| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
Lockfree connection pool (#54) * Modify connection pool to lockfree. * Use blocking queue in connection. * Dequeue limited number of reqs to alleviate hol blocking. --------- Co-authored-by: qiuziyan1998 <qiuziyan1@huawei.com> | 8 个月前 | |
init commit | 1 年前 | |
init commit | 1 年前 | |
Bugfix for falcon meta server (#18) * use try catch in abortion; start all services as process starts * avoid throwing error if self not found in shard table * fix the bug: cn has 3 replicas; add the check replication tool * if metaserver is in recovery status, dont start background service * avoid throwing error while aborting * add gdb and pstack in falcon-store docker image * add some logs in router.cpp * add missing head * modified the falcom cm * get local server id under user-controlled transaction * add the func to check extension is ready * modified the location of transaction * modified the docker build * modifed the log loc * start 2pc background through cm * change the log level * organize the condition to check the status of falcon extension loading --------- Co-authored-by: sth4nthL <liumingyu15@huawei.com> Co-authored-by: LEAKD98 <guo1049089770@gmail.com> | 1 年前 | |
Lockfree connection pool (#54) * Modify connection pool to lockfree. * Use blocking queue in connection. * Dequeue limited number of reqs to alleviate hol blocking. --------- Co-authored-by: qiuziyan1998 <qiuziyan1@huawei.com> | 8 个月前 | |
fix issue #61 Store cluster management (#62) * Fix bug: Multiple nodeIds uploaded caused by watcher thread reuploads before initialization. Reupload failure not killing the process. Delayed nullptr handling due to connection creation. Node id map fetch and update not considering same node id with different values. New feature: Using control file to dynamically determine whether to exit the process or not during heartbeat lost. New test: New test case in NodeUT. New integration test in regress test. * Fix typo. * Fix bug in candidate count. --------- Co-authored-by: qiuziyan <qiuziyan1998@icloud.com> | 6 个月前 | |
Bugfix for falcon meta server (#18) * use try catch in abortion; start all services as process starts * avoid throwing error if self not found in shard table * fix the bug: cn has 3 replicas; add the check replication tool * if metaserver is in recovery status, dont start background service * avoid throwing error while aborting * add gdb and pstack in falcon-store docker image * add some logs in router.cpp * add missing head * modified the falcom cm * get local server id under user-controlled transaction * add the func to check extension is ready * modified the location of transaction * modified the docker build * modifed the log loc * start 2pc background through cm * change the log level * organize the condition to check the status of falcon extension loading --------- Co-authored-by: sth4nthL <liumingyu15@huawei.com> Co-authored-by: LEAKD98 <guo1049089770@gmail.com> | 1 年前 | |
Bugfix for falcon meta server (#18) * use try catch in abortion; start all services as process starts * avoid throwing error if self not found in shard table * fix the bug: cn has 3 replicas; add the check replication tool * if metaserver is in recovery status, dont start background service * avoid throwing error while aborting * add gdb and pstack in falcon-store docker image * add some logs in router.cpp * add missing head * modified the falcom cm * get local server id under user-controlled transaction * add the func to check extension is ready * modified the location of transaction * modified the docker build * modifed the log loc * start 2pc background through cm * change the log level * organize the condition to check the status of falcon extension loading --------- Co-authored-by: sth4nthL <liumingyu15@huawei.com> Co-authored-by: LEAKD98 <guo1049089770@gmail.com> | 1 年前 | |
Lockfree connection pool (#54) * Modify connection pool to lockfree. * Use blocking queue in connection. * Dequeue limited number of reqs to alleviate hol blocking. --------- Co-authored-by: qiuziyan1998 <qiuziyan1@huawei.com> | 8 个月前 | |
init commit | 1 年前 | |
init commit | 1 年前 | |
Lockfree connection pool (#54) * Modify connection pool to lockfree. * Use blocking queue in connection. * Dequeue limited number of reqs to alleviate hol blocking. --------- Co-authored-by: qiuziyan1998 <qiuziyan1@huawei.com> | 8 个月前 |
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
| 8 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 8 个月前 | ||
| 6 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 8 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 8 个月前 |