11 Commits

Author	SHA1	Message	Date
allegroai	6142524b84	Fix resource name with a prefix matching a resource type may cause the AutoScaler to avoid spinning down idle instances ... Fix idle workers should contain resource name and not instance type (since it's later matched to a resource name)	2022-03-30 17:17:35 +03:00
allegroai	b1120195df	Add support for obtaining cloud console logs	2022-03-24 19:37:01 +02:00
allegroai	9744d63796	Fix autoscaler spin down stuck machine, ignore unknown stale workers	2022-02-07 13:29:51 +02:00
allegroai	051998800c	Fix incorrect timeout used for stale workers	2022-02-02 15:31:37 +02:00
allegroai	4b30ac6c31	Change auto-scaler confusing log message	2022-01-31 10:51:25 +02:00
allegroai	f620c6302a	Add last worker time to AutoScaler	2021-11-30 21:13:02 +02:00
allegroai	6a6cf111be	Refactor AutoScaler	2021-11-17 19:15:48 +02:00
allegroai	d769582332	Autoscaler improvements and optimizations ... Add customizable boot bash script Fix custom bash script should be called last before starting agent Fix auto scaler spins too many instances at onces then killing the idle ones (spin time is longer than poll time)	2021-06-12 23:11:48 +03:00
allegroai	71c45fcdd3	Update Auto Scaler default values and configuration format	2021-02-01 23:44:01 +02:00
allegroai	d327f2e2b9	clearml initial version 0.17.0	2020-12-22 23:25:37 +02:00
allegroai	a460df1e68	renamed trains -> clearml	2020-12-22 21:17:56 +02:00