Chris LuandClaude Opus 5.5 29fad5a468 mount: wait for the flush before merging a shrink over buffered writes ...

A write inside the file leaves its bytes in dirty pages without moving
the size. Merging a peer's truncate then adopted the smaller size, and the
chunk uploaded later sat past it, unreadable through the handle. The event
path and the periodic metadata flush now refuse a merge that shrinks the
handle; the close flush uploads the pages first and merges then.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-11-12 17:33:00 -08:00

..

admin

admin: calm console restyle with light and dark themes (#12568)

2026-11-11 15:32:48 -08:00

cluster

fix(mount): refuse stale whole-entry content flushes (#12594)

2026-11-12 16:15:08 -08:00

command

replication: keep a local sink path the source recreated after a delete (#12581)

2026-11-12 09:23:54 -08:00

credential

plugin worker: stop forwarding the stream's admin token to job RPCs (#12391)

2026-11-04 05:26:12 -08:00

ec

shell: parallel moves and a per-run limit for volumeServer.evacuate (#12561)

2026-11-11 10:18:51 -08:00

filer

filer: gzip metadata log pieces before uploading them (#12580)

2026-11-12 09:13:50 -08:00

filer_client

…

glog

go fmt

2026-04-10 17:31:14 -07:00

iam

s3: resolve a role session's aws:username from its user, not its session name (#12572)

2026-11-11 15:32:13 -08:00

iamapi

iam: optionally require an iam: grant for self-service access-key actions (#11937)

2026-10-21 20:06:30 -07:00

images

image: add an optional public image processing gateway (#11593)

2026-10-05 19:49:02 +08:00

kms

kms/azure: fix encrypt/decrypt round trip for Azure Key Vault (#11692)

2026-10-10 23:40:19 +08:00

mount

mount: wait for the flush before merging a shrink over buffered writes

2026-11-12 17:33:00 -08:00

mq

mq/kafka: wait for a filer to register instead of panicking on an empty list (#12019)

2026-10-24 07:09:43 -07:00

notification

filer,pb,mount: keep per-chunk cipher keys out of logs (#12085)

2026-10-26 11:52:56 -07:00

operation

operation: probe octet-stream uploads for gzip like untyped ones (#12584)

2026-11-12 09:16:55 -08:00

pb

filer.meta.backup: resume an interrupted metadata copy and add -entriesPerSecond (#12549)

2026-11-11 09:08:11 -08:00

placement

placement: let callers reject candidates placement cannot judge (#10581)

2026-08-05 00:18:22 -07:00

plugin/worker

plugin worker: stop forwarding the stream's admin token to job RPCs (#12391)

2026-11-04 05:26:12 -08:00

query

filer: fail the request, not the process, when chunks cannot be resolved (#11685)

2026-10-10 23:39:29 +08:00

remote_storage

filer.remote.sync: bounded filer lookups, one chunk cache per process (#12441)

2026-11-06 09:35:16 -08:00

replication

replication: keep a local sink path the source recreated after a delete (#12581)

2026-11-12 09:23:54 -08:00

s3api

s3api: verify signed request bodies against the signed payload hash (#12577)

2026-11-12 10:17:09 -08:00

security

security: scoped filer tokens are not admin credentials (#12551)

2026-11-10 21:18:59 -08:00

sequence

[nfs] Add NFS (#9067)

2026-04-14 20:48:24 -07:00

server

fix(mount): refuse stale whole-entry content flushes (#12594)

2026-11-12 16:15:08 -08:00

sftpd

sftp: fail rm and rmdir of a missing path, and mkdir of an existing one (#12168)

2026-10-28 08:11:21 -07:00

shell

shell: confirm trim survivors still serve the volume before deleting (#12590)

2026-11-12 09:32:37 -08:00

static

adjust logo

2026-04-23 10:05:51 -07:00

stats

filer: a full remote-deletion tombstone table keeps the lazy fetch off (#12440)

2026-11-06 13:28:47 -08:00

storage

volume: refuse a tier upload whose .dat was written during the copy (#12562)

2026-11-11 10:19:06 -08:00

telemetry

telemetry: keep only clusters that store at least 10 GiB (#11138)

2026-09-03 12:36:00 -07:00

topology

topology: take dn.Lock when a heartbeat links a new disk type (#12535)

2026-11-10 12:17:21 -08:00

util

s3, filer: bound the wait for an upload slot (#12559)

2026-11-11 10:17:18 -08:00

wdclient

wdclient: keep a shared volume lookup alive when its first caller leaves (#12579)

2026-11-12 09:15:01 -08:00

worker

ec_balance: make the per-rack cap on global moves a worker setting (#12478)

2026-11-07 16:01:12 -08:00

Makefile

…

weed.go

shell: exit non-zero when a piped command fails (#11117)

2026-09-02 22:12:51 -07:00