cmor
cmor copied to clipboard
PCMDI
Exposing latest netcdf 4.9.x library functionality: quantize, zstandard
The latest versions of libnetcdf include new functions to further squash data using lossy compression, see Charlie Zender: Why & How to Increase Dataset Compression in CMIP7 - in particular the quantize and zstandard operations.
How easy is it to expose this in CMOR 3.9?
ping @taylor13 @matthew-mizielinski @sashakames @piotrflorek @czender
Also see discussion https://github.com/PCMDI/cmor/discussions/724
There is ongoing discussion on how Charlie Z's approach might get standardized under CF. See https://github.com/cf-convention/cf-conventions/issues/403. Not sure if the above is advocating that or not.
I was able to add nc_def_var_quantize to the code without issue (placed above lines that call nc_def_var_deflate). However, I'm encountering an issue when using nc_def_var_zstandard. Some of the NetCDF files produced by the CMOR tests are giving the following error when you try to read them with ncdump.
$ncdump data.nc
...
NetCDF: HDF error
Location: file ?; fcn ? line 478
@mauzey1 great to see progress here! Do you need to have an updated version of ncdump that can deal with the *zstandard "compressed" data? I am unfamiliar with all these new functions, so not sure I'm going to be all that helpful. Charlie Zender (@czender) might be able to guide the new feature use somewhat, or point us at some docs that guide best practice use
Yes, I think Charlie could save us much time if he can respond. He's always been very helpful in the past. If we don't hear from him in say a week, let's discuss further.
Hi All. @mauzey1 the ncdump symptom you report above could indeed be due to using an ncdump that is not linked to the zstandard library, as suggested by @durack1. First, please use the nc-config tool built with the same netCDF installation that created your ncdump and post the results of nc-config --all. Second, please post the results of ncdump -h -s foo.nc where foo.nc is the problematic dataset.
@czender
Below is the output of nc-config --all
$ nc-config --all
This netCDF 4.9.2 has been built with the following features:
--cc -> x86_64-apple-darwin13.4.0-clang
--cflags -> -I/Users/mauzey1/opt/anaconda3/envs/cmor_dev/include
--libs -> -L/Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib -lnetcdf
--static -> -lmfhdf -ldf -lhdf5_hl -lhdf5 -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lzip -lblosc -lzstd -lbz2 -lxml2
--has-dap -> yes
--has-dap2 -> yes
--has-dap4 -> yes
--has-nc2 -> yes
--has-nc4 -> yes
--has-hdf5 -> yes
--has-hdf4 -> yes
--has-logging -> no
--has-pnetcdf -> no
--has-szlib -> no
--has-cdf5 -> yes
--has-parallel4 -> no
--has-parallel -> no
--has-nczarr -> yes
--has-zstd -> yes
--has-benchmarks -> no
--has-multifilters -> no
--has-stdfilters -> deflate szip blosc zstd bzip2
--has-quantize -> no
--prefix -> /Users/mauzey1/opt/anaconda3/envs/cmor_dev
--includedir -> /Users/mauzey1/opt/anaconda3/envs/cmor_dev/include
--libdir -> /Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib
--plugindir -> /Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib/hdf5/plugin
--version -> netCDF 4.9.2
Below is the output of ncdump -h -s on a file with this problem.
$ ncdump -h -s CMIP6/CMIP6/ISMIP6/PCMDI/PCMDI-test-1-0/piControl-withism/r3i1p1f1/Amon/ta/gn/v20240610/ta_Amon_PCMDI-test-1-0_piControl-withism_r3i1p1f1_gn_200801-200806.nc
netcdf ta_Amon_PCMDI-test-1-0_piControl-withism_r3i1p1f1_gn_200801-200806 {
dimensions:
time = UNLIMITED ; // (6 currently)
plev = 19 ;
lat = 90 ;
lon = 180 ;
bnds = 2 ;
variables:
double time(time) ;
time:bounds = "time_bnds" ;
time:units = "days since 2008" ;
time:calendar = "360_day" ;
time:axis = "T" ;
time:long_name = "time" ;
time:standard_name = "time" ;
time:_Storage = "chunked" ;
time:_ChunkSizes = 512 ;
time:_Endianness = "little" ;
double time_bnds(time, bnds) ;
time_bnds:_Storage = "chunked" ;
time_bnds:_ChunkSizes = 1, 2 ;
time_bnds:_DeflateLevel = 1 ;
time_bnds:_Filter = "32015,0" ;
time_bnds:_Endianness = "little" ;
double plev(plev) ;
plev:units = "Pa" ;
plev:axis = "Z" ;
plev:positive = "down" ;
plev:long_name = "pressure" ;
plev:standard_name = "air_pressure" ;
plev:_Storage = "contiguous" ;
plev:_Endianness = "little" ;
double lat(lat) ;
lat:bounds = "lat_bnds" ;
lat:units = "degrees_north" ;
lat:axis = "Y" ;
lat:long_name = "Latitude" ;
lat:standard_name = "latitude" ;
lat:_Storage = "contiguous" ;
lat:_Endianness = "little" ;
double lat_bnds(lat, bnds) ;
lat_bnds:_Storage = "chunked" ;
lat_bnds:_ChunkSizes = 90, 2 ;
lat_bnds:_DeflateLevel = 1 ;
lat_bnds:_Filter = "32015,0" ;
lat_bnds:_Endianness = "little" ;
double lon(lon) ;
lon:bounds = "lon_bnds" ;
lon:units = "degrees_east" ;
lon:axis = "X" ;
lon:long_name = "Longitude" ;
lon:standard_name = "longitude" ;
lon:_Storage = "contiguous" ;
lon:_Endianness = "little" ;
double lon_bnds(lon, bnds) ;
lon_bnds:_Storage = "chunked" ;
lon_bnds:_ChunkSizes = 180, 2 ;
lon_bnds:_DeflateLevel = 1 ;
lon_bnds:_Filter = "32015,0" ;
lon_bnds:_Endianness = "little" ;
float ta(time, plev, lat, lon) ;
ta:standard_name = "air_temperature" ;
ta:long_name = "Air Temperature" ;
ta:comment = "Air Temperature" ;
ta:units = "K" ;
ta:cell_methods = "time: mean" ;
ta:cell_measures = "area: areacella" ;
ta:missing_value = 1.e+20f ;
ta:_FillValue = 1.e+20f ;
ta:history = "2024-06-11T01:26:28Z altered by CMOR: Converted type from \'d\' to \'f\'." ;
ta:_Storage = "chunked" ;
ta:_ChunkSizes = 1, 19, 90, 180 ;
ta:_DeflateLevel = 1 ;
ta:_Filter = "32015,0" ;
ta:_Endianness = "little" ;
// global attributes:
:Conventions = "CF-1.7 CMIP-6.2" ;
:activity_id = "ISMIP6" ;
:branch_method = "no parent" ;
:branch_time_in_child = 59400. ;
:branch_time_in_parent = 0. ;
:contact = "Python Coder ([email protected])" ;
:creation_date = "2024-06-11T01:26:28Z" ;
:data_specs_version = "01.00.33" ;
:experiment = "preindustrial control with interactive ice sheet" ;
:experiment_id = "piControl-withism" ;
:external_variables = "areacella" ;
:forcing_index = 1 ;
:frequency = "mon" ;
:further_info_url = "https://furtherinfo.es-doc.org/CMIP6.PCMDI.PCMDI-test-1-0.piControl-withism.none.r3i1p1f1" ;
:grid = "native atmosphere regular grid (3x4 latxlon)" ;
:grid_label = "gn" ;
:history = "2024-06-11T01:26:28Z ;rewrote data to be consistent with ISMIP6 for variable ta found in table Amon.;\n",
"Output from archivcl_A1.nce/giccm_03_std_2xCO2_2256." ;
:initialization_index = 1 ;
:institution = "Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA" ;
:institution_id = "PCMDI" ;
:mip_era = "CMIP6" ;
:nominal_resolution = "10000 km" ;
:parent_activity_id = "no parent" ;
:parent_experiment_id = "no parent" ;
:parent_mip_era = "no parent" ;
:parent_source_id = "no parent" ;
:parent_time_units = "no parent" ;
:parent_variant_label = "no parent" ;
:physics_index = 1 ;
:product = "model-output" ;
:realization_index = 3 ;
:realm = "atmos" ;
:references = "Model described by Koder and Tolkien (J. Geophys. Res., 2001, 576-591). Also see http://www.GICC.su/giccm/doc/index.html. The ssp245 simulation is described in Dorkey et al. \'(Clim. Dyn., 2003, 323-357.)\'" ;
:run_variant = "3rd realization" ;
:source = "PCMDI-test 1.0 (1989): \n",
"aerosol: none\n",
"atmos: Earth1.0-gettingHotter (360 x 180 longitude/latitude; 50 levels; top level 0.1 mb)\n",
"atmosChem: none\n",
"land: Earth1.0\n",
"landIce: none\n",
"ocean: BlueMarble1.0-warming (360 x 180 longitude/latitude; 50 levels; top grid cell 0-10 m)\n",
"ocnBgchem: none\n",
"seaIce: Declining1.0-warming (360 x 180 longitude/latitude)" ;
:source_id = "PCMDI-test-1-0" ;
:source_type = "AOGCM ISM AER" ;
:sub_experiment = "none" ;
:sub_experiment_id = "none" ;
:table_id = "Amon" ;
:table_info = "Creation Date:(18 November 2020) MD5:67956a9cc0ef05fb4b373ee8dcc6b433" ;
:title = "PCMDI-test-1-0 output prepared for CMIP6" ;
:tracking_id = "hdl:21.14100/072ecd15-09e2-4157-914e-59a711add511" ;
:variable_id = "ta" ;
:variant_label = "r3i1p1f1" ;
:license = "CMIP6 model data produced by Lawrence Livermore PCMDI is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Consult https://pcmdi.llnl.gov/CMIP6/TermsOfUse for terms of use governing CMIP6 output, including citation requirements and proper acknowledgment. Further information about this data, including some limitations, can be found via the further_info_url (recorded as a global attribute in this file) and at https:///pcmdi.llnl.gov/. The data producers and data providers make no warranty, either express or implied, including, but not limited to, warranties of merchantability and fitness for a particular purpose. All liabilities arising from the supply of the information (including any liability arising in negligence) are excluded to the fullest extent permitted by law." ;
:cmor_version = "3.8.0" ;
:_NCProperties = "version=2,netcdf=4.9.2,hdf5=1.14.3" ;
:_SuperblockVersion = 2 ;
:_IsNetcdf4 = 0 ;
:_Format = "netCDF-4 classic model" ;
}
That all looks nominal. Last requirements to check are contents of plugindir whether ncdump client actually searches there. Please post results of 1. echo ${HDF5_PLUGIN_PATH} and 2. ls -l /Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib/hdf5/plugin
The environment variable HDF5_PLUGIN_PATH is blank. Below is the HDF5 plugin directory.
ls -l /Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib/hdf5/plugin
total 536
-rwxrwxr-x 7 mauzey1 28918 26128 Dec 10 2023 lib__nch5blosc.dylib
-rwxrwxr-x 7 mauzey1 28918 25672 Dec 10 2023 lib__nch5bzip2.dylib
-rwxrwxr-x 7 mauzey1 28918 25496 Dec 10 2023 lib__nch5deflate.dylib
-rwxrwxr-x 7 mauzey1 28918 25480 Dec 10 2023 lib__nch5fletcher32.dylib
-rwxrwxr-x 7 mauzey1 28918 25416 Dec 10 2023 lib__nch5shuffle.dylib
-rwxrwxr-x 7 mauzey1 28918 25976 Dec 10 2023 lib__nch5szip.dylib
-rwxrwxr-x 7 mauzey1 28918 25576 Dec 10 2023 lib__nch5zstd.dylib
-rwxrwxr-x 7 mauzey1 28918 35040 Dec 10 2023 lib__nczhdf5filters.dylib
-rwxrwxr-x 7 mauzey1 28918 34864 Dec 10 2023 lib__nczstdfilters.dylib
@durack1 Good catch. I was focused on the plugin libraries for compression/decompression because that is usually the issue. I suppose the reported lack of quantization support could also mess things up. However, I'm a bit skeptical that the quantization is the issue because that should impact netCDF creation, not reading. (Quantized data are IEEE format so no special software is needed to read). And I thought @mauzey1 was just demonstrating that Zstandard was not working. @mauzey1 , is your test dataset also supposedly quantized? None of the variables' metadata includes the quantization attribute (e.g., _QuantizeBitRoundNumberOfSignificantBits = 9). I suppose it's possible there are two separate issues...
However, back to the plugin directory contents. @mauzey1 please retry your ncdump command after first setting export HDF5_PLUGIN_PATH=/Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib/hdf5/plugin.
@czender I did try setting export HDF5_PLUGIN_PATH=/Users/mauzey1/opt/anaconda3/envs/cmor_dev/lib/hdf5/plugin but ncdump still gave me the same error.
As for the quantized issue, I haven't seen any errors arise from it. I'm currently just debugging errors caused by zstandard. The issue first came up in a test that was reading a netcdf file to append more data to it. The error message contained the same message that I saw in the ncdump output.
C Traceback:
! In function: cmor_validateFilename
!
!!!!!!!!!!!!!!!!!!!!!!!!!
!
! Error: NetCDF Error (-101: NetCDF: HDF error) creating file: CMIP6/CMIP6/ISMIP6/PCMDI/PCMDI-test-1-0/piControl-withism/r3i1p1f1/Amon/ta/gn/v20240611/ta_Amon_PCMDI-test-1-0_piControl-withism_r3i1p1f1_gnUjtBI821241.nc
!
!!!!!!!!!!!!!!!!!!!!!!!!!
I edited the test to only run the part before the "appending" portion to get the file it was appending to. That's where I saw the error in the ncdump output.
@czender How do you use NetCDF with zstandard in your code? Did you needed to pass certain flags to the compiler? I tried rebuilding CMOR with the HDF5_PLUGIN_PATH variable defined but still got the error.
How do you install NetCDF and other libraries? I'm using an Anaconda environment on MacOS. All of my packages are from conda-forge.
Here's the branch of CMOR with the quantize/zstandard changes: https://github.com/PCMDI/cmor/tree/725_expose_netcdf_quantize_and_zstandard
@mauzey1 FWIW my Conda-forge installation of netCDF 4.9.2 also reports
--has-quantize -> no, and it correctly dumps Zstandard-compressed (and quantized) datasets.
The library names in my installation appear identical to yours, yet the dates and sizes differ. Mine are newer and larger than yours:
zender@spectral:~/anaconda/bin$ ls -l /Users/zender/anaconda/lib/hdf5/plugin
total 612
-rwxr-xr-x 2 zender staff 68928 Jun 6 04:03 lib__nch5blosc.dylib*
-rwxr-xr-x 2 zender staff 68480 Jun 6 04:03 lib__nch5bzip2.dylib*
-rwxr-xr-x 2 zender staff 68256 Jun 6 04:03 lib__nch5deflate.dylib*
-rwxr-xr-x 2 zender staff 68272 Jun 6 04:03 lib__nch5fletcher32.dylib*
-rwxr-xr-x 2 zender staff 68224 Jun 6 04:03 lib__nch5shuffle.dylib*
-rwxr-xr-x 2 zender staff 68784 Jun 6 04:03 lib__nch5szip.dylib*
-rwxr-xr-x 2 zender staff 68368 Jun 6 04:03 lib__nch5zstd.dylib*
-rwxr-xr-x 2 zender staff 69632 Jun 6 04:03 lib__nczhdf5filters.dylib*
-rwxr-xr-x 2 zender staff 69440 Jun 6 04:03 lib__nczstdfilters.dylib*
Any idea why this would be the case? Are you using conda-forge for all packages (otherwise library conflicts can easily arise)? In any case, please report results of the ncdump command after executing conda update libnetcdf.
I tried creating a new environment to reinstall libraries hoping it would upgrade everything. However, I got the same error as before. It appears that I've gotten the same version of the HDF5 plugins. I wonder if it's due to dependencies in my environment setup that are giving me older versions of libraries.
@czender If you tried creating the environments described in our source build instructions, do you get versions of the plugins different from your current environment?
@mauzey1 Our messages crossed in the ether. To expand on the above, the ncdump from my conda-forge installations (on both MacOS and Linux machines) correctly dump Zstandard-compressed, quantized files. I am trying to emulate your environment. I did not mention that my typical workflow/environment on MacOS is to use a locally compiled recent daily snapshot of libnetcdf from the Unidata repository. My default compilers and other libraries on MacOS are all from Homebrew. And that setup also works fine. I do use a conda-forge based netCDF on acme1.llnl.gov. Everything works fine with that, and those libraries do appear to match yours in date, name, and (roughly) size:
zender1@acme1:~$ !513
ls -l /home/zender1/anaconda/lib/hdf5/plugin
total 164
-rwxr-xr-x. 2 zender1 climate 16392 Dec 10 2023 lib__nch5blosc.so
-rwxr-xr-x. 2 zender1 climate 16024 Dec 10 2023 lib__nch5bzip2.so
-rwxr-xr-x. 2 zender1 climate 15912 Dec 10 2023 lib__nch5deflate.so
-rwxr-xr-x. 2 zender1 climate 16064 Dec 10 2023 lib__nch5fletcher32.so
-rwxr-xr-x. 2 zender1 climate 15896 Dec 10 2023 lib__nch5shuffle.so
-rwxr-xr-x. 2 zender1 climate 16248 Dec 10 2023 lib__nch5szip.so
-rwxr-xr-x. 2 zender1 climate 16000 Dec 10 2023 lib__nch5zstd.so
-rwxr-xr-x. 2 zender1 climate 21784 Dec 10 2023 lib__nczhdf5filters.so
-rwxr-xr-x. 2 zender1 climate 21520 Dec 10 2023 lib__nczstdfilters.so
Here is an example of that conda-forge netCDF dumping a Zstandard-compressed, quantized file:
zender1@acme1:~$ which ncdump
~/anaconda/bin/ncdump
zender1@acme1:~$ ncdump -h -s ~/foo1.nc
netcdf foo1 {
dimensions:
time = UNLIMITED ; // (10 currently)
lat = 2 ;
lon = 4 ;
variables:
float ps(time, lat, lon) ;
ps:_QuantizeBitRoundNumberOfSignificantBits = 9 ;
ps:lossy_compression = "compression_info" ;
ps:lossy_compression_nsb = 9 ;
ps:lossy_compression_maximum_relative_error = 0.0009765625f ;
ps:standard_name = "surface_air_pressure" ;
ps:units = "Pa" ;
ps:_Storage = "chunked" ;
ps:_ChunkSizes = 1145, 2, 4 ;
ps:_Shuffle = "true" ;
ps:_Filter = "32015,0" ;
ps:_Endianness = "little" ;
char compression_info ;
compression_info:family = "quantize" ;
compression_info:algorithm = "bitround" ;
compression_info:implementation = "libnetcdf version 4.9.2" ;
compression_info:_Storage = "contiguous" ;
float ts(time) ;
ts:_QuantizeBitRoundNumberOfSignificantBits = 9 ;
ts:lossy_compression = "compression_info" ;
ts:lossy_compression_nsb = 9 ;
ts:lossy_compression_maximum_relative_error = 0.0009765625f ;
ts:standard_name = "surface_temperature" ;
ts:units = "K" ;
ts:_Storage = "chunked" ;
ts:_ChunkSizes = 1024 ;
ts:_Shuffle = "true" ;
ts:_Filter = "32015,0" ;
ts:_Endianness = "little" ;
// global attributes:
:Conventions = "CF-1.5" ;
:history = "Tue Jun 11 10:46:26 2024: ncks -O -7 -C -v ps,ts --cmp=btr|shf|zst,0 /home/zender1/nco/data/in.nc /home/zender1/foo1.nc\n",
...
}
So the issue you have encountered is perplexing. Any ideas?
@czender Could you try building and running the tests of the CMOR branch I posted? Ideally using the environment described in the source build instructions.
Alternatively, could you send me a Zstandard-compressed file that I could try opening with my ncdump setup?
@mauzey1 It would be easier for me if you would first try (and add to your CMOR instructions) this first:
conda install blosc bzip2 zstd and report whether you then achieve success with:
ncdump -s -h ~zender1/foo1.nc
That file works for me without issue. Using my current dev environment, I got the following output.
$ ncdump -s -h foo1.nc
netcdf foo1 {
dimensions:
time = UNLIMITED ; // (10 currently)
lat = 2 ;
lon = 4 ;
variables:
float ps(time, lat, lon) ;
ps:_QuantizeBitRoundNumberOfSignificantBits = 9 ;
ps:lossy_compression = "compression_info" ;
ps:lossy_compression_nsb = 9 ;
ps:lossy_compression_maximum_relative_error = 0.0009765625f ;
ps:standard_name = "surface_air_pressure" ;
ps:units = "Pa" ;
ps:_Storage = "chunked" ;
ps:_ChunkSizes = 1145, 2, 4 ;
ps:_Shuffle = "true" ;
ps:_Filter = "32015,0" ;
ps:_Endianness = "little" ;
char compression_info ;
compression_info:family = "quantize" ;
compression_info:algorithm = "bitround" ;
compression_info:implementation = "libnetcdf version 4.9.3-development" ;
compression_info:_Storage = "contiguous" ;
float ts(time) ;
ts:_QuantizeBitRoundNumberOfSignificantBits = 9 ;
ts:lossy_compression = "compression_info" ;
ts:lossy_compression_nsb = 9 ;
ts:lossy_compression_maximum_relative_error = 0.0009765625f ;
ts:standard_name = "surface_temperature" ;
ts:units = "K" ;
ts:_Storage = "chunked" ;
ts:_ChunkSizes = 1024 ;
ts:_Shuffle = "true" ;
ts:_Filter = "32015,0" ;
ts:_Endianness = "little" ;
// global attributes:
:Conventions = "CF-1.5" ;
:history = "Tue Jun 11 09:54:36 2024: ncks -O -7 -C -v ps,ts --cmp=btr|shf|zst,0 /Users/zender/nco/data/in.nc /Users/zender/foo1.nc\n",
"History global attribute.\n",
"Textual attributes like history often have embedded newlines like this.\n",
"Such newlines should serve as linebreaks on the screen to enhance legibility like this.\n",
"Friendly CDL converters print a single NC_CHAR attribute as a comma-separated list of strings where each embedded delimiter marks a linebreak. This makes poetry embedded in CDL much nicer to read (except for the printed literal \\n\'s---those are an eyesore):\n",
"\n",
"A POET by Hafiz\n",
"\n",
"A poet is someone\n",
"Who can pour light into a cup,\n",
"Then raise it to nourish\n",
"Your beautiful parched, holy mouth\n",
"" ;
:lorem_ipsum = "The Lorem Ipsum attribute demonstrates the legibility of text without embedded linebreaks:\n",
"Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Lady Gaga amat indueris vestimento laetus. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum." ;
:julian_day = 200000.04 ;
:RCS_Header = "$Header$" ;
:NCO = "netCDF Operators version 5.2.5-alpha02 (Homepage = http://nco.sf.net, Code = http://github.com/nco/nco, Citation = 10.1016/j.envsoft.2008.03.004)" ;
:_NCProperties = "version=2,netcdf=4.9.3-development,hdf5=1.14.3" ;
:_SuperblockVersion = 2 ;
:_IsNetcdf4 = 0 ;
:_Format = "netCDF-4 classic model" ;
}
I'm trying to run conda install blosc bzip2 zstd in my base environment but it is taking a long time. Running it in my 'cmor_dev' environment worked but those libraries are already installed.
@mauzey1 Since you can dump the file I sent but not your original test file then my hunch is now that your original test file was somehow generate in a corrupted state. Do you agree? Put it somewhere on acme1 where I can read it and I'll take a look.
The file appears to be fine. Both ncdump and ncks can read it fine on my MacOS and Linux machines, with various netCDF libraries, including conda-forge. I think that narrows down the potential causes to the configuration of your reading environment. The writing appears to work fine. Umm...the only thing that comes to mind is a possible mis-configuration of LD_LIBRARY_PATH which, if it exists at all, should point to the a netCDF library with correctly linked codecs, such as the conda-forge netCDF library that wrote the file.
I tried reading the same file on a Linux machine with a conda environment made withconda create -n ncdf_env -c conda-forge netcdf4 hdf5 blosc bzip2 zstd and got the following output.
-bash-4.2$ ls -lh ~/anaconda3/envs/ncdf_env/lib/hdf5/plugin/
total 164K
-rwxr-x---. 2 mauzey1 climate 17K Jun 6 04:00 lib__nch5blosc.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5bzip2.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5deflate.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5fletcher32.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5shuffle.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5szip.so
-rwxr-x---. 2 mauzey1 climate 16K Jun 6 04:00 lib__nch5zstd.so
-rwxr-x---. 2 mauzey1 climate 22K Jun 6 04:00 lib__nczhdf5filters.so
-rwxr-x---. 2 mauzey1 climate 22K Jun 6 04:00 lib__nczstdfilters.so
-bash-4.2$ pwd
/home/mauzey1
-bash-4.2$ export HDF5_PLUGIN_PATH=/home/mauzey1/anaconda3/envs/ncdf_env/lib/hdf5/plugin/
-bash-4.2$ ncdump bad_file.nc
netcdf bad_file {
dimensions:
time = UNLIMITED ; // (6 currently)
plev = 19 ;
lat = 90 ;
lon = 180 ;
bnds = 2 ;
variables:
double time(time) ;
time:bounds = "time_bnds" ;
time:units = "days since 2008" ;
time:calendar = "360_day" ;
time:axis = "T" ;
time:long_name = "time" ;
time:standard_name = "time" ;
double time_bnds(time, bnds) ;
double plev(plev) ;
plev:units = "Pa" ;
plev:axis = "Z" ;
plev:positive = "down" ;
plev:long_name = "pressure" ;
plev:standard_name = "air_pressure" ;
double lat(lat) ;
lat:bounds = "lat_bnds" ;
lat:units = "degrees_north" ;
lat:axis = "Y" ;
lat:long_name = "Latitude" ;
lat:standard_name = "latitude" ;
double lat_bnds(lat, bnds) ;
double lon(lon) ;
lon:bounds = "lon_bnds" ;
lon:units = "degrees_east" ;
lon:axis = "X" ;
lon:long_name = "Longitude" ;
lon:standard_name = "longitude" ;
double lon_bnds(lon, bnds) ;
float ta(time, plev, lat, lon) ;
ta:standard_name = "air_temperature" ;
ta:long_name = "Air Temperature" ;
ta:comment = "Air Temperature" ;
ta:units = "K" ;
ta:cell_methods = "time: mean" ;
ta:cell_measures = "area: areacella" ;
ta:missing_value = 1.e+20f ;
ta:_FillValue = 1.e+20f ;
ta:history = "2024-06-11T18:43:28Z altered by CMOR: Converted type from \'d\' to \'f\'." ;
// global attributes:
:Conventions = "CF-1.7 CMIP-6.2" ;
:activity_id = "ISMIP6" ;
:branch_method = "no parent" ;
:branch_time_in_child = 59400. ;
:branch_time_in_parent = 0. ;
:contact = "Python Coder ([email protected])" ;
:creation_date = "2024-06-11T18:43:28Z" ;
:data_specs_version = "01.00.33" ;
:experiment = "preindustrial control with interactive ice sheet" ;
:experiment_id = "piControl-withism" ;
:external_variables = "areacella" ;
:forcing_index = 1 ;
:frequency = "mon" ;
:further_info_url = "https://furtherinfo.es-doc.org/CMIP6.PCMDI.PCMDI-test-1-0.piControl-withism.none.r3i1p1f1" ;
:grid = "native atmosphere regular grid (3x4 latxlon)" ;
:grid_label = "gn" ;
:history = "2024-06-11T18:43:28Z ;rewrote data to be consistent with ISMIP6 for variable ta found in table Amon.;\n",
"Output from archivcl_A1.nce/giccm_03_std_2xCO2_2256." ;
:initialization_index = 1 ;
:institution = "Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA" ;
:institution_id = "PCMDI" ;
:mip_era = "CMIP6" ;
:nominal_resolution = "10000 km" ;
:parent_activity_id = "no parent" ;
:parent_experiment_id = "no parent" ;
:parent_mip_era = "no parent" ;
:parent_source_id = "no parent" ;
:parent_time_units = "no parent" ;
:parent_variant_label = "no parent" ;
:physics_index = 1 ;
:product = "model-output" ;
:realization_index = 3 ;
:realm = "atmos" ;
:references = "Model described by Koder and Tolkien (J. Geophys. Res., 2001, 576-591). Also see http://www.GICC.su/giccm/doc/index.html. The ssp245 simulation is described in Dorkey et al. \'(Clim. Dyn., 2003, 323-357.)\'" ;
:run_variant = "3rd realization" ;
:source = "PCMDI-test 1.0 (1989): \n",
"aerosol: none\n",
"atmos: Earth1.0-gettingHotter (360 x 180 longitude/latitude; 50 levels; top level 0.1 mb)\n",
"atmosChem: none\n",
"land: Earth1.0\n",
"landIce: none\n",
"ocean: BlueMarble1.0-warming (360 x 180 longitude/latitude; 50 levels; top grid cell 0-10 m)\n",
"ocnBgchem: none\n",
"seaIce: Declining1.0-warming (360 x 180 longitude/latitude)" ;
:source_id = "PCMDI-test-1-0" ;
:source_type = "AOGCM ISM AER" ;
:sub_experiment = "none" ;
:sub_experiment_id = "none" ;
:table_id = "Amon" ;
:table_info = "Creation Date:(18 November 2020) MD5:67956a9cc0ef05fb4b373ee8dcc6b433" ;
:title = "PCMDI-test-1-0 output prepared for CMIP6" ;
:tracking_id = "hdl:21.14100/665a2970-e412-4ddf-8e43-2b7ee12b577d" ;
:variable_id = "ta" ;
:variant_label = "r3i1p1f1" ;
:license = "CMIP6 model data produced by Lawrence Livermore PCMDI is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Consult https://pcmdi.llnl.gov/CMIP6/TermsOfUse for terms of use governing CMIP6 output, including citation requirements and proper acknowledgment. Further information about this data, including some limitations, can be found via the further_info_url (recorded as a global attribute in this file) and at https:///pcmdi.llnl.gov/. The data producers and data providers make no warranty, either express or implied, including, but not limited to, warranties of merchantability and fitness for a particular purpose. All liabilities arising from the supply of the information (including any liability arising in negligence) are excluded to the fullest extent permitted by law." ;
:cmor_version = "3.8.0" ;
data:
time = 15, 45, 75, 105, 135, 165 ;
time_bnds =
0, 30,
NetCDF: HDF error
Location: file ?; fcn ? line 478
Output of nc-config --all
-bash-4.2$ nc-config --all
This netCDF 4.9.2 has been built with the following features:
--cc -> x86_64-conda-linux-gnu-cc
--cflags -> -I/home/mauzey1/anaconda3/envs/ncdf_env/include
--libs -> -L/home/mauzey1/anaconda3/envs/ncdf_env/lib -lnetcdf
--static -> -lmfhdf -ldf -lhdf5_hl -lhdf5 -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lzip -lblosc -lzstd -lbz2 -lxml2
--has-dap -> yes
--has-dap2 -> yes
--has-dap4 -> yes
--has-nc2 -> yes
--has-nc4 -> yes
--has-hdf5 -> yes
--has-hdf4 -> yes
--has-logging -> no
--has-pnetcdf -> no
--has-szlib -> no
--has-cdf5 -> yes
--has-parallel4 -> no
--has-parallel -> no
--has-nczarr -> yes
--has-zstd -> yes
--has-benchmarks -> no
--has-multifilters -> no
--has-stdfilters -> deflate szip blosc zstd bzip2
--has-quantize -> no
--prefix -> /home/mauzey1/anaconda3/envs/ncdf_env
--includedir -> /home/mauzey1/anaconda3/envs/ncdf_env/include
--libdir -> /home/mauzey1/anaconda3/envs/ncdf_env/lib
--plugindir -> /home/mauzey1/anaconda3/envs/ncdf_env/lib/hdf5/plugin
--version -> netCDF 4.9.2
Packages installed in this environment:
-bash-4.2$ conda list
# packages in environment at /home/mauzey1/anaconda3/envs/ncdf_env:
#
# Name Version Build Channel
_libgcc_mutex 0.1 conda_forge conda-forge
_openmp_mutex 4.5 2_gnu conda-forge
blosc 1.21.5 hc2324a3_1 conda-forge
bzip2 1.0.8 hd590300_5 conda-forge
c-ares 1.28.1 hd590300_0 conda-forge
ca-certificates 2024.6.2 hbcca054_0 conda-forge
certifi 2024.6.2 pyhd8ed1ab_0 conda-forge
cftime 1.6.4 py312h085067d_0 conda-forge
hdf4 4.2.15 h2a13503_7 conda-forge
hdf5 1.14.3 nompi_hdf9ad27_105 conda-forge
icu 73.2 h59595ed_0 conda-forge
keyutils 1.6.1 h166bdaf_0 conda-forge
krb5 1.21.2 h659d440_0 conda-forge
ld_impl_linux-64 2.40 hf3520f5_3 conda-forge
libaec 1.1.3 h59595ed_0 conda-forge
libblas 3.9.0 22_linux64_openblas conda-forge
libcblas 3.9.0 22_linux64_openblas conda-forge
libcurl 8.8.0 hca28451_0 conda-forge
libedit 3.1.20191231 he28a2e2_2 conda-forge
libev 4.33 hd590300_2 conda-forge
libexpat 2.6.2 h59595ed_0 conda-forge
libffi 3.4.2 h7f98852_5 conda-forge
libgcc-ng 13.2.0 h77fa898_7 conda-forge
libgfortran-ng 13.2.0 h69a702a_7 conda-forge
libgfortran5 13.2.0 hca663fb_7 conda-forge
libgomp 13.2.0 h77fa898_7 conda-forge
libiconv 1.17 hd590300_2 conda-forge
libjpeg-turbo 3.0.0 hd590300_1 conda-forge
liblapack 3.9.0 22_linux64_openblas conda-forge
libnetcdf 4.9.2 nompi_h135f659_114 conda-forge
libnghttp2 1.58.0 h47da74e_1 conda-forge
libnsl 2.0.1 hd590300_0 conda-forge
libopenblas 0.3.27 pthreads_h413a1c8_0 conda-forge
libsqlite 3.46.0 hde9e2c9_0 conda-forge
libssh2 1.11.0 h0841786_0 conda-forge
libstdcxx-ng 13.2.0 hc0a3c3a_7 conda-forge
libuuid 2.38.1 h0b41bf4_0 conda-forge
libxcrypt 4.4.36 hd590300_1 conda-forge
libxml2 2.12.7 hc051c1a_1 conda-forge
libzip 1.10.1 h2629f0a_3 conda-forge
libzlib 1.3.1 h4ab18f5_1 conda-forge
lz4-c 1.9.4 hcb278e6_0 conda-forge
ncurses 6.5 h59595ed_0 conda-forge
netcdf4 1.6.5 nompi_py312h39d4375_102 conda-forge
numpy 1.26.4 py312heda63a1_0 conda-forge
openssl 3.3.1 h4ab18f5_0 conda-forge
pip 24.0 pyhd8ed1ab_0 conda-forge
python 3.12.3 hab00c5b_0_cpython conda-forge
python_abi 3.12 4_cp312 conda-forge
readline 8.2 h8228510_1 conda-forge
setuptools 70.0.0 pyhd8ed1ab_0 conda-forge
snappy 1.2.0 hdb0a2a9_1 conda-forge
tk 8.6.13 noxft_h4845f30_101 conda-forge
tzdata 2024a h0c530f3_0 conda-forge
wheel 0.43.0 pyhd8ed1ab_1 conda-forge
xz 5.2.6 h166bdaf_0 conda-forge
zlib 1.3.1 h4ab18f5_1 conda-forge
zstd 1.5.6 ha6fb4c9_0 conda-forge
All of our CI jobs, running on both Linux and Mac, are failing on the same test that produced that file when testing the zstandard/quantize branch.
Here is said test: https://github.com/PCMDI/cmor/blob/725_expose_netcdf_quantize_and_zstandard/Test/test_python_appending.py
@mauzey1 My bad. You are not crazy :) I had only been dumping metadata, under the mistaken impression that the issue would show itself printing metadata. I can reproduce the error you encounter when printing the full data with ncdump:
zender@spectral:~$ ~/anaconda/bin/ncdump bad_file.nc
...
time_bnds =
0, 30,
NetCDF: HDF error
Location: file ?; fcn ? line 478
Using the latest netCDF snapshot the error comes from a different source line:
zender@spectral:~$ ncdump bad_file.nc
...
time_bnds =
0, 30,
NetCDF: HDF error
Location: file vardata.c; fcn print_rows line 483
NCO's ncks gives a more informative error message:
zender@spectral:~$ ncks bad_file.nc
...
time = 15, 45, 75, 105, 135, 165 ;
ERROR: nco_get_vara() failed to nc_get_vara() variable "time_bnds"
ERROR NC_EHDFERR Error at HDF5 layer
HINT: NC_EHDFERR errors indicate that the HDF5-backend to netCDF is unable to perform the requested task. NCO can receive this devilishly inscrutable error for a variety of possible reasons including: 1) The run-time dynamic linker attempts to resolve calls from the netCDF library to the HDF library with an HDF5 libhdf5.a that is incompatible with the version used to build NCO and netCDF. 2) The file system does not allow the HDF5 flock() function, as of HDF5 1.10.x, to enable multiple processes to open the same file for reading, a feature known as SWMR (Single Write Multiple Read). The fix is to disable the HDF5 flock() by setting an environment variable thusly: "export HDF5_USE_FILE_LOCKING=FALSE". 3) An incorrect netCDF4 library implementation of a procedure (e.g., nc_rename_var()) in terms of HDF function calls (e.g., HDF5Lmove()) manifests an error or inconsistent state within the HDF5 layer. This often occurs during renaming operations (https://github.com/Unidata/netcdf-c/issues/597). 4) Attempting to compress or decompress a netCDF4 dataset with a non-standard (i.e., non-DEFLATE) filter when the requisite shared library to encode/decode that compression filter is not present in either the default location (/usr/local/hdf5/lib/plugin) or in the user-configurable location referred to by the HDF5_PLUGIN_PATH environment variable. One can determine if missing plugin libraries are the culprit by dumping the hidden attributes of the dataset with, e.g., ncks --hdn -m in.nc or ncdump -s -h in.nc. Any variables with (hidden) "_Filter" attributes require the corresponding shared libraries to be located in HDF5_PLUGIN_PATH. Some HDF5 implementations (at least MacOSX with MacPorts as of 20200907) may also require explicitly setting the plugin path in the environment, even for the default location! To test this, re-try your NCO command after doing this: "export HDF5_PLUGIN_PATH=/usr/local/hdf5/lib/plugin". 5) Bad vibes.
nco_err_exit(): ERROR Short NCO-generated message (usually name of function that triggered error): nco_get_vara()
nco_err_exit(): ERROR Error code is -101. Translation into English with nc_strerror(-101) is "NetCDF: HDF error"
nco_err_exit(): ERROR NCO will now exit with system call exit(EXIT_FAILURE)
Based on this error message, I looked more carefully at the metadata for time_bnds, which is where the HDF error occurs:
zender@spectral:~$ ncks -D 2 --hdn -m -C -v time_bnds bad_file.nc
netcdf bad_file {
// ncgen -k "netCDF-4 classic model" -b -o bad_file.nc bad_file.cdl
dimensions:
bnds = 2 ; // ID = 4
time = UNLIMITED ; // (6 currently) ID = 0
variables:
double time_bnds(time,bnds) ; // RAM size = 6*2*sizeof(NC_DOUBLE) = 12*8 = 96 bytes, ID = 1
time_bnds:_Storage = "chunked" ; // char
time_bnds:_ChunkSizes = 1, 2 ; // int
time_bnds:_Filter = "32015,3" ; // char codec(s): Zstandard
time_bnds:_DeflateLevel = 1 ; // int
time_bnds:_Endianness = "little" ; // char
} // group /
Next, I verified that ncks can dump the entire file except time_bnds:
zender@spectral:~$ ncks -C -x -v time_bnds bad_file.nc
netcdf bad_file {
...
} // group /
A few points about all this:
First, the HDF error occurs after successfully reading and printing the other compressed variables, including ta. Second, time_bnds is a small variable with only 12 values. Are you sure that all the values written are "normal"? Related to this, the chunksize of time_bnds is only 1*2=2 values per chunk. I wonder if there is a bug in one of the codecs (or their combination) that only reveals itself when chunksizes are super small. Maybe try not compressing variables with chunksizes smaller than a minimum threshold (e.g., 16). Third, I did not notice earlier the strange fact that both DEFLATE and Zstandard codecs have been applied. Why does CMOR use two lossless codecs? It should be possible to do this, yet little if anything is gained in compression by doing so. Moreover, the Shuffle filter has not been activated. Doing that would save much more space than "double compressing" the data. Please try applying only one lossless codec and see if that resolves the HDF error. Again, the netCDF filter code is written to support an arbitrary number of codecs, and I have tested that capability with 3+ lossless codecs in the past, yet it does seem possible that doing so is related to this bug.
Given all this, my hunch is that the small chunksize of time_bnds, not the "double compression" is what triggers the bug that causes the HDF error.
@czender
I tried disabling the use of zstandard compression on the bounds variables and ncdump read the generated file without issue. I also tried disabling deflate and that also made the file get read by ncdump without issue.
I noticed that some files generated by the tests didn't give ncdump issues. I'll look at the other files to see what could be causing the issue.
Thank you for helping debug this issue.
@mauzey1 If it is easy to do, I would be curious to see what the values of the time_bnds are in the case where you disabled use of zstandard compression. Could you print those out? This would help answer the question "Are you sure that all the values written are "normal"? Of course if the same input bounds are used for other tests that are successfully written, this would be unlikely to be the cause of any problems.
@taylor13 Here is the ncdump output with the time bounds displayed.
$ ncdump -v time_bnds CMIP6/CMIP6/ISMIP6/PCMDI/PCMDI-test-1-0/piControl-withism/r3i1p1f1/Amon/ta/gn/v20240611/ta_Amon_PCMDI-test-1-0_piControl-withism_r3i1p1f1_gn_200801-200806.nc
netcdf ta_Amon_PCMDI-test-1-0_piControl-withism_r3i1p1f1_gn_200801-200806 {
dimensions:
time = UNLIMITED ; // (6 currently)
plev = 19 ;
lat = 90 ;
lon = 180 ;
bnds = 2 ;
variables:
double time(time) ;
time:bounds = "time_bnds" ;
time:units = "days since 2008" ;
time:calendar = "360_day" ;
time:axis = "T" ;
time:long_name = "time" ;
time:standard_name = "time" ;
double time_bnds(time, bnds) ;
double plev(plev) ;
plev:units = "Pa" ;
plev:axis = "Z" ;
plev:positive = "down" ;
plev:long_name = "pressure" ;
plev:standard_name = "air_pressure" ;
double lat(lat) ;
lat:bounds = "lat_bnds" ;
lat:units = "degrees_north" ;
lat:axis = "Y" ;
lat:long_name = "Latitude" ;
lat:standard_name = "latitude" ;
double lat_bnds(lat, bnds) ;
double lon(lon) ;
lon:bounds = "lon_bnds" ;
lon:units = "degrees_east" ;
lon:axis = "X" ;
lon:long_name = "Longitude" ;
lon:standard_name = "longitude" ;
double lon_bnds(lon, bnds) ;
float ta(time, plev, lat, lon) ;
ta:standard_name = "air_temperature" ;
ta:long_name = "Air Temperature" ;
ta:comment = "Air Temperature" ;
ta:units = "K" ;
ta:cell_methods = "time: mean" ;
ta:cell_measures = "area: areacella" ;
ta:missing_value = 1.e+20f ;
ta:_FillValue = 1.e+20f ;
ta:history = "2024-06-11T22:31:01Z altered by CMOR: Converted type from \'d\' to \'f\'." ;
// global attributes:
:Conventions = "CF-1.7 CMIP-6.2" ;
:activity_id = "ISMIP6" ;
:branch_method = "no parent" ;
:branch_time_in_child = 59400. ;
:branch_time_in_parent = 0. ;
:contact = "Python Coder ([email protected])" ;
:creation_date = "2024-06-11T22:31:01Z" ;
:data_specs_version = "01.00.33" ;
:experiment = "preindustrial control with interactive ice sheet" ;
:experiment_id = "piControl-withism" ;
:external_variables = "areacella" ;
:forcing_index = 1 ;
:frequency = "mon" ;
:further_info_url = "https://furtherinfo.es-doc.org/CMIP6.PCMDI.PCMDI-test-1-0.piControl-withism.none.r3i1p1f1" ;
:grid = "native atmosphere regular grid (3x4 latxlon)" ;
:grid_label = "gn" ;
:history = "2024-06-11T22:31:01Z ;rewrote data to be consistent with ISMIP6 for variable ta found in table Amon.;\n",
"Output from archivcl_A1.nce/giccm_03_std_2xCO2_2256." ;
:initialization_index = 1 ;
:institution = "Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA" ;
:institution_id = "PCMDI" ;
:mip_era = "CMIP6" ;
:nominal_resolution = "10000 km" ;
:parent_activity_id = "no parent" ;
:parent_experiment_id = "no parent" ;
:parent_mip_era = "no parent" ;
:parent_source_id = "no parent" ;
:parent_time_units = "no parent" ;
:parent_variant_label = "no parent" ;
:physics_index = 1 ;
:product = "model-output" ;
:realization_index = 3 ;
:realm = "atmos" ;
:references = "Model described by Koder and Tolkien (J. Geophys. Res., 2001, 576-591). Also see http://www.GICC.su/giccm/doc/index.html. The ssp245 simulation is described in Dorkey et al. \'(Clim. Dyn., 2003, 323-357.)\'" ;
:run_variant = "3rd realization" ;
:source = "PCMDI-test 1.0 (1989): \n",
"aerosol: none\n",
"atmos: Earth1.0-gettingHotter (360 x 180 longitude/latitude; 50 levels; top level 0.1 mb)\n",
"atmosChem: none\n",
"land: Earth1.0\n",
"landIce: none\n",
"ocean: BlueMarble1.0-warming (360 x 180 longitude/latitude; 50 levels; top grid cell 0-10 m)\n",
"ocnBgchem: none\n",
"seaIce: Declining1.0-warming (360 x 180 longitude/latitude)" ;
:source_id = "PCMDI-test-1-0" ;
:source_type = "AOGCM ISM AER" ;
:sub_experiment = "none" ;
:sub_experiment_id = "none" ;
:table_id = "Amon" ;
:table_info = "Creation Date:(18 November 2020) MD5:67956a9cc0ef05fb4b373ee8dcc6b433" ;
:title = "PCMDI-test-1-0 output prepared for CMIP6" ;
:tracking_id = "hdl:21.14100/2c77d159-cda7-4009-a9e7-3bb813978041" ;
:variable_id = "ta" ;
:variant_label = "r3i1p1f1" ;
:license = "CMIP6 model data produced by Lawrence Livermore PCMDI is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Consult https://pcmdi.llnl.gov/CMIP6/TermsOfUse for terms of use governing CMIP6 output, including citation requirements and proper acknowledgment. Further information about this data, including some limitations, can be found via the further_info_url (recorded as a global attribute in this file) and at https:///pcmdi.llnl.gov/. The data producers and data providers make no warranty, either express or implied, including, but not limited to, warranties of merchantability and fitness for a particular purpose. All liabilities arising from the supply of the information (including any liability arising in negligence) are excluded to the fullest extent permitted by law." ;
:cmor_version = "3.8.0" ;
data:
time_bnds =
0, 30,
30, 60,
60, 90,
90, 120,
120, 150,
150, 180 ;
}
It seems to be due to an interaction of "double compression" (i.e., two lossless codecs) with microscopic chunk sizes.