Showing results for 
Show  only  | Search instead for 
Did you mean: 

This product reached the end of support date on March 31, 2021.

Log failures on AMD 12.4 - Daily cron list


AMD 12.4 on RHL v7.2 is throwing log errors on the daily cron list.

Is there any way to stop this? or find out why those files don't exist.


Dynatrace Pro
Dynatrace Pro

(Linux admin perspective)

The quickest/easiest way to eliminate those messages is to touch each file so it exists, though that does not address why it does not exist to begin with.

Past experience is this usually happens when the AMD has bee "hardened" a little too much, and some of the components are no longer given write permission to their log files. Resetting the permissions on /var/log/adlex/ to world writable and then restarting the AMD software should allow the files to be created if it is a permissions issue, and then also give you the service account/permissions reference information to adjust your permissions appropriately.

-- Erik

Dynatrace Pro
Dynatrace Pro

I get the same on a fresh install ((RHEL7.2 AMD12.4.1) using the kickstart build, and no additional lock downs.

The root cause, for my instance at least, is because the AMD isn't running the CBA processes (what provides the packet capture services for the RUMC wizards), the cba* logs have never been created.

page2trans and v2page are missing because I'm not running any software services that generate ADS data, so those logs aren't being created either.

pldmonitor.log I'm not sure about.

probeng.log is a problem, I feel. I'm running this AMD in NG mode (beta), and it's creating a rtmng.log, not probeng.log.

To properly fix the issue, the labs should fix the logrotate configs adding a 'missingok' line to the logs which aren't always generated, and either change the logrotate settings from probeng to rtmng, or fix the code to create probeng, instead of rtmng files.

in /etc/logrotate.d/ there'll be a few files relating to the AMD processes. Ones of interest here are.

adlex - pldmonitor is here
cba - cba-start is here
page2trans - page2trans-start is here
rtmng - has the wrong log filename in it. rename the top line, probeng.log to rtmng.log
v2page - v2page-start is here

To 'fix' before the labs do. Add a 'missingok' line to each logrotate config file. e.g. to fix the v2page_start.log

That should stop the logrotate process sending those errors. They're not critical at all, but no point being alerted to them all the time.

Note this fix is probably temporary, as it will most likely be overwritten by any AMD upgrades. The labs should correct the logrotate configs properly.

Based on past experience, these modules should always be running, even if only to generate empty (but required present) data files. Did you disable page2trans and v2page?

I ran into a similar issue with the new ndata file not being generated, and the CAS delaying ten minutes waiting for the missing files.

-- Erik