ambry
ambry copied to clipboard
linkedin
Distributed object store
I use Ambry for storing media files (sizes ranges from few megabytes to tens). And the ability to handle many requests simultaneously is quite important for me. I did testing...
Make its default value false. Some v1 blob id are experiencing issues.
GetRequests with Include_All and Include_Deleted_Blobs will return blobs from storage node even if they are deleted irrespective they are hard deleted or not. But from a customer standpoint, returning a...
Currently, store server authoriztion failutre is mapped to ServerErrorCode.Blob_Not_Found. However, router sends request for every replicas for Blob_Not_Found. Create a new server error code, let router handle it, and finally...
Got the following test failure on the most recent travis build. ``` com.github.ambry.store.BlobStoreCompactorTest > interruptionDuringLastCommitTest FAILED java.lang.AssertionError: Last modified time of an older index segment > newer index segment at...
Helix based cluster manager does not dynamically recognize newly added nodes. It assumes that `InstanceConfig` change notifications come in only for nodes that it already knows of. When such notifications...
BlobId getVersion() should be public to support tools using BlobId that need to know the version of the blobs they are working on.
Undoubtedly, ambry is one of the most useful distributed storage systems. But i do not know why there is no more detailed documents (no more management documents, no more tools...
This enhancement should allow us to 1. get the list of blob ids of the constituent data chunks of a composite blob. 2. determine if a given blob id is...
With pull request #827 , we are able to reorder replicas in the order of local, originating and then other dc replicas. One more enhancement can be done based on...