mirror of
https://github.com/versity/scoutfs.git
synced 2026-04-29 01:16:55 +00:00
64200ed61c
The srch compact and orphan scan workers called sync request RPCs that would block indefinitely if the server stopped answering. Both workers are idempotent and reschedule on error, so blocking forever buys nothing compared to treating a stalled RPC as a failure and trying again on the next tick. Add scoutfs_net_sync_request_timeout, a bounded-wait variant that returns -ETIMEDOUT if the response doesn't arrive in time. Response state lives on a refcounted heap allocation rather than the caller's stack so a late callback can't scribble into freed memory. On timeout we race with an arriving response for the msend under conn->lock: if find_request wins we queue_dead_free and drop the callback's ref; otherwise we wait for the in-flight callback to complete before returning. Add _timeout typed wrappers for the four RPCs these workers use and thread a 5 minute bound in from each worker. All other callers keep the unbounded client_sync_request path with its reconnect retries. Signed-off-by: Auke Kok <auke.kok@versity.com>
Introduction
scoutfs is a clustered in-kernel Linux filesystem designed to support large archival systems. It features additional interfaces and metadata so that archive agents can perform their maintenance workflows without walking all the files in the namespace. Its cluster support lets deployments add nodes to satisfy archival tier bandwidth targets.
The design goal is to reach file populations in the trillions, with the archival bandwidth to match, while remaining operational and responsive.
Highlights of the design and implementation include:
- Fully consistent POSIX semantics between nodes
- Atomic transactions to maintain consistent persistent structures
- Integrated archival metadata replaces syncing to external databases
- Dynamic seperation of resources lets nodes write in parallel
- 64bit throughout; no limits on file or directory sizes or counts
- Open GPLv2 implementation
Community Mailing List
Please join us on the open scoutfs-devel@scoutfs.org mailing list hosted on Google Groups