getting the following error on the router, but I don't know how to clear it.

I think the error is preventing any other scripts from being activated.

Aug 20 12:13:04.796: %HA_EM-7-FMS_POLICY_MAX_ENTRIES: fh_schedule_policy: Class default: Maximum number of 64 script publish entries exceeded; some events have been discarded

Can someone please show me how to clear it?

Cheers

Carlton

[Carlton Patterson]

Hi Community,

Can I get some help with this please?

Is there a way I can prevent this error message appearing?

Cheers

[Carlton Patterson]

Community,

I don't know if the following is related to the above error, but when I do a debug of tcl I see the following:

fh_tty_read_cmd] read not ready

I'm not sure what that means

[Carlton Patterson]

The thing I don't understand is that the error says

Maximum number of 64 script publish entries exceeded

However, I only have 9 scripts, see below.

Any thoughts would be greatly appreciated.

No.  Class     Type    Event Type          Trap  Time Registered           Name

1    script    user    timer watchdog      Off   Fri Aug 16 10:55:04 2013  stateInterfacestatus.tcl

name {stateInterfacestatus} time 60.000

nice 1 queue-priority low maxrun 20.000 scheduler rp_primary

2    script    user    timer watchdog      Off   Fri Aug 16 11:39:28 2013  utilisationHighInterfaceUt

ilization.

tcl

name {utilisationHighInterfaceU

tilizat} time 60.000

nice 1 queue-priority low maxrun 60.000 scheduler rp_primary

3    script    user    timer watchdog      Off   Fri Aug 16 13:14:10 2013  cpuHighCPUUsage.tcl

name {cpuHighCPUUsage} time 60.000

nice 1 queue-priority low maxrun 20.000 scheduler rp_primary

4    script    user    timer watchdog      Off   Fri Aug 16 13:16:09 2013  memoryHighMemory.tcl

name {memoryHighMemory} time 60.000

nice 1 queue-priority low maxrun 20.000 scheduler rp_primary

5    script    user    syslog              Off   Fri Aug 16 13:29:04 2013  configCONFCHG.tcl

occurs 1 pattern {SYS-5-CONFIG_I.*}

nice 0 queue-priority normal maxrun 20.000 scheduler rp_primary

6    script    user    syslog              Off   Fri Aug 16 17:29:09 2013  envAssetChassisIssue.tcl

severity-warning severity-minor severity-major severity-critical severity-fatal occurs 1 pattern {ENV.*}

nice 0 queue-priority normal maxrun 60.000 scheduler rp_primary

7    script    user    syslog              Off   Fri Aug 16 18:57:07 2013  syslogOSPF.tcl

priority all pattern {OSPF}

nice 0 queue-priority normal maxrun 20.000 scheduler rp_primary

8    script    user    syslog              Off   Fri Aug 16 19:22:52 2013  syslogIPNAT.tcl

priority all pattern {IPNAT}

nice 0 queue-priority normal maxrun 20.000 scheduler rp_primary

9    script    user    syslog              Off   Wed Aug 21 12:49:20 2013  syslogIOSVLAN.tcl

priority all pattern {SW_VLAN|VLAN|SW_VLAN_ISSU

|SPANTREE_

VLAN_SW|PL

ATFORM_VLA

N}

nice 0 queue-priority normal maxrun 20.000 scheduler rp_primary

Cheers

[Joe Clarke · Accepted solution]

As I've said before, this message means you're receiving more events that you can service.  This could mean that you get so many matching syslog messages at one time that you cannot run enough concurrent policies to service them all.  Since by default, only one Tcl policy can run at a time, that would mean if you generate 66 matching syslogs at one time, the 66th message will be dropped out of the event queue.  You can play with the number of policies that can execute at once using the "event manager scheduler script thread class default number" command, but you have to be careful that you don't overwhelm the device.

[Carlton Patterson]

Joseph,

I fully understand now. I didn't get it the first time.

Thanks again mate..

Cheers

[Carlton Patterson]

Hi Joseph,

One last question regarding this issue:

Can you let me know how to prevent the syslog message from appearing:

%HA_EM-7-FMS_POLICY_MAX_ENTRIES: fh_schedule_policy: Class default: Maximum number of 64 script publish entries exceeded; some events have been discarded

Or better still, can you show me how to have the above syslog message appear only once in syslog messages?

Cheers

[Joe Clarke]

There is no one fix for this.  I told you one way to prevent it in my reply already.

[Carlton Patterson]

Hi Joseph,

Thanks again for responding, if you mean the following;

You can play with the number of policies that can execute at once using the "event manager scheduler script thread class default number"

Then this won't prevent the message from appearing in the syslog over and over....

[Joe Clarke]

It may help.  It's all about message rate, how long it takes the EEM policy to run, and how many threads you can get away with on your device.  You'll have to test to find the right balance.  And if the rate of messages is still too high, you'll have to reconsider your policy design.