NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.

To learn about what's going on, see About the Archive Site.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[netCDF #GLM-296638]: netCDF write failure after a certain number of writes



Hello Dominik,

Thank you very much for providing this test file; it will be very useful.  We 
have had intermittent reports of this (or a similar) issue but it has been very 
hard to duplicate.  

We are mostly out of the office this week due to a Holiday, but I will be 
following up on this the best that I can until then.  Please don't worry if you 
haven't heard back from us before next week.

If I might ask, which version of linux are you using, and which MPI libraries 
are yo uusing (openmpi, mpich)?  I'll be testing this in a virtual environment, 
and would like to replicate your system as closely as possible.

Thanks once again, this will be very useful!  Have a good day,

-Ward

> Full Name: Dominik Reitzle
> Email Address: address@hidden
> Organization: ILM (Ulm, Germany)
> Package Version: netCDF 4.2.0 - 4.3.3-rc2
> Operating System: Linux
> Hardware:
> Description of problem: After a certain number of writes, netCDF fails with 
> the following error:
> NetCDF: Error initializing for parallel access
> This happens with put_var1 as well as with put_varm with stride > 1. put_vara 
> and put_varm with stride = 1 work.
> The attached C program reproduces this failure. I reproduced this with netCDF 
> 4.3.2 and 4.3.3-rc2. NetCDF 4.2.0, 4.2.1 and 4.2.1.1 show a similar failure, 
> but instead of returning with an error, the calls do not return at all.
> This seems to be the same problem that was reported here before:
> http://www.unidata.ucar.edu/support/help/MailArchives/netcdf/msg12543.html
> 
> 


Ticket Details
===================
Ticket ID: GLM-296638
Department: Support netCDF
Priority: Normal
Status: Closed