Commit graph

222 commits

Author SHA1 Message Date
Viktor Kuzmin 06e723f629 Waiting for writes to complete with acquired locks may lead to deadlock, fixes #4952 2023-11-17 07:38:14 -08:00
chrislu 310eec9a9c skip if path changed
mitigate https://github.com/seaweedfs/seaweedfs/issues/4968
2023-11-01 16:27:59 -07:00
chrislu cbc24c7b24 disconnect from old subscribers 2023-10-13 16:48:14 -07:00
chrislu b02fdeabff RetryForever => RetryUntil 2023-10-01 11:33:56 -07:00
chrislu 186b1f029b read source uses shared lock 2023-09-23 10:33:44 -07:00
chrislu 31fc165715 mount switch to ordered lock requests 2023-09-21 11:08:26 -07:00
chrislu 27af11f1e8 Revert "Revert "Merge branch 'master' into sub""
This reverts commit 0bb97709d4.
2023-09-18 18:47:34 -07:00
chrislu 0bb97709d4 Revert "Merge branch 'master' into sub"
This reverts commit 4d414f54a2, reversing
changes made to 4827425146.
2023-09-18 16:13:20 -07:00
wang wusong 9e5b1eef9b avoid passe lock by value 2023-09-14 06:51:12 -07:00
chrislu f5c4b642bb re-ordering 2023-08-16 23:54:23 -07:00
chrislu 3852307e94 renaming 2023-08-16 23:47:43 -07:00
chrislu 6c7fa567d4 add separate cache directory for write buffers 2023-08-16 23:39:21 -07:00
wusong 33b9192536
Fix large file chunks deleted by mistake (#4678) 2023-07-17 07:52:14 -07:00
wusong 0cb9ddd8ec
Fix data loss: add lock for metacache (#4664)
Co-authored-by: wang wusong <wangwusong@virtaitech.com>
2023-07-11 22:23:32 -07:00
wusong c44a6552d2
Fix buff use (#4628) 2023-07-03 08:33:34 -07:00
zemul 68f1cc34a5
[mount]Fix copy file range (#4407)
* fix:mount deadlock

* fix

---------

Co-authored-by: zemul <zhouzemiao@ihuman.com>
2023-04-23 22:34:38 -07:00
Chengyu Liu 0f653d2153
update fuse client log which should output the fh of FileHandle. (#4404) 2023-04-15 22:15:16 -07:00
zemul 0122e022ea
Mount concurrent read (#4400)
* fix:mount deadlock

* feat: concurrent read

* fix

* Remove useless code

* fix

---------

Co-authored-by: zemul <zhouzemiao@ihuman.com>
2023-04-13 22:32:45 -07:00
chrislu 5db9fcccd4 refactoring 2023-03-21 23:01:49 -07:00
Patrick Schmidt 71b33faef0
Skip parent directory creation in mount (#4310)
A POSIX filesystem does not implicitly create parent directories when
they do not exist. Directories must be explicitly created and
permissions be set.

This also fixes a bug where asynchronous operations would create a
file in the filer before the parent directory was created. If the
file was a symlink or another special type of file the directory
would inherit that type and become unusable in the mounted FS.
2023-03-15 12:37:25 -07:00
wusong 79af6b41d4
[mount] Fix read file-data in entry.content (#4207) 2023-02-15 07:18:56 -08:00
wusong fe573f26a8
memchunk save content twice (#4174) 2023-01-31 19:36:59 -08:00
wusong 665917fdf1
[mount]: delete redundant code (#4173) 2023-01-31 18:54:46 -08:00
wusong de081c0d64
[mount] fix metacache update (#4161) 2023-01-29 07:55:27 -08:00
chrislu 81fdf3651b grpc connection to filer add sw-client-id header 2023-01-20 01:48:12 -08:00
chrislu bfe5d910c6 use one readerCache for the whole file 2023-01-16 22:43:02 -08:00
chrislu c24678153c use memory chunks only for sequential writes 2023-01-15 21:10:08 -08:00
chrislu 313fd17ae4 allow random writes for mem chunk 2023-01-13 20:32:18 -08:00
Chris Lu 1cd2e64aac
merge chunks during upload (#4130)
* merge chunks during upload

* fix test
2023-01-12 23:07:36 -08:00
chrislu efcffe4f77 no need to set chunks during flush 2023-01-10 10:01:52 -08:00
chrislu 86657ed3a8 avoid repeatedly adding manifest chunks 2023-01-10 01:35:10 -08:00
chrislu 28fe578944 avoid possible nil entry 2023-01-10 01:07:34 -08:00
chrislu 5423790b2c add back previous chunk upload selection algo 2023-01-03 22:50:39 -08:00
Chris Lu d4566d4aaa
more solid weed mount (#4089)
* compare chunks by timestamp

* fix slab clearing error

* fix test compilation

* move oldest chunk to sealed, instead of by fullness

* lock on fh.entryViewCache

* remove verbose logs

* revert slat clearing

* less logs

* less logs

* track write and read by timestamp

* remove useless logic

* add entry lock on file handle release

* use mem chunk only, swap file chunk has problems

* comment out code that maybe used later

* add debug mode to compare data read and write

* more efficient readResolvedChunks with linked list

* small optimization

* fix test compilation

* minor fix on writer

* add SeparateGarbageChunks

* group chunks into sections

* turn off debug mode

* fix tests

* fix tests

* tmp enable swap file chunk

* Revert "tmp enable swap file chunk"

This reverts commit 985137ec47.

* simple refactoring

* simple refactoring

* do not re-use swap file chunk. Sealed chunks should not be re-used.

* comment out debugging facilities

* either mem chunk or swap file chunk is fine now

* remove orderedMutex  as *semaphore.Weighted

not found impactful

* optimize size calculation for changing large files

* optimize performance to avoid going through the long list of chunks

* still problems with swap file chunk

* rename

* tiny optimization

* swap file chunk save only successfully read data

* fix

* enable both mem and swap file chunk

* resolve chunks with range

* rename

* fix chunk interval list

* also change file handle chunk group when adding chunks

* pick in-active chunk with time-decayed counter

* fix compilation

* avoid nil with empty fh.entry

* refactoring

* rename

* rename

* refactor visible intervals to *list.List

* refactor chunkViews to *list.List

* add IntervalList for generic interval list

* change visible interval to use IntervalList in generics

* cahnge chunkViews to *IntervalList[*ChunkView]

* use NewFileChunkSection to create

* rename variables

* refactor

* fix renaming leftover

* renaming

* renaming

* add insert interval

* interval list adds lock

* incrementally add chunks to readers

Fixes:
1. set start and stop offset for the value object
2. clone the value object
3. use pointer instead of copy-by-value when passing to interval.Value
4. use insert interval since adding chunk could be out of order

* fix tests compilation

* fix tests compilation
2023-01-02 23:20:45 -08:00
chrislu 77b7c9f7cc lock and unlock chunks for read 2022-12-21 14:15:35 -08:00
chrislu d1797deccb remove duplicated logic of removing from writableChunks 2022-12-21 14:15:11 -08:00
chrislu f7beba8515 unlock before submitting the uploading jobs 2022-12-20 12:52:30 -08:00
chrislu 28d479e5c0 mount: adjust locking for upload pipeline 2022-12-19 15:07:22 -08:00
wusong 549354e324
Fix hardlink counting (#4042)
Signed-off-by: wusong <wangwusong@virtaitech.com>

Signed-off-by: wusong <wangwusong@virtaitech.com>
Co-authored-by: wusong <wangwusong@virtaitech.com>
2022-12-08 10:50:57 -08:00
chrislu 659f28b855 simple refactoring 2022-12-05 23:19:38 -08:00
chrislu ca49ef8ed8 duplicated logic 2022-12-05 23:19:22 -08:00
Chris Lu dac9c28d05
Revert "refactor: moved to locked entry" (#4035)
* Revert "refactor: moved to locked entry"

This reverts commit 94bc9afd9d.

* only add LockedEntry, no changes to entryLock

* fix compilation
2022-12-05 12:32:27 -08:00
chrislu 94bc9afd9d refactor: moved to locked entry 2022-12-04 23:33:05 -08:00
chrislu 70a4c98b00 refactor filer_pb.Entry and filer.Entry to use GetChunks()
for later locking on reading chunks
2022-11-15 06:33:36 -08:00
chrislu ea2637734a refactor filer proto chunk variable from mtime to modified_ts_ns 2022-10-28 12:53:19 -07:00
chrislu 5c2eb602c9 Revert "mount: remove mount directory from generating cache directory id"
This reverts commit 109dc7fdf6.
2022-10-14 22:24:14 -07:00
chrislu 109dc7fdf6 mount: remove mount directory from generating cache directory id
fix https://github.com/seaweedfs/seaweedfs/discussions/3811
2022-10-10 00:12:10 -07:00
chrislu 31922b2bf2 s3 to watch specific directories 2022-09-20 09:30:05 -07:00
Ryan Russell d65bdeef08
docs(meta_cache): readability fixes (#3691)
Signed-off-by: Ryan Russell <git@ryanrussell.org>

Signed-off-by: Ryan Russell <git@ryanrussell.org>
2022-09-15 02:04:57 -07:00
chrislu 21c0587900 go fmt 2022-09-14 23:06:44 -07:00