Common IT Asset Management / SAM system down cases

book

Article ID: 145592

calendar_today

Updated On:

Products

CA Service Management - Asset Portfolio Management ASSET PORTFOLIO MGMT- SERVER

Issue/Introduction

Some common troubleshooting advice for when an IT Asset Management system is failing.

As well as ITAM itself, many sites use CA Software Asset Manager (SAM) alongside, so this document will cover problems with both.

Resolution

IT Asset Management problems

Installation issues

When facing an installation that has failed, the first thing to do is check the install logs. This registry key will show where the logs created during the installation process – typically the user’s temp folder:

HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\CA\Shared\ITAM\LogFileFullPath

One of the most common causes for a problem with the install is a missed Server Component, e.g. with IIS. These are documented in the Operating System Requirements, but a more step-by-step guide is in Article Id 138024.

If there is indeed an issue that requires a second run of the installer then it is vital to clean up the existing failed run first. The process is in Article Id 125170.

Password login issues

There are two levels of login issue; ITAM connecting to the database, and users trying to log in to the application.

If the problem is with the “mdbadmin” database user connecting to the database, then Article Id 139049 contains the steps for updating the ITAM configuration files that store the encrypted password. Typically, this is necessary if the password expires or is changed by another installer. You may also find that this has only been set incorrectly for some of the files, if a previous attempt was made to update and others were missed; this would result in certain functionality returning “You do not have required privileges to perform the action” even if you are an administrator user in the UI.

User login failure

For user authentication, ITAM users EEM. The default application instance in the EEM server is “APM”; if you need to move EEM to a new server or recreate the instance, see Article Id 133653.

However, there is an exception to this – the Super Administrator user “uapmadmin” is stored in the MDB database. The table “al_process_account” contains certificate_key and admin_password entries for the user. Article Id 35820 contains further details on how to reset this.

 

 

Software Asset Manager problems

Installation / upgrade issues

CA SAM presently lacks an installer. Instead, it must be set up manually, by following the instructions in the Administration manual that comes in the zip file. For further detail on these, Article Id 131421 can be helpful.

An issue that can be seen while following this is that there is an incompatibility with certain releases of the SourceGuardian package. There is a known working release attached to Article Id 137908. This results in “unknown type: [h]” errors to the screen at the Database Installation step.

Meanwhile, a full upgrade of SAM and/or PHP is documented in Article Id 10481 as preserving configuration while doing so requires careful copying of files.

Once installed, the patching process is defined in both the manual and Article Id 139818.

Problems with the cron job system

If cron jobs are failing, then check the errors to see why. Go to Admin > Configuration > Cron jobs, click the … and select Cron Log.

“WARNING: ASP-W0.  This process is already running and cannot be started more than once” means that a previous run did not finish before the new one was launched. Details for clearing this up in the case of a hung/stuck job are in Article Id 56968.

If this error is seen because the first job is sometimes taking too long for a genuine reason, rather than “stuck” this can be avoided by setting “Enable parallel run”.

However, if jobs are not being launched, and so no error is logged in its cron log, then the most likely cause is an issue with the cronmaster task itself. This is defined in the Windows Task Scheduler during installation; if it has not been created at all, or does not have the parallel run setting “Run a new instance in parallel” then multiple tasks cannot be launched. In addition, ensure that the repeat interval of the cronmaster task reflects the granularity expected of the jobs defined – a job cannot run every 5 minutes if the cronmaster only looks for new jobs every 10. For further info, see Article Id 142109.