What is this?

This knowledgebase contains questions and answers about PRTG Network Monitor and network monitoring in general.

Learn more

PRTG Network Monitor

Intuitive to Use. Easy to manage.
More than 500,000 users rely on Paessler PRTG every day. Find out how you can reduce cost, increase QoS and ease planning, as well.

Free Download

Top Tags


View all Tags

When PRTG Probe Service restart is triggered, what is the procedure?

Votes:

0

Hello,

When PRTG Probe restart is triggered what is the procedure and what are conditions when restart is going to fail or stop?

br, Olli Jokinen

probe prtg restart service

Created on Aug 4, 2010 6:05:11 AM



3 Replies

Accepted Answer

Votes:

0

PRTG restarts the probe when serious problems have been observed.

This can be hanging scan threads (e.g. we have seen WMI to cause stalled threads) or problems with the SSL connection to the core server (e.g. we have seen SSL connections to the probe that were broken by packet loss).

It is the final and last resort of what PRTG Probe can do to "stay alive". If it fails we recommend to reboot the system. It should not be a usual thing to occur.

Created on Aug 6, 2010 7:43:42 AM by Dirk Paessler [Founder Paessler AG] (11,025) ●3 ●6



Votes:

0

Ok,

[we have trouble running the probe service on some devices]

After examining several probe logs we found out the following:

Problem Behavior Case 1 (W7 ent x86): Probe starts to terminate the threads and we can see 13 entries of this in the log file. When it's supposed to kill the rest of the threads it stops and is left in stopped state, no additional entries in log.

Problem Behavior Case 2 (XP pro x86): There is a sudden rise on alerts and message queue on the probe (400-1400 of 3000 sensors, 3k-7k messages in queue). Also at this time the probe memory goes to lowest. After several minutes (up to 20-120 mins) the probe service is able to restart by itself. This seems to be related to avalable memory which keeps declining until the service reboot after that its in original amount. If we use any "memory cleaning" software this seems to prolong this cycle significantly.

Problem Behavior Case 3 (W7 ent x86): Probe starts to terminate the threads and we can see some entries in the log file and in the middle of the thread process: "Trying to restart probe. Reason: connection unresponsive for more than 60 seconds.". The the remaining threads are killed and service is restarted properly.

Overall it seems that in x64 operating systems (xp/w7) the memory is in more stable situation than in x86 systems eg it keeps declining until service reboot. We do not have similar behavior on 2008R2 or 2003 x86 servers.

Created on Aug 6, 2010 10:24:11 AM



Votes:

0

These memory fragmentation is likely caused by one sensor type or one sensor type group. Could you try to pause sensor type groups (e.g. WMI, SNMP, etc.) to see if this changes the game?

Please contact us at support at paessler.com for further support (this knowledgebase is not meant as a forum :-) )!

Created on Aug 6, 2010 10:39:57 AM by Dirk Paessler [Founder Paessler AG] (11,025) ●3 ●6




Disclaimer: The information in the Paessler Knowledge Base comes without warranty of any kind. Use at your own risk. Before applying any instructions please exercise proper system administrator housekeeping. You must make sure that a proper backup of all your data is available.