On 8/31/2012 6:58 AM, Dave Love wrote:
In the absence of any knowledge about that cluster, that doesn't confirm that it's reported for the specific hosts that the scheduler complained about, just that it's reported for some. Look explicitly at the load parameters from one of the hosts in question. Is mem_free there? Is anything else missing (see load_parameters(5))?
It is indeed there. Here is one node: $ qhost -F -h compute-3-1 HOSTNAME ARCH NCPU NSOC NCOR NTHR LOAD MEMTOT MEMUSE SWAPTO SWAPUS ---------------------------------------------------------------------------------------------- global - - - - - - - - - - compute-3-1 lx-amd64 8 2 8 8 0.01 7.8G 916.7M 16.6G 13.8M hl:arch=lx-amd64 hl:num_proc=8.000000 hl:mem_total=7.815G hl:swap_total=16.600G hl:virtual_total=24.415G hl:m_topology=SCCCCSCCCC hl:m_socket=2.000000 hl:m_core=8.000000 hl:m_thread=8.000000 hl:load_avg=0.010000 hl:load_short=0.000000 hl:load_medium=0.010000 hl:load_long=0.050000 hl:mem_free=6.920G hl:swap_free=16.587G hl:virtual_free=23.507G hl:mem_used=916.703M hl:swap_used=13.754M hl:virtual_used=930.457M hl:cpu=0.000000 hl:m_topology_inuse=SCCCCSCCCC hl:cores_in_use=0.000000 hl:np_load_avg=0.001250 hl:np_load_short=0.000000 hl:np_load_medium=0.001250 hl:np_load_long=0.006250 $ qrsh -q space1@compute-3-1 -l mem_free=1G error: no suitable queues It apparently does know about "mem_free" because if I request a non existing resource "mem_noexist" it complains: $ qrsh -q space1@compute-3-1 -l mem_noexist=1G unknown resource "mem_noexist" With no resources it works: $ qrsh -q space1@compute-3-1 Last login: Fri Aug 31 17:48:19 2012 from sys.local Rocks Compute Node Rocks 5.4.3 (Viper) [me@compute-3-1 ~]$ _______________________________________________ users mailing list users@gridengine.org https://gridengine.org/mailman/listinfo/users