Probes not processing large number of CIs
We are seeing probes in probe status not processing large counts of CIs (300-500K related to desktop inventory scannng via the UD agent). Totaldiscovered CI in sending queue counts are not dropping and probe to app server and vice-versa communication is good as when we run dropunsent data on probe JMX we see counts in UI probe status drop. The counts in probe JMX unsent seem to align with what we see in UI on probe status. We havee also recently deployed the probe touchtime process to true and set start and endtimes to off hours as recommended by R&D. We have an open ticket with Support for this but has anyone else experienced this and would appreciate any ideas on where to look.
One of the probes in our environment (10.31) recently started doing something that sounds similar. When we ran the Host Connection by SNMP job, about 97% of the trigger CIs failed with uable to connect message withing minutes. But, if i ran some of the failed CIs individually, it would connect and run successfully. We tried a few options, clearing the probe data batch job, etc, but it would not run correctly.
I ended up uninstalling the probe ucmdb software and then simply reinstalled. This fixed whatever problems there were. Don't have an explanation but it solved out issue.