15.12.2018, 20:49
Hello,
I think LMs should be prepared for dealing with looped telegrams (should have some watchdog).
I have a few LMs in one installations and I have a problem with 1 script in specific moment. Probably it generates some loop of telegrams but I cannot see this on LM, it looks like there is normal load. But all LMs in this moment goes more and more loaded (CPU and Memory load). CPU up to 4 and 5, normal <1 load.
After disable script Load balancer goes to normal. LM3 V2, LM4, LM5 must be rebooted because their still has CPU load on this level until reboot, and memory ~80%, and using them is very, very, very slow and unstable(including send reboot command).
All LMs (excluding LB) hasn't TP connected and are in IP Routing mode with IP selected.
I think that all LMs should have be prepared for such issue and should can go back to normal state after loop or some other error will be solved.
I think LMs should be prepared for dealing with looped telegrams (should have some watchdog).
I have a few LMs in one installations and I have a problem with 1 script in specific moment. Probably it generates some loop of telegrams but I cannot see this on LM, it looks like there is normal load. But all LMs in this moment goes more and more loaded (CPU and Memory load). CPU up to 4 and 5, normal <1 load.
After disable script Load balancer goes to normal. LM3 V2, LM4, LM5 must be rebooted because their still has CPU load on this level until reboot, and memory ~80%, and using them is very, very, very slow and unstable(including send reboot command).
All LMs (excluding LB) hasn't TP connected and are in IP Routing mode with IP selected.
I think that all LMs should have be prepared for such issue and should can go back to normal state after loop or some other error will be solved.
Done is better than perfect