Hi David, Baker D.J. <[email protected]> writes: > Hi Loris, > > I've pasted the output of "sshare -la" below. It's not too easy to > read as the lines are looped around. Does this make any sense to you? > It is disconcerting that a lot of the values are shown as undefined > (nan). > > Best regards, > David. > > [root@blue30 etc]# sshare -la > Account User RawShares NormShares RawUsage > NormUsage EffectvUsage FairShare GrpTRESMins > TRESRunMins > -------------------- ---------- ---------- ----------- ----------- > ----------- ------------- ---------- ------------------------------ > ------------------------------ > root 1.000000 9223372036854775808 > 1.000000 0.500000 > cpu=15838674,mem=0,energy=0,n+ > root root 1 0.055556 0 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > gpuusers 2 0.111111 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > gpuusers djb1 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > gpuusers hpc 1 0.055556 0 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research 15 0.833333 9223372036854775808 > nan nan nan > cpu=15838674,mem=0,energy=0,n+ > research ab24g12 1 0.055556 875866239 > nan nan nan > cpu=5347208,mem=0,energy=0,no+ > research cica1d14 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research djb1 1 0.055556 9223372036854775808 > nan nan nan > cpu=80,mem=0,energy=0,node=80 > research dpm1u13 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research gtj1y12 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research hpc 1 0.055556 0 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research icw 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research jag1g13 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research jec1f12 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research lmr1u16 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research mb1a10 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research mjp1m12 1 0.055556 9223372036854775808 > nan nan nan > cpu=10491385,mem=0,energy=0,n+ > research ph1m12 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research srw1g10 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0 > research tp1v09 1 0.055556 9223372036854775808 > nan nan nan > cpu=0,mem=0,energy=0,node=0
That looks very odd. The RawUsage values can't be correct and are probably causing the NaNs as calculating the NormUsage is probably failing because the sum of the RawUsages is not a sensible value. The number 9223372036854775808 equals 2**63, i.e. 1 larger than the largest signed 64-bit integer, which looks like some sort of overflow or type mismatch. Unless anyone else has any ideas, I would be tempted to say that your database is borked and you need to start over again. Sorry not be more helpful :-( Loris -- Dr. Loris Bennett (Mr.) ZEDAT, Freie Universität Berlin Email [email protected]
