Sapelo2 changes after Oct.24 maintenance - Revision history

Shtsai at 01:59, 30 October 2020

2020-10-30T01:59:09Z

← Older revision		Revision as of 21:59, 29 October 2020
Line 73:		Line 73:

	When connecting from Windows for the first time after the maintenance, users might encounter an error like '''POTENTIAL SECURITY BREACH''' or '''HOST IDENTIFICATION HAS CHANGED'''. Users can click '''Yes'''to continue the connection and have a new host key saved on their local machines.		When connecting from Windows for the first time after the maintenance, users might encounter an error like '''POTENTIAL SECURITY BREACH''' or '''HOST IDENTIFICATION HAS CHANGED'''. Users can click '''Yes'''to continue the connection and have a new host key saved on their local machines.

			===Compute nodes have new hostnames===

			As we expand the Sapelo2 cluster by adding more compute nodes, it is helpful to identify compute nodes by its node rack location and node slot in the rack. Therefore we now incorporate this information in the hostname of the compute nodes. The new hostnames have the format '''rackname'''-'''nodeslot''', for example, node ra6-8 refers to a node on node rack ra6 and that occupies slot number 8 in this rack.

2020-10-29T23:32:26Z

← Older revision		Revision as of 19:32, 29 October 2020
Line 16:		Line 16:

	When you connect to Sapelo2 using the hostname sapelo2.gacrc.uga.edu, you will be connected to the same environment as Sap2test, before the maintenance window. The sap2test.gacrc.uga.edu hostname is not active anymore.		When you connect to Sapelo2 using the hostname sapelo2.gacrc.uga.edu, you will be connected to the same environment as Sap2test, before the maintenance window. The sap2test.gacrc.uga.edu hostname is not active anymore.

			===New Toolchains and Software Environment installed===

			Sapelo2's operating system was updated from CentOS 7.5 to CentOS 7.8. Compiler toolchains and the software application modules were updated as well. For more information, please see [[Software on Sapelo2]].

	===Error connecting to Sapelo2===		===Error connecting to Sapelo2===

2020-10-29T23:28:04Z

← Older revision		Revision as of 19:28, 29 October 2020
Line 6:		Line 6:

	Sapelo2 runs the Slurm queueing system now.		Sapelo2 runs the Slurm queueing system now.

			To help users familiarize with Slurm and the test cluster environment, we have prepared some training videos that are available from the GACRC's Kaltura channel at https://kaltura.uga.edu/channel/GACRC/176125031 (login with MyID and password is required). Please also refer to [[Running Jobs on Sapelo2]] and [[Migrating from Torque to Slurm]].

	===Changes in the maximum walltime limit of the partitions (queues)===		===Changes in the maximum walltime limit of the partitions (queues)===

2020-10-29T23:17:03Z

← Older revision		Revision as of 19:17, 29 October 2020
Line 1:		Line 1:


			Below are some user visible changes implemented on Sapelo2 during this maintenance window.

	===Replaced queueing system===		===Replaced queueing system===

2020-10-29T21:06:36Z

← Older revision		Revision as of 17:06, 29 October 2020
Line 8:		Line 8:
	===Changes in the maximum walltime limit of the partitions (queues)===		===Changes in the maximum walltime limit of the partitions (queues)===

	The batch, highmem_p, and gpu_p partitions now have a 7-day maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].		The batch, highmem_p, and gpu_p partitions now have a 7-day maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p partitions allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].

	===Sap2test not available anymore===		===Sap2test not available anymore===

2020-10-29T20:25:18Z

← Older revision		Revision as of 16:25, 29 October 2020
Line 2:		Line 2:


	===~~Changes in the~~ queueing system===		===Replaced queueing system===

	Sapelo2 runs the Slurm queueing system now.		Sapelo2 runs the Slurm queueing system now.

2020-10-29T19:07:51Z

← Older revision		Revision as of 15:07, 29 October 2020
Line 1:		Line 1:



Line 8:		Line 9:

	The batch, highmem_p, and gpu_p partitions now have a 7-day maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].		The batch, highmem_p, and gpu_p partitions now have a 7-day maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].

			===Sap2test not available anymore===

			When you connect to Sapelo2 using the hostname sapelo2.gacrc.uga.edu, you will be connected to the same environment as Sap2test, before the maintenance window. The sap2test.gacrc.uga.edu hostname is not active anymore.

	===Error connecting to Sapelo2===		===Error connecting to Sapelo2===

2020-10-29T19:05:24Z

Changes in the maximum walltime limit of the partitions (queues)

← Older revision		Revision as of 15:05, 29 October 2020
Line 7:		Line 7:
	===Changes in the maximum walltime limit of the partitions (queues)===		===Changes in the maximum walltime limit of the partitions (queues)===

	The batch, highmem_p, and gpu_p partitions now have a 7 ~~days~~ maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].		The batch, highmem_p, and gpu_p partitions now have a 7-day maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].


	===Error connecting to Sapelo2===		===Error connecting to Sapelo2===

2020-10-29T19:04:51Z

Changes in the queueing system

← Older revision		Revision as of 15:04, 29 October 2020
Line 3:		Line 3:
	===Changes in the queueing system===		===Changes in the queueing system===

	Sapelo2 ~~run~~ the Slurm queueing system now.		Sapelo2 runs the Slurm queueing system now.

	===Changes in the maximum walltime limit of the partitions (queues)===		===Changes in the maximum walltime limit of the partitions (queues)===

2020-10-29T19:02:29Z

← Older revision		Revision as of 15:02, 29 October 2020
Line 1:		Line 1:


			===Changes in the queueing system===

			Sapelo2 run the Slurm queueing system now.

			===Changes in the maximum walltime limit of the partitions (queues)===

			The batch, highmem_p, and gpu_p partitions now have a 7 days maximum walltime limit. New batch_30d, highmem_30d_p, and gpu_30d_p allow up to 30 days of walltime limit, but these queues have a more limited availability. Please see [[Running Jobs on Sapelo2]].


	===Error connecting to Sapelo2===		===Error connecting to Sapelo2===