Re: IO error when running NAMD on new cluster

From: Edward Patrick Obrien (edobrien_at_Glue.umd.edu)
Date: Mon Jun 07 2004 - 09:21:44 CDT

• Next message: Edward Patrick Obrien: "Re: psfgen issue"
• Previous message: Hyonseok Hwang: "question on a chared system MD simulation"
• In reply to: Alex V.: "Re: IO error when running NAMD on new cluster"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Hi Alex,
  I have not tried running with dcd output off. This problem comes up only
occasionally; sometimes the simulation runs to completion, other times
this IO error occurs.

  If all outputs were turned off, I think I would not get this error.

  My guess is that NFS if failing to write remotely to disk, for whatever
reason.

Thanks,
Ed

On Sun, 6 Jun 2004, Alex V. wrote:

> Hi Edward!
> Do you get the same erro when DCD output is dissabled?
>
> Alex
>
> Edward Patrick Obrien wrote:
>
> >Hi All,
> >
> >I get this error after NAMD runs for awhile:
> >
> >
> >
> >>WRITING COORDINATES TO DCD FILE AT STEP 56000
> >>FATAL ERROR: Input/output error
> >>------------- Processor 0 Exiting: Called CmiAbort ------------
> >>Reason: FATAL ERROR: Input/output error
> >>
> >>Stack Traceback:
> >> [0] vfprintf+0x1e6a [0x80ec4f6]
> >> [1] _Z13write_dcdstepiiPfS_S_Pd+0x61 [0x80ebfb7]
> >> [2] _ZN6Output14output_dcdfileEiiP11FloatVectorPK7Lattice+0x575
> >>[0x82002e1]
> >> [3] _ZN6Output10coordinateEiiP6VectorP11FloatVectorR7Lattice+0x86
> >>[0x81ff7ae]
> >> [4]
> >>
> >>
> >>
> >_ZN16CollectionMaster16disposePositionsEPNS_21CollectVectorInstanceE+0x60
> >
> >
> >>[0x80f64d0]
> >> [5] _ZN16CollectionMaster16receivePositionsEP16CollectVectorMsg+0x564
> >>[0x80f56a8]
> >> [6]
> >>
> >>
> >>
> >_ZN24CkIndex_CollectionMaster39_call_receivePositions_CollectVectorMsgEPvP16CollectionMaster+0xe
> >
> >
> >>[0x80f513a]
> >> [7] CkDeliverMessageFree+0x21 [0x827c1f9]
> >> [8] _Z15_processHandlerPvP11CkCoreState+0x3ee [0x827d63e]
> >> [9] CmiHandleMessage+0x1d [0x82b5e25]
> >> [10] _ZN9ScriptTcl7suspendEv+0xc [0x8228c68]
> >> [11] _ZN9ScriptTcl13runControllerEi+0x3f [0x8228c4f]
> >> [12] _ZN9ScriptTcl3runEPc+0xad [0x8228c09]
> >> [13] main+0x1d2 [0x80ed052]
> >> [14] __libc_start_main+0xf0 [0x834750]
> >> [15] sinh+0x65 [0x80eaed1]
> >>req_handle_abort called
> >>Fatal error on PE 0> FATAL ERROR: Input/output error
> >>
> >>
> >
> >It appears it may be a problem with our NFS... any ideas or suggestions,
> >
> >Thanks!
> >Ed
> >
> >
> >
> >
> >
>
>

• Next message: Edward Patrick Obrien: "Re: psfgen issue"
• Previous message: Hyonseok Hwang: "question on a chared system MD simulation"
• In reply to: Alex V.: "Re: IO error when running NAMD on new cluster"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:40 CST