Hello,
I'm using an old version of Dataminer 9.6 with SQL Database (Single Node/No Offload).
Since 20th December the server is facing a lot of timeout alarms which happear randomly on many elements.
There is no network issue.
I noticed as well some error alarms related to SLDataminer
dLooking at the SLDataminer logging file i can see this error sometimes
SLDataMiner.exe|5340|CDataMiner::OffloadFile|ERR|-1|Failed offloading file C:\Skyline DataMiner\System Cache\Local\DataAvg\32310-316_current2025_01_03_07_44_23_1112699.dat. 0x80131500
But i have no clue it is related.
Unfortunately, I cannot make a diagnosis based on the provided information, but I hope the following information will be helpful.
A timeout or loss of communication alarm typically occurs when a device does not respond or responds with unexpected results. Depending on the type of connector (protocol/driver), it is always recommended to verify whether the communication channels between the DataMiner agent hosting the affected element and the device are open. You can use third-party tools such as an MIB browser for an SNMP connector or Wireshark for network analysis. Additionally, the Stream Viewer tool Collecting data to report an issue to Technical Support | DataMiner Docs can help check if any response is received from the device.
If you believe that the loss of communication alarms are not caused by communication issues between DataMiner and the device (i.e., all necessary ports are open), we recommend opening a ticket with the TechSupport team by emailing techsupport@skyline.be.
Please use the SLLogCollector tool - DataMiner Dojo to collect logging data. However, note that this tool may not work in DataMiner version 9.6.0.0.
PS: You are currently running an unsupported DataMiner version. We highly recommend consulting Tech Support regarding an upgrade to a supported version.
Regarding the Thread problem in SLDataMiner Database Offload Thread
These alarms occur weekly, every Friday, due to a scheduled task configured to optimize the database.
The Database Offload Thread [local] pushes data from the buffer (C:\Skyline DataMiner\System Cache\Local\..) into the local MySQL database.
Every Friday at 04:00, DataMiner performs optimization queries on the tables of its local database. These queries ensure that when DataMiner has cleaned up old records from a database table, the size-on-disk of that table is properly updated. Without these queries, the size-on-disk of the database would never shrink.
This alarm was common in older DataMiner versions. However, we have made significant improvements(and are continuously improving) since 9.6.0.0.0, including migrating from a local database to a distributed clustered storage system (Cassandra and OpenSearch).
Given this, we highly recommend upgrading to the latest DataMiner version. Please review the following documentation:
The left panel of the Stream Viewer contains the connections, you should see the HTTP and SNMP connection trees. From there, you should be able to see on which connection the timeout occurs.
Hello Christine,
Thanks for your reply. Very helpful as always 🙂
I notice some timeout on the backup agent as well but with different timestamp.
I will do the wireshark capture.
But there is something i never get completely about timeout. For an Element which is using both HTTP and SNMP then how Dataminer is dealing with two kind of timeout. How to differenciate timeout SNMP and API from the Alarm ?