I have a workflow in which I am downloading a large volume of files from a CVMFS repository using the cvmfs_shrinkwrap utility, then storing those files in a squashfs image.
On my machine with limited disk space, I find I am running out of space on the device although I should have plenty of space to support the files that I intend to download. It appears that all of the files are being duplicated in various caches that I don’t believe I need.
I tried to set the cache quota limit with CVMFS_QUOTA_LIMIT, but I get the following error:
LibCvmfs version 2.8, revision 30 Initialization failed: Failure: libcvmfs supports only unmanaged exclusive cache or alien cache. (2)
Is there no way to use the quota limit with cvmfs_shrinkwrap?
If I don’t need a proper cache, how would you recommend I handle this situation to keep disk space utilization to a minimum?
Any help is greatly appreciated.
Here are more details of my configuration.
- Repository configuration file:
CVMFS_REPOSITORIES=[Repository Name] CVMFS_REPOSITORY_NAME=[Repository Name] CVMFS_CLAIM_OWNERSHIP=yes CVMFS_CACHE_BASE=/shrinkwrap_cvmfs/src-cache CVMFS_SERVER_URL="http://cvmfs-s1goc.opensciencegrid.org:8001/cvmfs/@fqrn@" CVMFS_HTTP_PROXY=DIRECT CVMFS_KEYS_DIR=/etc/cvmfs/keys/opensciencegrid.org CVMFS_CATALOG_WATERMARK=64 CVMFS_MAX_RETRIES=3 CVMFS_TIMEOUT=60 CVMFS_TIMEOUT_DIRECT=60
- cvmfs and cvmfs-shrinkwrap were installed with
sudo yum install -y https://ecsft.cern.ch/dist/cvmfs/cvmfs-release/cvmfs-release-latest.noarch.rpm \ && yum install -y cvmfs.x86_64 cvmfs-shrinkwrap.x86_64