gian/cubefs
1
0
Fork 0
mirror of https://github.com/cubefs/cubefs.git synced 2026-08-02 02:00:56 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
5,495 Commits 48 Branches 52 Tags 378 MiB
develop-v3.5.3
Commit Graph

5495 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
clinx
086034cb92 fix(flash): avoid returning network EOF 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-29 14:27:27 +08:00
clinx
6e73ed2cd0 fix(flash): change the write packet size to 64K to avoid timeout 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-29 14:27:27 +08:00
chihe
b5b2d3644c fix(client): do not wait for async flush if stream is error 
close:#1000363069

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-28 11:18:07 +08:00
clinx
b07e38d8b1 fix(flash): set connection deadline with ctx for put 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-28 10:00:24 +08:00
clinx
6503c0d7cc fix(flash): do not put cache miss error to warn 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-26 17:31:21 +08:00
shuqiang-zheng
45d91e1387 fix(master): when querying the decommissioning progress of a disk that has not yet been decommissioned, it returns the initial decommissioning status. 
close: #1000362788

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-26 15:57:45 +08:00
clinx
e306c6f6fb fix(flash): print cache miss error to debug log file 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-26 14:56:32 +08:00
shuqiang-zheng
53ef100852 fix(master): fix an issue where the maximum rollback count determination in dp did not match the actual rollback count. 
close: #1000359057

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-26 11:01:25 +08:00
shuqiang-zheng
5635b02677 fix(master): fix panic caused by lack of locking during concurrent access to the rafts map. 
close: #1000356742

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-26 11:01:25 +08:00
shuqiang-zheng
65ac8ca7b2 fix(master): fix the issue of inaccurate disk and node decommission progress due to leader changes. 
close: #1000349627

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-26 11:01:25 +08:00
clinx
77eb7c264c fix(flash): free pre-allocated space after file creation 
with: #1000360870

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-26 09:49:22 +08:00
chihe
f49728b4be fix(clinet):shuffle the order of hosts for aheadread 
close:#1000340945

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-25 17:02:07 +08:00
clinx
750ea762d1 fix(flash): change keyToDiskMap type 
with: #1000356686

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-24 20:20:57 +08:00
clinx
00bf6d2389 fix(flash): the master distributes the flashkeylimit value 
with: #1000356686

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-24 11:09:27 +08:00
clinx
76dd0886b5 fix(flash): the read and write rate limit sizes using master config 
with: #1000356820

Signed-off-by: clinx <chenlin1@oppo.com>

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-24 10:46:05 +08:00
chihe
de243a2a87 fix(client):fix the issue with the timing of triggering the addNextTask task 
close:#1000354285

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-23 14:25:59 +08:00
Victor1319
d24e646004 fix(cluster): update getAllDataPartitionWithDiskPathByDataNode to handle discard partitions. #1000355377 
- Modify function signature to include ignoreDiscard parameter
- Update logic to skip discarded data partitions when requested
- Ensure compatibility with existing data node retrieval logic

Signed-off-by: Victor1319 <zengxuewei@oppo.com>
 2025-09-23 10:41:19 +08:00
clinx
b6771c282a fix:(flash): check key limit with no wait 
with: #1000355286

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-23 10:37:36 +08:00
chihe
5657874b3f fix(client):fixing the prefetching sliding window logic 
close:#1000354285

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-22 19:59:43 +08:00
chihe
9d45588a13 fix(client):add unit test for aheadread 
@formatter:off

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-22 19:59:43 +08:00
clinx
52a3674321 feat(flash): atomic update request result 
with: #1000347719

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-22 15:33:37 +08:00
clinx
b5555ab40f fix(flash): config client flow limit by master 
with: #1000347719

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-19 15:16:11 +08:00
clinx
dfa3e22413 fix(flash): pull handleReadTimeout when flashnode start 
with: #1000347719

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-19 10:46:51 +08:00
clinx
53edbd9d19 fix(flash): change log level to warn 
with: #1000347719

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-18 16:09:44 +08:00
chihe
19ea486af1 fix(client): fix bugs in partly hit scenarios of the prefetch function 
@formatter:off

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-18 10:05:10 +08:00
dengxiaochao
a5f9ecd148 fix(sdk): incorrect offset when aheadread copy cacheBlock to req data buffer 
@formatter:off

Signed-off-by: dengxiaochao <dengxiaochao@bilibili.com>
 2025-09-18 09:57:43 +08:00
clinx
6685b34b2e fix(flash): to load blockv2 need to init key limieter 
with: #1000347719

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-17 18:04:21 +08:00
clinx
d4c2520e8f fix(client:) add flow limit for object operation 
with: #1000343199

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-17 09:48:47 +08:00
clinx
951e600424 feat(flash): add key limiter for cache read 
with: #1000274394

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-16 19:07:04 +08:00
shuqiang-zheng
dfcc603a03 fix(master): Optimize the reporting metrics for bad disks that fail to decommission promptly. 
close: #1000336199

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-16 19:01:21 +08:00
shuqiang-zheng
654d5bfcf2 fix(master): fix an issue where token might consume twice during the decommission repair process of two replicaNum dp, due to the master changing the leader. 
close: #1000336199

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-16 19:01:21 +08:00
chihe
a59c4c71ed fix(client): read from leader when excuting ahead read 
close:#1000340945

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-16 14:13:25 +08:00
zhumingze
2f64db405b fix(master): Don't show disk decommissionSuccess alerts triggered by node decommission. 
.#1000336575

Signed-off-by: zhumingze <zhumingze@oppo.com>
 2025-09-16 09:56:15 +08:00
clinx
9808c33d84 fix(cache): modify the retry logic to adapt to the first packet timeout duration 
with: #1000343199

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-15 19:19:30 +08:00
clinx
e2b5003956 fix(cache): improve read concurrency 
with: #1000326585

Signed-off-by: clinx <chenlin1@oppo.com>
 2025-09-15 11:41:13 +08:00
Victor1319
f47bc8ba9e fix(master): enhance error handling in getVolSimpleInfo by logging failures and updating metrics. #1000337685 
Signed-off-by: Victor1319 <zengxuewei@oppo.com>
 2025-09-12 16:19:23 +08:00
shuqiang-zheng
37db5cf3b1 feat(master): when traversing the dp decommission queue, add a check for the deletion status of its associated volume. 
close: #1000329496

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-11 15:41:53 +08:00
shuqiang-zheng
a89871ce64 feat(master): add status update records to dp decommission process for querying. 
close: #1000329528

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-11 15:41:53 +08:00
shuqiang-zheng
e1165a242e feat(master): add retries for ignore dataPartitions during disk decommission. 
close: #1000329518

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-11 15:41:53 +08:00
shuqiang-zheng
83748a2a03 fix(master): add display of remainingDpCnt when querying disk or dataNode decommission progress. 
close: #1000329505

Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-11 15:41:53 +08:00
chihe
d1e7e412e0 fix(client):cleanup eh when flush is completed 
close:#1000310987

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-10 09:59:50 +08:00
chihe
65c1b4a878 fix(client): fix error for ltp 
close:#1000310987

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-09 10:52:24 +08:00
chihe
428255293e fix(client); Cache blocks will not be evicted when they are not read. 
close:#1000310987

Signed-off-by: chihe <chihe@oppo.com>
 2025-09-08 09:59:24 +08:00
shuqiang-zheng
cee19c2f90 fix(master): if the bad disk is not completed decommission within the time limit, the disk is reported in metrics. 
close:#1000242504
Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-04 10:01:19 +08:00
zhumingze
bdcc980d75 fix(master): Fixed the problem of repeated execution of disk decommission. #1000247984 
Signed-off-by: zhumingze <zhumingze@oppo.com>
 2025-09-04 10:01:19 +08:00
zhumingze
6c244dcf69 fix(data): Determine the disk lost based on the DP information on the master and the locally loaded DP information. #1000244723 
Signed-off-by: zhumingze <zhumingze@oppo.com>
 2025-09-04 10:01:19 +08:00
shuqiang-zheng
69d75fee82 fix(master): ignore dp that has the same srcAddr but are in the autoAddReplica status when the disk decommission. 
close:#1000242504
Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-04 10:01:19 +08:00
shuqiang-zheng
f0a5cd5fc9 fix(master): remove the interface that resets the decommission state of the disk to avoid misuse. 
close:#1000242504
Signed-off-by: shuqiang-zheng <zhengshuqiang@oppo.com>
 2025-09-04 10:01:19 +08:00
zhumingze
a535f6eed1 fix(data): Fix autoComputeCRC log issue. #1000232936 
Signed-off-by: zhumingze <zhumingze@oppo.com>
 2025-09-04 10:01:19 +08:00
zhumingze
da39949aa8 fix(master): Optimize fileStats related interfaces. #1000240718 
Signed-off-by: zhumingze <zhumingze@oppo.com>
 2025-09-04 10:01:19 +08:00