gian/cubefs
1
0
Fork 0
mirror of https://github.com/cubefs/cubefs.git synced 2026-08-02 10:06:14 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
5,086 Commits 48 Branches 52 Tags 378 MiB
40f67d570b
Commit Graph

5086 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
yanghonggang
40f67d570b feat(blobstore): add blobstore bench tool 
The bench tool is used to evaluate the performance of blobstore.
It supports performance evaluation for put, get, and del operations.

Fixes: #3795

Signed-off-by: yanghonggang <yanghonggang_yewu@cmss.chinamobile.com>
Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-22 10:07:55 +08:00
yanghonggang
513c04c21f feat(bssdk): support log file configuration for sdk client 
This commit adds support for logging to a file in the blobstore sdk client,
allowing configuration of log file path, level, and size.

close: #3862

Signed-off-by: yanghonggang <yanghonggang_yewu@cmss.chinamobile.com>
Signed-off-by: zhangjianwei2 <zhangjianwei2_yewu@cmss.chinamobile.com>
 2025-07-21 16:57:35 +08:00
slasher
131b332136 chore(ci): split run test for cubefs and blobstore 
@formatter:off

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-09 10:05:10 +08:00
xiejian
b363b75d99 fix(proxy): avoid alloc 0 volume from clustermgr 
with: #1000217440

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:32:17 +08:00
Cloudstriff
ebb82972f6 fix(clustermgr): skip disk not found error when disk is dropping, avoid disk drop task blocking 
with #1000200308

Signed-off-by: Cloudstriff <chenjiong@oppo.com>
 2025-07-08 11:32:12 +08:00
slasher
514f9ac3b7 feat(access): add ordered configration of codemode to order reading shards 
EC24P8 in 1AZ
Matrix: C(32, 8) = 10518300
Memory: (24 + 24*24 + 24*24*8) * 10518300 ~= 51 GB

EC12P9 in 3AZ, 7 constant shards in local AZ
Matrix: C(14, 5) = 18018
Memory: (24 + 24*12 + 12*12*8) * 18018 ~= 25 MB

. #1000199588

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:32:07 +08:00
slasher
e56f5ccd96 fix(access): using last token in client 
. #1000203638

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
8ede90565d fix(shardnode): skip readIndex in shard.Stats() when service start 
with: #1000176274

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
dcdb7fb681 fix(shardnode): do not compare idc and rack info with clustermgr when init 
with: #1000176986

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
03cb728729 fix(shardnode): prevent seal part slices of local 
with: #1000183861

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
0c6d15042d fix(shardnode): 1.remove raft group when apply remove member 
avoid removed member still leader and clear raft data after all raft log applied
2. skip load disk which status is not normal in clustermgr
3.import gomonkey package

with: #1000176986

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
mawei029
b99da2112f fix(bssdk): fix list blob count uint overstep the boundary, modify panic to span error 
with: #1000179782

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
24f9066d35 feat(raft): print runtime stack when raft worker handle error 
with: #1000167662

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
413c1de3c2 fix(shardnode): check readIndexFunc loaded in shard when preRWCheck 
with: #1000167662

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
slasher
daf016636f fix(access): return origin error if context cancel 
. #1000042638

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:16 +08:00
slasher
2ebab23f6c fix(access): remove token if last slice only has one blob 
. #1000155215

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:16 +08:00
slasher
d4288318de feat(ci): add docker for go 1.18 
@formatter:off

close: #3782

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:16 +08:00
slasher
9ea30f5ac5 chore(docs): add rpc2 configuration 
@formatter:off

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:16 +08:00
Cloudstriff
21ef7c2a21 refactor(clustermgr): do refactor for v1.4.4 change 
with #1000138128

Signed-off-by: Cloudstriff <chenjiongwendao@qq.com>
 2025-07-08 11:30:16 +08:00
mawei029
fdf69e2596 fix(blobnode): fix the unstable test case 
with: #1000102964

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:16 +08:00
tangdeyi
78df429486 fix(clustermgr): drop node/disk apply err return 
with #1000102800

Signed-off-by: tangdeyi <tangdeyi@oppo.com>
 2025-07-08 11:30:16 +08:00
tangdeyi
ca68b3f33f fix(clustermgr): rename ak and sk 
with #1000095073

Signed-off-by: tangdeyi <tangdeyi@oppo.com>
 2025-07-08 11:30:16 +08:00
tangdeyi
d48ca7debc fix(clustermgr): fix snapshot member nodehost store 
with #1000095073

Signed-off-by: tangdeyi <tangdeyi@oppo.com>
 2025-07-08 11:30:16 +08:00
mawei029
d29e116edc fix(bssdk): support multiple clusters can be configured when consul is empty 
with: #1000045453

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
836738b7e1 feat(shardnode): distinguish item and blob key in storage 
with: #22989410

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:16 +08:00
JasonHu520
ed2e209b9c perf(raft): entry cache supports quickly get 
with: #23065907
Signed-off-by: JasonHu520 <huzongchao@oppo.com>
 2025-07-08 11:30:16 +08:00
JasonHu520
41bf952511 fix(raft): the entry cache supports overwriting by index 
with: #23065907
Signed-off-by: JasonHu520 <huzongchao@oppo.com>
 2025-07-08 11:30:16 +08:00
mawei029
4867b6ca6d fix(bssdk): old leader disk is broken, choose new leader from the old units 
with: #1000022240

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:16 +08:00
mawei029
c577b51612 fix(bssdk): old leader disk is broken, change leader disk 
with: #1000014262, #1000010625

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:16 +08:00
JasonHu520
ff9a307699 fix(blobnode): remove bad unit which update id failed with cm 
with: #1000010625

Signed-off-by: JasonHu520 <huzongchao@oppo.com>
 2025-07-08 11:30:16 +08:00
xiejian
27327ab3ce fix(shardnode): replay raft log by ReadIndex() before start serve, remove load shard asynchronous 
with: #1000004366

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
slasher
1ca6fc0554 feat(util): add util statistic for time delay 
@formatter:off

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
858895b18e fix(cli): cli support local file path as request body when PutBlob 
with: #23109197

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
b6a2be88ec refactor(bssdk): not support GetBlob data when not sealed blob 
modify alloc slice args count is real success count

with: #23109197

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
8274a65907 fix(shardnode): record slice count when SealBlob 
with #23109197

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
fd185ed965 test(shardnode): optimize shardnode ut 
with #22357426

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
74c69ccd41 fix(shardnode): remove failedSlice.Count check of AllocSlice api 
full check of SealBlob args, update location when SealBlob

with #23109197

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
slasher
f7664eec0f feat(rpc2): cache server request and response in pool 
. #23076156

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
e0a59713ff chore(bssdk): update code comment 
@formatter:off

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
f83500da17 feat(shardnode): clear shardnode list blob log, print blob when seal failed 
with #22357426

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
a7ffe01736 feat(raft): clear raft log 
with #22357426

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
fe107d3304 feat(shardnode): sync route version from clustermgr before start serve 
with #23092404

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
42d0bbfa37 fix(bssdk): prevent route version rollback 
with: #23031244

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
4b75603a06 fix(bssdk): fix location size after alloc slices 
with: #23053900

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
slasher
f7472eeb2d fix(access): remove marker of listing broken disk 
. #23065780

Signed-off-by: slasher <shenjie1@oppo.com>
 2025-07-08 11:30:15 +08:00
tangdeyi
6ad085725c fix(clustermgr): loadroute when apply snapshot 
with #23075869

Signed-off-by: tangdeyi <tangdeyi@oppo.com>
 2025-07-08 11:30:15 +08:00
tangdeyi
89219627e3 tools(clustermgr): get inconsistent volume between leader and follower 
with #23074437

Signed-off-by: tangdeyi <tangdeyi@oppo.com>
 2025-07-08 11:30:15 +08:00
JasonHu520
cdb9a30565 fix(scheduler): if get dest disk shard info failed, task should be redo 
with:#23080463

Signed-off-by: JasonHu520 <huzongchao@oppo.com>
 2025-07-08 11:30:15 +08:00
mawei029
6acfc1ad4b feat(tools): cli disk drop stat, support specify some disk, use cmd context, find vuid not in cm 
@formatter:off

Signed-off-by: mawei029 <mawei2@oppo.com>
 2025-07-08 11:30:15 +08:00
xiejian
a05df55013 fix(shardnode): prevent reopen broken disk as normal cause raft panic 
with #23075546

Signed-off-by: xiejian <xiejian3@oppo.com>
 2025-07-08 11:30:15 +08:00