seaweedfs

History

Jayshan Raghunandan 1f1eac4f08 feat: improve aio support for admin/volume ingress and fix UI links (#8679 ) * feat: improve allInOne mode support for admin/volume ingress and fix master UI links - Add allInOne support to admin ingress template, matching the pattern used by filer and s3 ingress templates (or-based enablement with ternary service name selection) - Add allInOne support to volume ingress template, which previously required volume.enabled even when the volume server runs within the allInOne pod - Expose admin ports in allInOne deployment and service when allInOne.admin.enabled is set - Add allInOne.admin config section to values.yaml (enabled by default, ports inherit from admin.) - Fix legacy master UI templates (master.html, masterNewRaft.html) to prefer PublicUrl over internal Url when linking to volume server UI. The new admin UI already handles this correctly. fix: revert admin allInOne changes and fix PublicUrl in admin dashboard The admin binary (`weed admin`) is a separate process that cannot run inside `weed server` (allInOne mode). Revert the admin-related allInOne helm chart changes that caused 503 errors on admin ingress. Fix bug in cluster_topology.go where VolumeServer.PublicURL was set to node.Id (internal pod address) instead of the actual public URL. Add public_url field to DataNodeInfo proto message so the topology gRPC response carries the public URL set via -volume.publicUrl flag. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: use HTTP /dir/status to populate PublicUrl in admin dashboard The gRPC DataNodeInfo proto does not include PublicUrl, so the admin dashboard showed internal pod IPs instead of the configured public URL. Fetch PublicUrl from the master's /dir/status HTTP endpoint and apply it in both GetClusterTopology and GetClusterVolumeServers code paths. Also reverts the unnecessary proto field additions from the previous commit and cleans up a stray blank line in all-in-one-service.yml. * fix: apply PublicUrl link fix to masterNewRaft.html Match the same conditional logic already applied to master.html — prefer PublicUrl when set and different from Url. * fix: add HTTP timeout and status check to fetchPublicUrlMap Use a 5s-timeout client instead of http.DefaultClient to prevent blocking indefinitely when the master is unresponsive. Also check the HTTP status code before attempting to parse the response body. * fix: fall back to node address when PublicUrl is empty Prevents blank links in the admin dashboard when PublicUrl is not configured, such as in standalone or mixed-version clusters. * fix: log io.ReadAll error in fetchPublicUrlMap --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: Chris Lu <chris.lu@gmail.com>		8 hours ago
..
admin	feat: improve aio support for admin/volume ingress and fix UI links (#8679)	8 hours ago
cluster	admin: auto migrating master maintenance scripts to admin_script plugin config (#8509)	2 weeks ago
command	Make `weed-fuse` compatible with systemd-based mount (#6814)	9 hours ago
credential	fix(s3): omit NotResource:null from bucket policy JSON response (#8658)	2 days ago
filer	improve: large file sync throughput for remote.cache and filer.sync (#8676)	1 day ago
filer_client	fix: resolve gRPC DNS resolution issues in Kubernetes #8384 (#8387)	4 weeks ago
glog	glog: add --log_rotate_hours flag for time-based log rotation (#8685)	8 hours ago
iam	Add data file compaction to iceberg maintenance (Phase 2) (#8503)	3 days ago
iamapi	fix(s3): omit NotResource:null from bucket policy JSON response (#8658)	2 days ago
images	chore: execute goimports to format the code (#7983)	2 months ago
kms	S3 API: Add integration with KMS providers (#7152)	7 months ago
mount	Add data file compaction to iceberg maintenance (Phase 2) (#8503)	3 days ago
mq	mount: make metadata cache rebuilds snapshot-consistent (#8531)	2 weeks ago
notification	go fix	4 weeks ago
operation	master: return 503/Unavailable during topology warmup after leader change (#8529)	1 week ago
pb	improve: large file sync throughput for remote.cache and filer.sync (#8676)	1 day ago
plugin/worker	iceberg: add sort-aware compaction rewrite (#8666)	2 days ago
query	mount: make metadata cache rebuilds snapshot-consistent (#8531)	2 weeks ago
remote_storage	improve: large file sync throughput for remote.cache and filer.sync (#8676)	1 day ago
replication	improve: large file sync throughput for remote.cache and filer.sync (#8676)	1 day ago
s3api	fix(s3): omit NotResource:null from bucket policy JSON response (#8658)	2 days ago
security	fix: port in SNI address when using domainName instead of IP for master (#8500)	2 weeks ago
sequence	chore: execute goimports to format the code (#7983)	2 months ago
server	feat: improve aio support for admin/volume ingress and fix UI links (#8679)	8 hours ago
sftpd	Fix SFTP file upload failures with JWT filer tokens (#8448)	3 weeks ago
shell	improve: large file sync throughput for remote.cache and filer.sync (#8676)	1 day ago
static	Fix Broken Links (#5287)	2 years ago
stats	Persist S3 bucket counter metrics across idle periods (#8595)	1 week ago
storage	fix(ec): gather shards from all disk locations before rebuild (#8633)	4 days ago
telemetry	Prevent split-brain: Persistent ClusterID and Join Validation (#8022)	2 months ago
topology	Add data file compaction to iceberg maintenance (Phase 2) (#8503)	3 days ago
util	fix(replication): resume partial chunk reads on EOF instead of re-downloading (#8607)	7 days ago
wdclient	master: return 503/Unavailable during topology warmup after leader change (#8529)	1 week ago
worker	Add data file compaction to iceberg maintenance (Phase 2) (#8503)	3 days ago
Makefile	Move SQL engine and PostgreSQL server to their own binaries (#8417)	3 weeks ago
weed.go	Fix the issue where fuse command on a node cannot specify multiple configuration directory paths (#7874)	3 months ago