[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[LDM #OFP-197738]: debugging LDM 6.10.1
- Subject: [LDM #OFP-197738]: debugging LDM 6.10.1
- Date: Thu, 21 Jun 2012 14:01:42 -0600
Sue,
> No worries. I think I may have it figured out. I think the checksum
> signature for "unique" products is the same. We are transferring files
> from our research system and from our operational system through the
> same LDM gateway servers. The product id's are unique but similar.
Whew! I'm glad to hear that it's probably not the LDM's fault. :-)
> ldmd.log:Jun 21 08:38:19 salt cospa3.rap.ucar.edu[11360] INFO:
> d9de5af0dc90916b21c4702136fb3ed5 62225 20120621143756.657 EXP
> 000
> ncar.ral.CoSPA.Shadow.BlendNcarPcEchoTops.2012-06-21T14:30:00.PT02:15.nc
>
> ldmd.log:Jun 21 08:38:51 salt cospa9.rap.ucar.edu[11359] INFO: hereis:
> duplicate: d9de5af0dc90916b21c4702136fb3ed5 62225
> 20120621143844.913 EXP 000
> ncar.ral.CoSPA.Live.BlendNcarPcEchoTops.2012-06-21T14:30:00.PT02:15.nc
Those two data-products do have the same signature, so the product-queue will
reject the second one as being a duplicate.
> I am not sure how the checksum is computed but I will look at that.
Typically, the signature is the MD5 checksum of the contents of the
data-product.
> Does this seem reasonable? How can we insure unique signatures for our
> products.
If you're using pqinsert(1) to insert the data-products into an LDM
product-queue, then, by default, the signature will be computed as the MD5
checksum of the contents of the file. The "-i" option will cause the signature
to be computed as the MD5 checksum of the product-identifier, so if the
identifiers are unique, then so will the signatures.
> Thanks,
> Sue
Regards,
Steve Emmerson
Ticket Details
===================
Ticket ID: OFP-197738
Department: Support LDM
Priority: Normal
Status: Closed