NetXMS Support Forum

Update: I just checked, and seems that there is another bug with same conversion (of instance filter scripts), that may lead to script corruption. We are currently building version 5.0.2 with the fix in upgrade procedure. If possible, restore database from backup and run upgrade again with version 5.0.2. Alternatively, restore only content of column "instd_filter" in table "items", and do script conversion manually where needed.

Best regards,
Victor

Hi!
This is instance filter script, number in script name is DCI ID. Please find this DCI and show the script.
Best regards,
Victor

Why do most data collection entries appear as yellow after upgrade, but they are still collecting metrics?

and i see error "Script(DCI::NODE-NAME::177748::InstanceFilter) execution error: Error in line 1: sytnax error, unexpected '?', expecting $end" Where can i find the script that is generating this error ?

update was done directly to 5.0.1.

At my installation I'm using a networkmap for every floor with a jpg or png image as background and some other informations. These images are of different sizes. Til now it worked fine getting the image from image-library and it was displayed perfectly.
With version 5.0.1 mapsize within map properties has to be changed from width=0 and height=0 to a value that fits imagesize to be displayed correctly.
Is this a bug or intended behavior?
If this is intended behavior I have to change the values for many, many maps....

Hi. Version 5.0.1 for macOS (Apple Silicon) is not available. Error 404 on download.

Thanks for the answer. It's better

Hi,

"Scripted Pie Charts" had the script removed after migrating to version 5.0.1. At least visually they do not appear in the scripts tab and the graph is no longer rendered.

"Pie Chart for table DCI" appears with all items in black.



Congratulations to everyone on the improvements in the latest version.

This is intended change. We keep list of breaking changes in the documentation, it's worth checking prior to upgrade:
https://www.netxms.org/documentation/adminguide/installation.html#major-changes-between-releases

Hi,

$node.snmpOID
is no longer returning the OID starting with dot (.) , many scripts will need adjustments in version 5.0.
Is it a bug or implementation change?

Good Morning, I updated a ubuntu/mssql based server to 5.0.0 then 5.0.1 and have lost the ability to poll icmp targets through an agent serving as a proxy.

• Nodes were added through discovery, but also fails on new, manually added nodes
• The proxy agents are otherwise working (including other subagents like mqtt)
• I upgraded the agents to 5.0.1
• I have added and removed proxy nodes from the zone (we have several zones)
• The only error I see logged on server or agent is "Unable to establish connection with proxy node"
• I can manually ping from the server where the agent is installed to the other nodes
• I also checked the database with the dbmgr utility
• All other aspects, other than icmp, appear to work correctly

Update:  The agent using the proxy is on a windows server, the proxy function works correctly with linux-based agents.

This happened right after the upgrade, so I wanted to flag it here. 

Thank you!