Problem when checking profile with manual addition of servers

[obtim]

1. I have a folder with manually added profiles with 23729 servers in it
2. I run a check of this folder for “dead servers”. The check finds many dead servers from the first step of the job
3. Since there are a lot of servers, the program crashes during the process: always different. Sometimes it has time to check 500 servers, sometimes several thousand.
4. After the program crashes, the number of servers 23729 does not decrease, taking into account the checked servers. That is, it remains 23729. Although 2000 dead servers could have been found.
5. In the folder settings, the “Automatically delete servers that fail validation” mode is enabled.

[GooRingX]

1. "Automatically delete servers that fail validation" is only valid for manually triggered delay detection
2. You can limit the number of servers actually used by settings-auto select-maximum number of server
3. With more than 20,000 nodes, if the node quality is poor, it may take more than 20,000s=5.5hours to complete the delay detection. Will you wait that long? So you need to pre-screen the nodes. More than 20,000 nodes are too extreme

[obtim]

1. "Automatically delete servers that fail validation" is only valid for manually triggered delay detection
2. You can limit the number of servers actually used by settings-auto select-maximum number of server
3. With more than 20,000 nodes, if the node quality is poor, it may take more than 20,000s=5.5hours to complete the delay detection. Will you wait that long? So you need to pre-screen the nodes. More than 20,000 nodes are too extreme

The program crashes during the test. The results are not saved (not counted). So I have to run it again. It crashes again, and so on and so forth.
If the program would save the results and reduce the initial number of servers, taking into account the retired (not working), then I am ready to restart it.

[GooRingX]

1. "Automatically delete servers that fail validation" is only valid for manually triggered delay detection
2. You can limit the number of servers actually used by settings-auto select-maximum number of server
3. With more than 20,000 nodes, if the node quality is poor, it may take more than 20,000s=5.5hours to complete the delay detection. Will you wait that long? So you need to pre-screen the nodes. More than 20,000 nodes are too extreme

The program crashes during the test. The results are not saved (not counted). So I have to run it again. It crashes again, and so on and so forth. If the program would save the results and reduce the initial number of servers, taking into account the retired (not working), then I am ready to restart it.

The delay detection will not be saved to the file until it is completed. Otherwise, it will cause a lot of disk reading and writing, affecting device performance