Start a Conversation

Unsolved

This post is more than 5 years old

GU

87969

July 25th, 2013 12:00

Inventory and Compliance Issues

I am experiencing some inventory and update compliancy issues.  I recently upgraded from OME 1.1 to 1.2, and have run the "Clear catalogue" patch, which did resolve the problem where compliancy was wrong in cases where BIOS 6.3.0 is installed and 6.3.0 is required.  Now I am experincing a few other isues, highlighted below (The following relates to 4 R710s, all purchased at the same time last year, all running the same FW/drivers versions):

 

 

  1. Inconsistent information has been provided in the OME console.  For example, when I look at a server's details in  Manage - Devices, it shows BIOS 6.3.0 is installed.  BUT this same server is uncompliant because it has BIOS 6.1.0 installed and 6.1.3 is available.  I have verified that 6.3.0 IS installed.  I do not know if this is an Inventory issue, as I had initiated inventory many, many times to try to resolve this over the period of a few days.  How can inventory info be different for the same server in differnet lpaces within the console?
  2. These same servers also listed about 5 other updates available.  I verified what versions of these components were installed looking in OMSA on the server, and verified that all of the other required updates were valid.
  3. To try to resolve the issue, I deleted these four servers from OME, and re-ran Discovery (of the DRAC and SNMP on the servers themselves.)
  4. Three of the four servers were re-discovered.  one was not rediscovered, not sure why -- are there any logs I can look at when discovery runs to see what happens for each IP it tries to discover?
  5. For the three servers that were discovered, they are showing as noncompliant, BUT only that the DRAC firmware needs to be updated from v. 1.80 to 1.95.  The other 4 updates that were showing up as required prior to deleting them do not show up as needed now.  I have re-run Discovery and inventory on these three servers several times over the past 2 days.
  6. I updated the DRAC firmware on one of the servers, and it succeeded.  Then it did not show up as either compliant or not compliant.  30 minutes later I looked in Manage - Devices and it was missing.  I had to re-run disocvery for it to show up in Mane - Devices and in the compliancy section.  its almost as if the server got deleted automatically after running the DRAC firmware update.

Another issue, with another set of servers.  I discovered two R620s this morning.  Neither of them showed up under the Compliant, Non-compliant, or Non-inventoried systems tabs.  So I ran Discovery and Inventory again,  As of now (about 2 1/2 hours later) they still do not show up under any of the complaincy tabs.  Inventory has run on them, under Manage - Devices I can see all of their details such as health status, connection staus, Device Name, Model, Service tag, OS, firmware info, etc.)

Community Manager

 • 

711 Posts

July 25th, 2013 13:00

Hi,

Thanks for your post.

Few questions for you. Are you using in band system updates using OMSA or are you using out of band system updates using iDrac? Reading the description it looks like the inventory information was not getting updated in OME for some reason. That resulted in compliance report based on stale inventory information. Looking at #5, it does look like OME received the updated inventory information and the compliance report was updated.

For #1, you can see Firmware Information table and Software Inventory Information table when you look at inventory data using Manage->Devices option. Was the same BIOS version showing up in both tables. From the details, it looks like Firmware Information table was showing 6.3.0. If Software Inventory Information was showing 6.1.0, that part is not getting refreshed. If you are gathering inventory in band using OMSA and SNMP you might have to restart OMSA service. If you are using iDrac and WS-MAN you might want to check if CSIOR option is enabled.

Regarding the missing iDrac, if you discover the iDrac and server both, OME correlates those devices and you get a single entry for server in the device tree. If you click on the server, you should be able to see the iDrac IP address and health status as part of server inventory. You will also be able to launch the iDrac console by right clicking on the server.

Regards

Abhijit

July 26th, 2013 10:00

I can not check to see if the information in the Firmware Information table matched what is in the Software Inventory table, because we had a maintenance window for the servers yesterday afternoon and had to update them (we had to do it manually since OME wasn't working as we expected it to.)

Hopefully you can answer a few questions for me, so I will have some additional information for when this happens again.

1.) What would cause the information to differ in the Firmware Information table and Software Inventory tables?

2.) Are there any logs created for the discovery process?  As I mentioned before, we have four servers that I deleted from OME and when I tried to re-discover them, only 3 are getting discovered.  I would like to see what happened when it tried to discover the 4th IP.

3.) Is there a way to enable CSIOR without a server restart (usinf racadm or in the iDRAC web GUI)?  The only info I could find instructs you to boot to the iDRAC config and turn it on there.

July 28th, 2013 06:00

Hi,

1. We have not seen information mismatch between Firmware information and Software inventory table.

2. We need to enable logging in dconfig file to get discovery logs:

C:\Program Files\Dell\SysMgt\Essentials\configuration\dconfig.ini”

Change “INFO_LOGGING=N” to “INFO_LOGGING=FS;I;C:\OMElog

under task management

ENABLE_COMMAND_LINE_LOGGING=1

Restart Netmon and Taskmanager services. Logs will be updated in 'C' drive.

3. You can enable CSIOR using RACADM without server restart. Use the below command :

C:\rac>racadm –r xxx.xxx.xxx.xxx –u root –p calvin set LifecycleController.LCAttributes.CollectSystemInventoryOnRestart Enables. However inventory is collected only after server reboot even though CSIOR is enabled.

July 30th, 2013 11:00

There was no line "INFO_LOGGING=N" in dconfig.ini.  I did find a whole logging section (pasted below for reference), which I was able to edit and get logging to work.

I got the fourth server to be discivered and inventory correctly after re-installing OMSA.

I will go ahead and enable CSIOR on our servers.  it would be nice if this were enabled by default.

=====================

Logging section I found in dconfig.ini:

[LOGGING]

; The master switch (true/false).

LOG_ENABLED=false

; Log to file (true/false), if true Output file is $(LOG_FILE_PREFIX)$(SEPARATOR) $(LOG_FILE_SUFFIX).

LOG_TO_FILE=false

LOG_FILE_PREFIX=c:\windows\temp

LOG_FILE_SUFFIX=_log.txt

; Filter based on tags (true/false), if true  log only the statements with Tag name specified.  

LOG_FILTER=false

LOG_FILTER_TAG=PPA

; Log levels are on a scale 1-7 where 7 is full verbose, trace (7), debug (6), info (5), warning (4), error (3), fatal (2), and output (1),

; and range is from Minimum scale to Maximuml scale of logging levels, you get all logging information between the range of levels specified.

LOG_LEVEL_MIN=info

LOG_LEVEL_MAX=output

; Log to Visual Studio and CMD Screen (true/false).

LOG_TO_VS=false

LOG_TO_SCREEN=false

July 30th, 2013 12:00

I was able to use the racadm command provided to enable CSIOR on an iDRAC7 (Poweredge R620)

racadm –r xxx.xxx.xxx.xxx –u root –p calvin set LifecycleController.LCAttributes.CollectSystemInventoryOnRestart Enabled

But, when I try to run it on an iDRAC6 (R710) I get the following error: "ERROR: Invalid subcommand specified."

July 31st, 2013 05:00

11G does not support enabling CSIOR using Racadm. It can be done using CTRL E only.

Lakshmi.

October 29th, 2013 13:00

Nevermind I reread the post and realized the logging command was on the OME server,  not the OMSA client.

October 29th, 2013 13:00

I'm experiencing these exact same issues as the original poster but under RHEL 5 and RHEL 6.  What commands would we use to enable logging as well as I see the Windows commands?

Unfortunately we have ~500 R710's all requiring the Lifecycle controller setting to enabled.  All were being updated to 6.3.0 and unfortunately after applying firmware updates they do not report properly in OME 1.2 unless I press F10 to enter the Lifecycle controller, exit out and then boot into the OS normally.  In our situation we're only able to use the iDRAC for monitoring even though we're in the process of deploying OMSA to each of the hosts.  The security team won't allow us to open TCP/1278 directly to each machine in the DMZ's only via the iDRAC.

No Events found!

Top