| Description | We should have a cache folder with all files extracted. It would be useful for troubleshooting. The files could have a well known name using the batch ID. It could skip extracting if the cache file exists. The PurgeJob could have a setting to delete cache files that get so old. A clustered system could specify the cache folder on a network file system. Without a specified folder, it would default to a well known folder within the system temporary folder. One potential problem with using cached files is if the system crashes in the middle of generating one. To avoid that, we could extract to a temporary file first, then on successful extraction, rename it to the well-known name. |
|---|