LMS 4: DFM issues

As suggested by Joseph Clarke, I started a new thread regarding Fault manager issues in our LMS 4 installation.
The original post: https://supportforums.cisco.com/message/3233282
The following menus do not open (error messages are shown instead):
Monitor> Fault Settings> Setup
Monitor> Threshold Settings> Fault
Reports> Fault and Event> History

I just checked the affected menus and it now seems the issues are gone. Could it be that the IPM patch (which fixed SNMP v3 and IPSLA issue) also fixed DFM?

Similar Messages

LMS 4.0.1 - DFM issue

HI,
I installed the LMS 4.0.1 and I'm facing a problem with DFM. None Fault is showed.
The Fault Discovery Process never finishes, it is always Running, according attached file.
I checked the LMS Process MAnagement and 4 process are as "Transient Terminated". The process are: CSDiscovery, DCRDevicePoll, FHPurgeTask,UTMajorAcquisition. I started all process but 2 minutes later the process goes to "Transient Terminated".
All Windows Services related with CiscoWorks are UP.
Someone knows this issue?
Thanks for your help.
Luis

Hi duong,
I'm using the ping sweep. not using the "Use DCR as Seed List" as there were more than 5000 devices.
Earliar LMS pick the devices. once the discovery was running for a long time . to stop the same we have restarted
the CSDiscovry process. After this we are facing the issue.
Channa

LMS 4.0 DFM issue

Hi Experts, Recently I have installed LMS4.0 and we are facing issues with DFM. All devices are in DFM in known state. recently our link from HQ to branch went down and all branch devices were unreachable but we did not receive a sinlge alert from DFM and when I go to DFM to see if alerts are there but I dont see any alerts related to reachability..its only showing interface up/down message....even though i checked the availability report in HUM and uptime report in HUM it does not show there as well that device was not reachable or some downtime....
attached is the version of applications installed in LMS4.0... Please help me to resolve this issue ?

MDF isn't something that needs to be configured. It is an architectural component of LMS. The most likely reason you cannot create VLANs is that LMS does not know what your read-write community string is. Discovery only populates DCR with the read-only string. Go to Inventory > Add/Import/Manage Devices and edit the credentials for your switches. Add the proper read-write community string, and then you should be able to configure the VLANs.

LMS 3.1 - DFM Issues

Hi There,
I'm having a few problems with my setup, some more important than others, but causing troubles for me all the same (I've listed them in priority)
1) Right now I have the default Scheduled rediscovery set in DFM to run weekly. That's all well and good, except that when it runs, less than 10 of my 664 devices get put into the "known" state, while the rest - after 6-8 hrs - get thrown into the "questioned" state. When I try to suspend the schedule, I get an error (see attached: rediscovery_error.jpg, and associated Rediscovery.txt). I have a case open with the TAC right now about this, but it has not been resolved. The only fix I have at the moment is to rebuild the DFM DB, then re-add all the devices, and then I have to unmanage all the interfaces/ips that I'm not using. Seems like an incredible waste of time.
2) When adding a new device to the DFM (automatically pulled when I add it to the DCR) all the existing devices return to their default managed/unmanaged state. Again, at this point, I have to go back and unmanage all the interfaces/ips, etc... (May be related to above, I'm not sure.)
3) My jrm.log file is currently sitting at over 650MB (not KB). This equates to roughly 7.3 million lines of text, where 99% of the lines are "Test Exception is : java.sql.SQLException: JZ0S2: Statement object has already been closed." (see attached: jrm.jpg)
4) I would also like to know if the emails that are sent out by the DFM can be changed - a lot of the information included in them is not relevant to me, and I would like to add the information that is.
I've also attached the output of pdshow and systeminfo, which seems to be the first thing most people are asked for.
I'm running LMS 3.1, with an updated DFM (3.1.3) on Windows 2003 Enterprise.
Please let me know if there is anything else I can provide that may help.

I should also point out that once you upgrade to LMS 3.2, there is a consolidated patch for DFM to apply at http://tools.cisco.com/support/downloads/go/ImageList.x?relVer=3.2.0&mdfid=282640771&sftType=CiscoWorks+Device+Fault+Manager+Patches&optPlat=Windows&nodecount=2&edesignator=null&modelName=CiscoWorks+Device+Fault+Manager+3.2&treeMdfId=268439477&treeName=Network+Management&modifmdfid=&imname=&hybrid=Y&imst=N&lr=Y which will fix quite a few DFM bugs which are still present in 3.2. You should download, and apply this patch, and that should give you a much better DFM experience.

Email Notifications DFM issue

HI all
I have a Cisco Works LMS 3.2 with DFM 3.2 on a Server 2008 Service Pack 2, and the email notifications on the DFM are not working, I recived emails from others modules but didnt from DFM, And I checked all the configurations and seems ok for me.
I have the same CiscoWorks on a Server 2003 and the notifications are working properly.
Could you help me with this issue
Thanks for your help
Regards
Rafael Turriago

Hi,
Have you set the SMTP server under DFM > Configuration > Other Configurations > SMTP Default Server & set up a Notification Subscription?
Regards.

LMS 3.2 : issue with custom portal : high CPU

Hi,
Customer has a big issue with LMS.
When he connects to the standard portal, it works fine.
pdshow seems ok.
But when he wants to work with custom portal (portal.jpg), it makes a long time to display information and when he wants select something it hangs.
CPU is very high. He must stoped/restart LMS to have a normal CPU.
I asked him to :
- restart LMS, he made at 14h30
- CPU was ok and he worked with standard portal withou problems but when he tried to put debug for portal CPU increase at 15h30:
Common Services -> server -> Admin -> CS Log Configurations: debug on CiscoWorks Portal, CiscoWorks portlets and CS Home
Process which takes a lot of CPU is tomcat, you will find pdshow.txt, stdout.log and stderr.log.
He tries to go to the custom portal but it hangs, you will find logs from debug : CSHomePage.log and CSPortal.log
he sent me AAD.log because there was " high severity alert".
Many thanks for your help,
Elisabeth

The most obvious problem is that DfmBroker is down. Check to make sure tcp/9002 is not in user, then try a server reboot. See if the AAD problems go away (and perhaps the CPU problem will as well). If not, post the NMSROOT/objects/local/logs/broker.log.

LMS 4.0 issue

Hi,
I am facing issue with LMS 4.0. The Core Switch is showing in RED color,and device type as UNKNOWN. It was working fine but some how it is showing this problem. Can anyone help me out with this issue.
Thanks,
Shakir
Sent from Cisco Technical Support Android App

I see your issue has two set of problems, but one is not relevant for now.
1. Device shows unknown
2. Shows Red in Topology
You should fix the unknown cause first as a device is shown unknown till the time inventory collection isnt succeeded.
Once it is done, the device should show the known state and the device icon.
Try, to run inventory collection job on your device and check if succeeds/fails and share the result. If failing, the failure resutl should show the underlying issue, why the device is still failing.
Once a device is in known state, the other modules will only then start working with that device, and hence it should come in the Green state thereafter.
-Thanks

LMS 4.1 issue with collection summary board

Hello,
something strange, collection summary says 2 config archives failed but don't say succeeded, see below:
Thanks,

The other are probably "partial"
This probably means I did not get the vlan.dat file but got startup and running config.
You can see this in: Configuration > Configuration archive > Summary
I find the collection summary very confusing since it doesn't provide all info needed
Cheers,
Michel

Cisco LMS 3.2 SYSLOG not storing after 10 days

Hi ,
Im facing one issue with Cisco LMS 3.2
Issue : The logs is generating only for 10 days and post that im not able to see the logs. I have not done any config changes. The only change i have done is i have completely reinstalled the LMS. i did multiple troubleshoot but not able to resolve this isse. It would be great If any some one is able to help me in this isse. Thanks.
Regards,
Juliet

Dear Vinod
Thanks for ur response and the problem has been resolved.
The purge policy was set to 60 days only .The problem in reports viewing setting.
Syslog folder under LMS would store syslog reports of both the device as well as applications for defined folder size , which in your case was 1 MB ( same can be viewed under log generator option). The older reports would get deleted from the folder upon reaching the limit.
The only way to view device syslog is under following option : Reports -> Reports Generator in LMS GUI where we will have to choose syslog with desired attribute.
Regards,
Juliet

Issues with coverted files of captivate from CP2 to CP4

Hi Adobe,
We have a captivate files made using CP2 and saved it as CP4 and published it for Flash8, AS2 and SCORM 1.2.
This a multi-sco course and each SCO had one CP file
1st CP file contains the simulation with few self-checks in it and the
2nd CP file contain the assessment questions.
When we tested it on the Docent LMS the following issues were identified.
1.      for 1st CP file à The bookmarking is not working
a.      Bookmarking worked fine until the course was not completed. But after completion of the course bookmarking did not work properly. Since we have multi SCO structure the leaner can revisit the completed SCO.
b.      The course did not mark as completed on the LMS on the first attempt but on 2nd attempt it did.
c.      If the course is again launched after completion and the Rewind button is clicked or slider is dragged the course gets marked as "incomplete" on the LMS and does not get changed to complete even after the last slide is viewed and we were not able to re-enroll again.
d.    The interactivities built using the inbuilt captivate templates (MCSS/MCMS/True-False) can not be re-attempted.
2.      For 2nd CP file – Score value does not get properly recorded on LMS
a. The LMS does not record the accuracy value on the LMS, due to which the completion status does not get fired.
b. If we launch and exit the module without attempting any of the questions then a Failed status is sent to the LMS. Is this something that can be resolved using the Captivate publish settings?
b. Even when the bookmarking is set to false, when the assessment is re-launched, the starts from the bookmarked position and are not able to go back to reattempt from the beginning.
Can you guide us on how we can overcome these issues.
Thanks
Regards,
Siddharth Lodha

Thanks for the reply, Bartok, but those didn't seem to work for me. Maybe I applied them incorrectly as that thread was for the Win platform and not Xcode and I might have translated the solution poorly.
Basically I went to the "Release_Cocoa64" target, and scrolled down the "Build Settings" page. I then edited the BOOST_PREPROCESSOR_DEFINITIONS field.
First I tried just appending BOOST_SYSTEM_NO_DEPRECATED.
Second I thought that this might be a macro of some sort so I tried $(BOOST_SYSTEM_NO_DEPRECATED).
Finally, I deleted all the previous values and added "BOOST_ALL_DYN_LINK BOOST_REGEX_USE_CPP_LOCALE BOOST_REGEX_RECURSIVE BOOST_HAS_ICU=1 BOOST_ FILESYSTEM_VERSION=3 BOOST_CHRONO_HEADER_ONLY BOOST_SIGNALS_NO_DEPRECATION_WARNING".
As you might guess from the experimentation I got none of these to work. The error that I was getting has remained exactly the same.
One difference that I assumed between the Win and Xcode version is that Win separates the values with a semi-colon whereas Xcode separates with a space.
My only other thought is maybe I edited the wrong field. Is BOOST_PREPROCESSOR_DEFINITIONS the one?
As a final comment, the thread you gave me and my issue really comparable? I have other projects that don't use the boost::regex libraries and they build and run without issue. It's only when I try to use boost's regular expressions in a release build that I get problems when linking (debug build is still fine).
Also the link error looks different. The thread that you gave me points at "boost::system::error_category" as not being resolved, while I have problems resolving "boost::re_detail::get_mem_block".
I don't seem to have problems with basic boost - just the regular expression stuff.

Update apache of lms 3.1

Hi,
According to this link (https://issues.apache.org/bugzilla/show_bug.cgi?id=48359) Apache has some vulnerabilities which we need to address. Is it possible to upgrade apache on the lms server?
Regards Marco Kerklaan

It turns out this vulnerablity does not affect LMS.
This issue is not affecting the LMS.
Please have a look at the following bug ID.
http://tools.cisco.com/Support/BugToolKit/search/getBugDetails.do?method=fetchBugDetails&bugId=CSCtf54872
as you can see as I suspected we are not using This module in the LMS.
So as a workaround you can safely comment the line
LoadModule isapi_module modules/mod_isapi.so
From httpd.conf file.
Regards Marco.

Cisco Prime LMS compared to NCS

What is the difference between the 2. Can NCS do the same as LMS, but with support for wireless.

LMS and NCS will be merged in the next release, most likely in the coming few months. So the "LMS VS NCS" issue will be moot.
"LMS Vs other monitoring tools" is a huge question. It really depends on what you have in your infrastructure and what you want to do with your network management system.
If you're multi-vendor, LMS falls short as it is 98% designed for all-Cisco systems. On the other hand, few (or no) competitors offer the breadth of solution for a Cisco LAN infrastructure that LMS does. Also, LMS will of course have access to internal development resources and systems that no other vendor will ever offer. If LMS has a bug or IOS has a bug, it's one vendor who needs to fix it - no finger-pointing.
Many people use Solarwinds, for example. To begin to approach LMS, one would need both Solarwinds NPM and NCM. Even then you would not have the topology service or workflow that LMS gives you. You would also not have CiscoView. Likewise HP OpenView NNM falls short. As does What's Up Gold. They all do their intended function well - they just don't have the breadth of services built-in for a Cisco shop.
If all you need is configuration file management / archiving, then sure stand up an open source RANCID installation. It might be a bit cruder and require more Linux admin skills but it's perfectly functional. You won't get any support and that's all it does, but it does it in a workable fashion.
If all you need is performance management, use Cacti.
If all you need is fault management, use Nagios.
If you need fault and performance and are multi-vendor, use NPM.
If you need it all wrapped together and are a Cisco shop, use LMS.

CiscoWorks LMS: not receveing certain SNMP traps

CiscoWorks LMS 4.0.1 (I know it's old and unsupported).
Problem: not receiving certain SNMP traps.
For example: I receive trap like "STP new root" but not like "port put to err-disabled" or my custom traps (produced by EEM scripts).
I've investigated my situation and found out that switch sends traps and they reach* LMS but somehow LMS ignores them (there's no trace of them in GUI). I've read that some traps just pass through LMS but my traps are very important and I need to know about them.
* I did Wireshark capture on LMS machine.
I'd like to know how to debug receiving of SNMP traps in LMS:
which specific debugs need to be enabled,
which specific log files need to be examined.

LMS uses DFM to process certain traps.
All traps it deems unimportant are dropped.
If you want to use the LMS GUI you can have you device send a SYSLOG message rather than a trap.
use logging source command to make the management interface send the message.
Then there is a GUI that allows you to launch an action on a message
Cheers,
Michel

Ciscoworks Server Performance Issue

Hi ,
I have a CiscoWorks Server running LMS 2.6 version.I have seen the cpu utilization is spiking always between 85 - 95%.Also observed there are atleast 15-20 instances of cwjava.exe are running. The process dbsrv9.exe and sm_server.exe is taking memory around 1.1GB.
Please let me know is this the normal behavior in ciscoworks server.We are managing around 650 devices through ciscoworks server.
Please let me know your observation.
Thanks in advance.
Thanks & Regds,
Lalit

Hi Lalit,
     When you say " the cpu utilization is spiking always between 85 - 95%." , do you mean it is sitting there continuously, or is it running at a lower level of CPU and then regularly spiking to 85 - 95 %. Has your LMS server always run with this CPU utilization, or has it changed recently for no apparent reason ?
     What platform are you running LMS 2.6 on, in particlar the CPU power and the amount of physical memory and swap space (Page file in windows) that you have configured ?
     In terms of multiple cwjava.exe processes, that is normal. To relate these processes to specific LMS process names, execute the command pdshow from the server cli and then match the process id to the process id for specific cwjava.exe processes.
     The process dbsrv9.exe is the database server. There should be multiple of these processes running, 1 for each database. If you follow the process above you will be able to relate the one that is holding a lot of memory, to a specific LMS Database process. Which one is holding the memory ?
     The process sm_server.exe is one of the main processes for DFM. It is normal for this process to hold a lot of memory and consume a lot of CPU (DFM is very resource intensive). In particular, each time you re-start the LMS daemons, DFM re-discovers all of the network elements that it is managing, and this is resource intensive and depending on the number of devices can take some time to complete.
     I hope this helps.
Regards
Derek Clothier

LMS 4.0 IP Address

Hello,
I changed the IP address and restarted my LMS server but some features don't work. I can do a system backup for example, discovery too but not Sync Archive or Manage Device State.
When I try, I have a message which say : DCMA0022: No device selected. Select one or more devices to continue.
Do I forget something?
Kind regards

There is BUG for the same issue :
lms 4.0:Issue in Sync Archive Job page
CSCte83028
seems to be fixed in 4.0 as per the BUG details
If you have a good backup try restoring it..
Thanks-
Afroz

LMS 4: DFM issues

Similar Messages

Maybe you are looking for