[webservices] defective miniseed

[Philip Crotwell]

Hi

Interesting, I ran your mseed file through my mseed file reading and
decompression routines and it read it fine. But looking more closely,
I see that the error is that the "last value" stored in the stiem1
data doesn't match the last value after decompression. This I think is
to be a check to make sure the compression/decompression didn't
introduce errors. Since this check is failing on multiple
decompression codes, my guess is that you are right that this is badly
compressed data. It likely also has nothing to do with the web
services, and is just "garbage in, garbage out".

Interestingly, the routines in my SeedCodec package  don't actually
check this, so I have not noticed this before. I guess the question
for me is should this check be made (presumably yes) and if it fails
is it more likely that the data are wrong or the header ending value.
It seems a bit odd to fail hard if these don't match in that the
decompression may be fine and the data usable, but the value at the
beginning of the stiem1 block is bad. Maybe should be more of a
warning??? I am not sure what the right thing to do in a decompression
code is, looks like obspy and mseedtosac have taken the "fail if
mismatch" route.

Philip



On Fri, Dec 16, 2011 at 1:50 AM, John D. West <john.d.west at asu.edu> wrote:
> Hi.
>
> When requesting data using the bulkdataselect service, I occasionally get a
> defective miniseed volume back. A sample is attached. (does this mail list
> allow attachments?).
>
> ObsPy fails to read it, and running mseed2sac on it gives me several error
> messages like "IU_NWAO__BHE_M: Warning: Data integrity check for Steim-1
> failed, last_data=2192, xn=-2165"
>
>      -- John
>
> _______________________________________________
> webservices mailing list
> webservices at iris.washington.edu
> http://www.iris.washington.edu/mailman/listinfo/webservices
>