Commit Graph

669 Commits

Author SHA1 Message Date
8b79abcaba Support setting exec_mem_limit in ExportJob (#1280) 2019-06-11 21:05:45 +08:00
ccf2e5bb9e Add page api for new format segment (#1270) 2019-06-11 10:37:16 +08:00
922fa28097 Add common and options for new format segment (#1269) 2019-06-11 09:34:58 +08:00
0900ab853f Add pb define for new format segment (#1268) 2019-06-11 09:34:32 +08:00
9d7f99a669 Add new file format design markdown (#1267) 2019-06-11 09:34:06 +08:00
84632cd062 Add BitMapIterator (#1277) 2019-06-11 09:23:02 +08:00
53062122ea Change strategy of incorrect data (#1255)
This change adds a load property named strict_mode which is used to prohibit the incorrect data.
When it is set to false, the incorrect data will be loaded by NULL just like before.
When it is set to true, the incorrect data which belongs to a column without expr will be filtered.
The strict_mode is supported in broker load v2 now. It will be supported in stream load later.
2019-06-10 20:39:45 +08:00
6a54464ee8 Fix bug that NullPredicate is not correctly handled by partition pruning (#1276) 2019-06-10 20:11:42 +08:00
2efd5a4d86 Fix bug: FE pid file has wrong content(#1273)
For example, we start the process for the first time. The pid is 12345. Due to the accident, the process is killed and the fe.pid exists. Then we start the process for the second time. The pid is 6789. The fe.pid shows 67895 , Because file.write only cover the first four digits. This case can happen easily when we use supervise. Then I add the file.setLength(0) and delete the old data.
2019-06-10 17:29:04 +08:00
8cd29d194e Fix += decimal error (#1272)
This change fix the +=decimal error when integer is zero or fraction is zero.
In this situation, the += operator will make a mistake.
2019-06-10 16:30:57 +08:00
038ddcfa0b Set timeout configuration for stream load (#1271) 2019-06-10 15:51:58 +08:00
d1b1fce92f Change LICENSE file (#1265) 2019-06-09 15:55:46 +08:00
3e1c70d1b7 Add coding function (#1264) 2019-06-08 21:02:31 +08:00
e4e04e8203 Make LZO support optional (#1263) 2019-06-07 22:26:54 +08:00
ff0dd0d2da Support SSL authentication with Kafka in routine load job (#1235) 2019-06-07 16:29:01 +08:00
cb91e15f1e Modify UDF docs (#1260) 2019-06-06 15:47:10 +08:00
934ca2481a Make MySQL support optional (#1248) 2019-06-05 12:28:15 +08:00
ece34fb838 Make hll function backward compatibility (#1251) 2019-06-05 11:12:36 +08:00
f424321625 Fix IllegalArgumentException in LoadManager (#1240) 2019-06-04 22:23:13 +08:00
6ce8087916 Fix bug that RowCusor do NOT match with RowBlock's layout (#1249) 2019-06-04 22:20:10 +08:00
9f5f44ec48 Reduce memory RowBlock needed (#1238)
Before RowBlock will reserve memory for all columns in schema, even if
it is not queried. Which will cause bad performance when quering wide
table.

In this patch, RowBlock will reserve memory for needed columns. In a
case, this reduce ConvertBatchTime from 10s to 60ms when quering a wide
table who has 178 columns.

 #1236
2019-06-04 12:58:41 +08:00
bedd94dca2 Upgrade brpc to 0.9.5 (#1243)
Change some ut
2019-06-04 11:13:23 +08:00
ae75e44e05 fixup leak memory (#1244)
When I declared that the compilation mode was BUILD_TYPE=LSAN, there was a memory leak after running doris.

be.out:
Direct leak of 32816 byte(s) in 1 object(s) allocated from:
    #0 0x1089666 in __interceptor_malloc ../../../../libsanitizer/lsan/lsan_interceptors.cc:53
    #1 0x7ff459547280 in __alloc_dir (/lib64/libc.so.6+0xc0280)

SUMMARY: LeakSanitizer: 32816 byte(s) leaked in 1 allocation(s).
2019-06-04 11:07:37 +08:00
6231fe0abc Fix FragmentMgrTest crash sometimes (#1232) 2019-06-01 18:10:24 +08:00
741539de91 Release udf headers & lib (#1231)
remove internal headers from udf.h
release udf headers & lib
2019-05-31 17:47:41 +08:00
7cdaba66dc Add spatial func (#1213)
Support some spatial functions, such as ST_Contains.
2019-05-31 14:23:09 +08:00
c20d62679e Add negative load from StreamLoad (#1227) 2019-05-31 07:14:06 +08:00
dc0cd5fd67 Fix the bug of += decimal in olap engine (#1226)
* Fix the bug of += decimal in olap engine
[ISSUE-1225] This change fix the olap engine bug of decimal agg. Using ^ instead of * to judge result is less then zero.
The result of * will be less then zero when the result is overflow. So the answer of += is incorrect.
2019-05-31 07:12:22 +08:00
309b779a7d Check colocate table name should be case-sensitive (#1224) 2019-05-30 22:47:22 +08:00
180d8e5cbd Modify some thirdparties (#1228)
1. Change Kafka java client from 2.0.0 to 0.10.1.1. Because high version client may not support low server server.
2. Enable SSL in librdkafka
2019-05-30 21:23:37 +08:00
fa4ac9f751 Replay GlobalVariable by Annotation (#1219) 2019-05-29 19:21:42 +08:00
319c71f69f Remove unnecessary import sun.security.ssl.Debug (#1215) 2019-05-29 12:35:20 +08:00
f648bdd968 Fix datediff function (#1208) 2019-05-28 15:55:31 +08:00
9d19c6c315 Support arbitrary kafka properties (#1204) 2019-05-28 10:03:50 +08:00
5ca2805701 Add some date time function doc (#1206) 2019-05-27 17:36:09 +08:00
08c8caeacf Add max cache size to ClientCache in BE (#1202)
Currently, unlimited client cache pool may cause too many connections in FE
2019-05-24 22:02:09 +08:00
41152f5c84 Modify FE sys log and audit log configuration (#1198)
All FE logs are now with suffix yyyyMMdd or yyyyMMddHH, specified by 2 new
configurations: 'sys_log_roll_interval' and 'audit_log_roll_interval'

All FE logs roll at max size of 1024MB(default), specified by new
configuration: 'log_roll_size_mb'

By default, the new FE logs will look like this:

log/
    fe.audit.log
    fe.audit.log.20190524-1
    fe.audit.log.20190523-2
    fe.log
    fe.log.20190524-1
    fe.log.20190524-2
    fe.log.20190523-3
    fe.warn.log
    fe.warn.log.20190524-1
    fe.warn.log.20190523-2
    fe.gc.log.20190524

Configurations 'sys_log_roll_mode' and 'audit_log_roll_mode' are deprecated.
2019-05-24 11:24:38 +08:00
85b4619d54 Change insert into to streaming (#1191)
The non-streaming hint of insert into will use the streamin plan which is same as the plan of stream insert.
It will also record the load info and return the label of insert stmt.
The partition is supportted in insert into stmt. The result which meet the target partitions will be loaded.
The introduction of example has been changed especially non-streaming insert.
Also, the param of partition_names is added in sql syntax which is used to declare the target partition_names in target table.

Change META_VERSION to 50
2019-05-23 20:53:30 +08:00
42601f2d88 Print unfinished tablet when cancel load job (#1186) 2019-05-23 19:22:54 +08:00
488e3825f7 Fix bug that restore process in BE causes BE crash (#1193)
When calling SnapshotLoader.move(), all files should be revoked if they
are in GC queue, or the file may be deleted after move() success.
2019-05-23 19:22:29 +08:00
cde315c9e9 Add date-function doc (#1190) 2019-05-23 15:29:08 +08:00
5d1457c0b6 Add check to create tablet upon alter tablet task (#1187)
When creating new tablet by alter tablet task,
next_unique_id will increase on the base old tablet.
If next_unique_id is eqaul to zero, it will cause that
ColumnDataMessage not match with tablet meta.
2019-05-23 14:17:22 +08:00
2fe1950d84 Fix minor string format bug (#1189) 2019-05-23 14:14:51 +08:00
d42409cc35 Fix short key not fill up all space (#1183) 2019-05-22 11:32:46 +08:00
5e245e031b Change source type to BATCH_LOAD_JOB (#1182) 2019-05-22 11:07:38 +08:00
592c2c24d9 Fix revoke files bug (#1181) 2019-05-22 11:06:31 +08:00
c5bf1a8da1 Fix prefix index comparison (#1180)
1. Upon prefix index comparison, it should only compare the fixed length of prefix index
2019-05-21 20:17:24 +08:00
b98b07efe2 Add persistence of load job v2 (#1176) 2019-05-21 19:30:26 +08:00
b132f4ac0c Add a configuration to force seek for block (#1179) 2019-05-21 14:53:37 +08:00
722a9e71c7 Optimize json functions (#1177)
1. get_json_xxx() now support using quoto to escape dot
2. Implement json_path_prepare() function to preprocess json_path

Performance of get_json_string() on 1000000 rows reduces from 2.27s to 0.27s
2019-05-21 09:13:12 +08:00