You have a large file, for example a large database dump file, which is greater than 10G after compression. When you tried to upload this file to eCuRep it always fails due to network interruption. What can you do to ensure this large file can be uploaded successfully?
So far, eCuRep does not have the function to continue uploading incomplete parts of a file after it has been uploaded partly. Especially for a large file, you will have to re-upload, and this will waste time.
A large file can be uploaded to eCuRep successfully after you split it into smaller files. If any one of the small files is not uploaded successfully, you can just upload that file again.
Perform the following steps:
Note: It is assumed that the big file is xxxxx.yyy.zzz.dmp.gz
PART A: on customer side.
1)Use unix/linux split command
$split -b 1024m xxxxx.yyy.zzz.dmp.gz &
This command will split the big file into some small files, and each of them is 1024MB, it is suggested to run this command in background mode.
2)Some small files with names xaa, xab, xac, ... can be found in the local directory.
3)Rename these small files as per eCuRep naming convention, for example: $mv xaa xxxxx.yyy.zzz.dmp.gz.xaa $mv xab xxxxx.yyy.zzz.dmp.gz.xab ... ...
Ensure to rename all small files. It is advised to use original file name as the prefix, and xaa (xab,...) as the suffix. We need the suffix because they will be used to merged into the original file.
After that, the customer can use https://www.ecurep.ibm.com/app/upload to upload those small files to eCuRep.
PART B: on IBM side.
1)Download these small files from eCuRep.
2)Use unix/linux command cat
$cat xxxxx.yyy.zzz.dmp.gz.* > xxxxx.yyy.zzz.dmp.gz &
This command will merge these small files into the original file.
3)Then you can uncompress and import the DB dump.
|Information Management||InfoSphere Master Data Management Server for Product Information Management||Not Applicable||AIX, HP-UX, Linux, Solaris||6.0, 6.5, 9.0, 9.1|
WebSphere Product Center
InfoSphere Master Data Management Server for Product Information Management