Automated Action Summary

[orsonjoon]

Hello,

When a new configuration is saved on a router the router sends a syslog message to the Ciscoworks server somewhat like this:

SYS

5

CONFIG_I

Configured from console by

The date and time the configuration that is changed are correct, and RME syslog. automated actions job "Config fetch" should initiate an automatic new config fetch job because of this:

Automated Action Summary
Name:	Config Fetch
Devices:	*
State:	Enabled
Action Type:	Invoke Config Collection
Messages:	*-*-*-RESTART:* RESTART-*-*-*:* SYS-*-5-ONLINE:* *-*-*-CONFIG_I:* SYS-*-5-RELOAD:* CONFIG-*-*-*:* *-*-*-CONFIG:* OIR-*-6-INSCARD:* Nodemgr-*-5-CE:*REBOOT* CPU_REDUN-*-6-BOOTED_AS_ACTIVE:* CPU_REDUN-*-5-SWITCHOVER:* CPU_REDUN-*-6-RUNNING_CONFIG_CHG :* CPU_REDUN-*-5-RCSF_SYNCED:* CPU_REDUN-*-6-STARTUP_CONFIG_CHG:* CPU_REDUN-*-5-STARTUP_CONFIG_SYNCED:* SNMP-*-5-COLDSTART:* SYS-*-6-CFG_CHG:*telnet* SYS-*-6-CFG_CHG:*Console* *-*-*-OIR:* PIX-*-5-111005:* SYS-*-6-CFG_CHG:*SNMP* SYS-*-6-CFG_CHG:*ssh* SNMP-*-5-CONF:* PORT-*-5-CONF:* CHAS-*-5-CONF:* DIAG-*-5-CONF:*

The problem is that this does not work, and the new config doesn't get picked up by RME.

Could anyone please help

[Joe Clarke]

Even though RME may receive a config change syslog message, it does not mean that the config will be updated in the archive.  RME will download the config, then compare it to the currently archived version.  If there are no interesting changes, RME will discard the downloaded version to prevent archive churn.  The interesting commands are those not listed in the Excluded Commands list under RME > Admin > Config Mgmt.

If the configuration is really changing, there may be a problem with the archive for this device.  Try performing a manual Sync Archive job for this device to make sure that works successfully.  If it does, you could be hitting CSCtc18888.  This is fixed in LSM 4.0 and will also be fixed in the upcoming LMS 3.2 SP release this spring.

[orsonjoon]

Hello Joseph,

Thanks for the answer.

I failed to mension that a manual sync job succeeds

without a hich.

I guess we're hitting another bug again .

Bad luck for us I guess.

Now I scheduled a manual sync job for al devices every day as an unwanted workaround for this bug.

[Joe Clarke]

If you want to rule this bug out, click the "Open Service Request" link in the actions panel of this thread and request the patch from TAC.  If

by chance it's not this bug, they can collect the AnalyzerDebug.log to see if there is another problem.  However, given the fact that messages are being written to the DB, I would believe this is the bug in question.