为什么同时执行任务失败并导致 Ignite 冻结？
Why are simultaneously executing tasks failing and causing Ignite to freeze?
OS: Ubuntu 18.04
阿帕奇点燃：2.8.1
其他信息：
Flask 应用程序 - 用于 API 个端点
pyignite - 作为 Apache Ignite 的瘦客户端
已加载数据：
5 张桌子：
A：1000万条记录
B：3750万条记录
C：1000万条记录
D：2530 万条记录
E：550 万条记录
Ignite Persistence 总共占用 29GB space。
我正在执行两个简单的 SQL 查询：
查询 1：
SELECT SUM(TABLE_B.ID) AS num_people, TABLE_B.PRODUCT, TABLE_C.CITY
JOIN TABLE_C
ON TABLE_B.ID = TABLE_C.ID
GROUP BY TABLE_B.PRODUCT, TABLE_C.CITY
ORDER BY num_people DESC
LIMIT 1000
查询 2：
SELECT SUM(TABLE_A.ID) AS num_people, TABLE_C.CITY
JOIN TABLE_C
ON TABLE_A.ID = TABLE_C.ID
GROUP BY TABLE_C.CITY
ORDER BY num_people DESC
LIMIT 1000
我 运行 在 docker 容器中安装 Apache Ignite。
这是配置：
<?xml version="1.0" encoding="UTF-8"?>
<!--
Licensed to the Apache Software Foundation (ASF) ...
-->
<beans xmlns="http://www.springframework.org/schema/beans"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="
http://www.springframework.org/schema/beans
http://www.springframework.org/schema/beans/spring-beans.xsd">
<!--
Alter configuration below as needed.
-->
<bean id="ignite.cfg" class="org.apache.ignite.configuration.IgniteConfiguration">
<!-- Enabling Apache Ignite Persistent Store. -->
<property name="dataStorageConfiguration">
<bean class="org.apache.ignite.configuration.DataStorageConfiguration">
<property name="defaultDataRegionConfiguration">
<bean class="org.apache.ignite.configuration.DataRegionConfiguration">
<!-- Setting the size of the default region to 48GB. -->
<property name="maxSize" value="#{48L * 1024 * 1024 * 1024}"/>
<!-- Set the page size to 4 KB -->
<property name="pageSize" value="#{4 * 1024}"/>
<!-- Enable Native Persistence - required for Authentication -->
<property name="persistenceEnabled" value="true"/>
</bean>
</property>
</bean>
</property>
<!-- Enabling authentication. -->
<property name="authenticationEnabled" value="true"/>
<!-- Enabling node discovery for Ignite Visor. -->
<property name="discoverySpi">
<bean class="org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi">
<property name="ipFinder">
<bean class="org.apache.ignite.spi.discovery.tcp.ipfinder.vm.TcpDiscoveryVmIpFinder">
<property name="addresses">
<list>
<value>ignite:47500</value>
<value>ignite:47501</value>
</list>
</property>
</bean>
</property>
</bean>
</property>
<property name="communicationSpi">
<bean class="org.apache.ignite.spi.communication.tcp.TcpCommunicationSpi">
<property name="sharedMemoryPort" value="-1"/>
</bean>
</property>
<!-- Configure internal thread pool. -->
<!-- <property name="publicThreadPoolSize" value="32"/>-->
<!-- Configure system thread pool. -->
<!-- <property name="systemThreadPoolSize" value="32"/>-->
<!-- Configure query thread pool. -->
<!-- <property name="queryThreadPoolSize" value="32"/>-->
<property name="cacheConfiguration">
<list>
<bean abstract="true" class="org.apache.ignite.configuration.CacheConfiguration"
id="cache-template-bean">
<!-- when you create a template via XML configuration, you must add an asterisk to
the name of the template -->
<property name="name" value="tbl_pll*"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="queryParallelism" value="32"/>
<!-- Other cache parameters -->
</bean>
<bean abstract="true" class="org.apache.ignite.configuration.CacheConfiguration"
id="cache-template-bean">
<!-- when you create a template via XML configuration, you must add an asterisk to
the name of the template -->
<property name="name" value="tbl_hf_pll*"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="queryParallelism" value="16"/>
<!-- Other cache parameters -->
</bean>
</list>
</property>
<property name="failoverSpi">
<bean class="org.apache.ignite.spi.failover.always.AlwaysFailoverSpi">
<property name="maximumFailoverAttempts" value="2"/>
</bean>
</property>
<!-- Execute one job at a time. -->
<property name="collisionSpi">
<bean class="org.apache.ignite.spi.collision.fifoqueue.FifoQueueCollisionSpi">
<property name="parallelJobsNumber" value="1"/>
</bean>
</property>
</bean>
</beans>
当我 运行 同时查询两个 SQL 时，它们都冻结并且第二个查询总是抛出类似的错误。有时是未知类型代码：b' ' 而这次是：
app_1 | ERROR:app:Unknown type code: `b'm'`
app_1 | DEBUG:app:Traceback (most recent call last):
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 275, in parse
app_1 | data_class = tc_map(type_code)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 108, in tc_map
app_1 | return _memo_map[key]
app_1 | KeyError: b'm'
app_1 |
app_1 | During handling of the above exception, another exception occurred:
app_1 |
app_1 | Traceback (most recent call last):
app_1 | File "/app/analytics/views.py", line 832, in post
app_1 | ignore_data_limit=ignore_data_limit
app_1 | File "/app/analytics/apply_rules.py", line 769, in create_view_ignite
app_1 | dataset_source_details=dataset_source_details)
app_1 | File "/app/analytics/apply_rules.py", line 685, in apply_rules_ignite
app_1 | explain = client.sql('EXPLAIN ' + QUERY)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/client.py", line 401, in sql
app_1 | max_rows, timeout,
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/api/sql.py", line 379, in sql_fields
app_1 | response_class, recv_buffer = response_struct.parse(connection)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/queries/__init__.py", line 146, in parse
app_1 | field_class, field_buffer = AnyDataObject.parse(client)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 277, in parse
app_1 | raise ParseError('Unknown type code: `{}`'.format(type_code))
app_1 | pyignite.exceptions.ParseError: Unknown type code: `b'm'`
点燃日志：
ignite_1 | [10:29:31,529][WARNING][query-#176][IgniteH2Indexing] Long running query is finished [time=6230ms, type=MAP, distributedJoin=false, enforceJoinOrder=true, lazy=false, schema=PUBLIC, node=TcpDiscoveryNode [id=a3f4gthv, consistentId=a4deb8, addrs=ArrayList [127.0.0.1, 172.19.0.5], sockAddrs=HashSet [/127.0.0.1:47500, 97ba4d812948/172.19.0.5:47500], discPort=47500, order=1, intOrder=1, lastExchangeTime=1603362363483, loc=true, ver=2.8.1#20200521-sha1:86422096, isClient=false], reqId=6, segment=31, sql='SELECT
ignite_1 | __Z0.ID __C0_0,
ignite_1 | __Z1.CITY __C0_1
ignite_1 | FROM PUBLIC.TABLE_C __Z1
ignite_1 | INNER JOIN PUBLIC.TABLE_A __Z0
ignite_1 | ON TRUE
ignite_1 | WHERE __Z0.ID = __Z1.ID', plan=SELECT
ignite_1 | __Z0.ID AS __C0_0,
ignite_1 | __Z1.CITY AS __C0_1
ignite_1 | FROM PUBLIC.TABLE_C __Z1
ignite_1 | /* PUBLIC."IDX_TABLE_C_proxy" */
ignite_1 | INNER JOIN PUBLIC.TABLE_A __Z0
ignite_1 | /* PUBLIC."_key_PK_proxy": ID = __Z1.ID */
ignite_1 | ON 1=1
ignite_1 | WHERE __Z0.ID = __Z1.ID]
ignite_1 | [10:29:34,890][WARNING][long-qry-#89][LongRunningQueryManager] Query execution is too long [time=3363ms, type=REDUCE, distributedJoin=false, enforceJoinOrder=false, lazy=false, schema=PUBLIC, reqId=6, sql='SELECT COUNT(DISTINCT `TABLE_A`.`id`) `num_people`,`TABLE_C`.`city` FROM `TABLE_A` JOIN `TABLE_C` ON `TABLE_A`.`id`=`TABLE_C`.`id` GROUP BY `TABLE_C`.`city` ORDER BY `num_people` DESC LIMIT 10000]
ignite_1 | [10:29:35,923][WARNING][client-connector-#211][IgniteH2Indexing] Long running query is finished [time=4400ms, type=REDUCE, distributedJoin=false, enforceJoinOrder=false, lazy=false, schema=PUBLIC, reqId=6, sql='SELECT COUNT(DISTINCT `TABLE_A`.`id`) `num_people`,`TABLE_C`.`city` FROM `TABLE_A` JOIN `TABLE_C` ON `TABLE_A`.`id`=`TABLE_C`.`id` GROUP BY `TABLE_C`.`city` ORDER BY `num_people` DESC LIMIT 10000]
ignite_1 | [10:30:03,730][INFO][grid-timeout-worker-#71][IgniteKernal]
ignite_1 | Metrics for local node (to disable set 'metricsLogFrequency' to 0)
ignite_1 | ^-- Node [id=a3f4gthv, uptime=00:04:00.021]
ignite_1 | ^-- H/N/C [hosts=1, nodes=1, CPUs=32]
ignite_1 | ^-- CPU [cur=0.03%, avg=7.18%, GC=0%]
ignite_1 | ^-- PageMemory [pages=3141162]
ignite_1 | ^-- Heap [used=5288MB, free=67.72%, comm=16384MB]
ignite_1 | ^-- Off-heap [used=12413MB, free=74.9%, comm=49352MB]
ignite_1 | ^-- sysMemPlc region [used=0MB, free=99.99%, comm=100MB]
ignite_1 | ^-- default region [used=12413MB, free=74.74%, comm=49152MB]
ignite_1 | ^-- metastoreMemPlc region [used=0MB, free=99.92%, comm=0MB]
ignite_1 | ^-- TxLog region [used=0MB, free=100%, comm=100MB]
ignite_1 | ^-- Ignite persistence [used=29178MB]
ignite_1 | ^-- sysMemPlc region [used=0MB]
ignite_1 | ^-- default region [used=29177MB]
ignite_1 | ^-- metastoreMemPlc region [used=0MB]
ignite_1 | ^-- TxLog region [used=0MB]
ignite_1 | ^-- Outbound messages queue [size=0]
ignite_1 | ^-- Public thread pool [active=0, idle=0, qSize=0]
ignite_1 | ^-- System thread pool [active=0, idle=5, qSize=0]
ignite_1 | [10:31:03,733][INFO][grid-timeout-worker-#71][IgniteKernal]
此后，Ignite Metrics 不断出现，节点不再 returns 任何输出或接受新任务。没有记录 JVM 暂停。
我做错了什么，如果同时查询 运行，为什么 Ignite 冻结并且不响应？
如果您一次 运行 一个查询 运行 就可以正确查询。
PS:
我在 Nabble 论坛上阅读了大量 Ignite 文档和故障排除。
任何帮助将不胜感激。
您很可能 运行 堆满了。使用该查询，特别是如果您没有正确的索引，您将使用大量堆，并且当您 运行 退出时，Java 几乎会冻结试图在循环。
你有什么指标？另外，如果您选择 sum(column1)，“按 column1 排序”对您意味着什么？
你能显示来自 Ignite 节点的日志吗？我会特别寻找“JVM 暂停”消息。
我能够解决问题！ （犯了一个愚蠢的错误）
我使用 python 瘦客户端连接到 Ignite，我没有为每个查询创建一个新客户端，而是重复使用了一个全局客户端。
我想事情是这样的：
由于客户端在发送一个查询执行后是 运行 第二个查询，因此未检索到第一个查询的输出。这阻止了 Ignite 运行 第二个查询和检索第一个查询的输出。
如果对 Ignite 有更多了解的人可以纠正我或告诉我我是否正确，将不胜感激。
OS: Ubuntu 18.04
阿帕奇点燃：2.8.1
其他信息： Flask 应用程序 - 用于 API 个端点 pyignite - 作为 Apache Ignite 的瘦客户端
已加载数据：
5 张桌子：
A：1000万条记录
B：3750万条记录
C：1000万条记录
D：2530 万条记录
E：550 万条记录
Ignite Persistence 总共占用 29GB space。
我正在执行两个简单的 SQL 查询：
查询 1：
SELECT SUM(TABLE_B.ID) AS num_people, TABLE_B.PRODUCT, TABLE_C.CITY
JOIN TABLE_C
ON TABLE_B.ID = TABLE_C.ID
GROUP BY TABLE_B.PRODUCT, TABLE_C.CITY
ORDER BY num_people DESC
LIMIT 1000
查询 2：
SELECT SUM(TABLE_A.ID) AS num_people, TABLE_C.CITY
JOIN TABLE_C
ON TABLE_A.ID = TABLE_C.ID
GROUP BY TABLE_C.CITY
ORDER BY num_people DESC
LIMIT 1000
我 运行 在 docker 容器中安装 Apache Ignite。
这是配置：
<?xml version="1.0" encoding="UTF-8"?>
<!--
Licensed to the Apache Software Foundation (ASF) ...
-->
<beans xmlns="http://www.springframework.org/schema/beans"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="
http://www.springframework.org/schema/beans
http://www.springframework.org/schema/beans/spring-beans.xsd">
<!--
Alter configuration below as needed.
-->
<bean id="ignite.cfg" class="org.apache.ignite.configuration.IgniteConfiguration">
<!-- Enabling Apache Ignite Persistent Store. -->
<property name="dataStorageConfiguration">
<bean class="org.apache.ignite.configuration.DataStorageConfiguration">
<property name="defaultDataRegionConfiguration">
<bean class="org.apache.ignite.configuration.DataRegionConfiguration">
<!-- Setting the size of the default region to 48GB. -->
<property name="maxSize" value="#{48L * 1024 * 1024 * 1024}"/>
<!-- Set the page size to 4 KB -->
<property name="pageSize" value="#{4 * 1024}"/>
<!-- Enable Native Persistence - required for Authentication -->
<property name="persistenceEnabled" value="true"/>
</bean>
</property>
</bean>
</property>
<!-- Enabling authentication. -->
<property name="authenticationEnabled" value="true"/>
<!-- Enabling node discovery for Ignite Visor. -->
<property name="discoverySpi">
<bean class="org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi">
<property name="ipFinder">
<bean class="org.apache.ignite.spi.discovery.tcp.ipfinder.vm.TcpDiscoveryVmIpFinder">
<property name="addresses">
<list>
<value>ignite:47500</value>
<value>ignite:47501</value>
</list>
</property>
</bean>
</property>
</bean>
</property>
<property name="communicationSpi">
<bean class="org.apache.ignite.spi.communication.tcp.TcpCommunicationSpi">
<property name="sharedMemoryPort" value="-1"/>
</bean>
</property>
<!-- Configure internal thread pool. -->
<!-- <property name="publicThreadPoolSize" value="32"/>-->
<!-- Configure system thread pool. -->
<!-- <property name="systemThreadPoolSize" value="32"/>-->
<!-- Configure query thread pool. -->
<!-- <property name="queryThreadPoolSize" value="32"/>-->
<property name="cacheConfiguration">
<list>
<bean abstract="true" class="org.apache.ignite.configuration.CacheConfiguration"
id="cache-template-bean">
<!-- when you create a template via XML configuration, you must add an asterisk to
the name of the template -->
<property name="name" value="tbl_pll*"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="queryParallelism" value="32"/>
<!-- Other cache parameters -->
</bean>
<bean abstract="true" class="org.apache.ignite.configuration.CacheConfiguration"
id="cache-template-bean">
<!-- when you create a template via XML configuration, you must add an asterisk to
the name of the template -->
<property name="name" value="tbl_hf_pll*"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="queryParallelism" value="16"/>
<!-- Other cache parameters -->
</bean>
</list>
</property>
<property name="failoverSpi">
<bean class="org.apache.ignite.spi.failover.always.AlwaysFailoverSpi">
<property name="maximumFailoverAttempts" value="2"/>
</bean>
</property>
<!-- Execute one job at a time. -->
<property name="collisionSpi">
<bean class="org.apache.ignite.spi.collision.fifoqueue.FifoQueueCollisionSpi">
<property name="parallelJobsNumber" value="1"/>
</bean>
</property>
</bean>
</beans>
当我 运行 同时查询两个 SQL 时，它们都冻结并且第二个查询总是抛出类似的错误。有时是未知类型代码：b' ' 而这次是：
app_1 | ERROR:app:Unknown type code: `b'm'`
app_1 | DEBUG:app:Traceback (most recent call last):
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 275, in parse
app_1 | data_class = tc_map(type_code)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 108, in tc_map
app_1 | return _memo_map[key]
app_1 | KeyError: b'm'
app_1 |
app_1 | During handling of the above exception, another exception occurred:
app_1 |
app_1 | Traceback (most recent call last):
app_1 | File "/app/analytics/views.py", line 832, in post
app_1 | ignore_data_limit=ignore_data_limit
app_1 | File "/app/analytics/apply_rules.py", line 769, in create_view_ignite
app_1 | dataset_source_details=dataset_source_details)
app_1 | File "/app/analytics/apply_rules.py", line 685, in apply_rules_ignite
app_1 | explain = client.sql('EXPLAIN ' + QUERY)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/client.py", line 401, in sql
app_1 | max_rows, timeout,
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/api/sql.py", line 379, in sql_fields
app_1 | response_class, recv_buffer = response_struct.parse(connection)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/queries/__init__.py", line 146, in parse
app_1 | field_class, field_buffer = AnyDataObject.parse(client)
app_1 | File "/usr/local/lib/python3.6/site-packages/pyignite/datatypes/internal.py", line 277, in parse
app_1 | raise ParseError('Unknown type code: `{}`'.format(type_code))
app_1 | pyignite.exceptions.ParseError: Unknown type code: `b'm'`
点燃日志：
ignite_1 | [10:29:31,529][WARNING][query-#176][IgniteH2Indexing] Long running query is finished [time=6230ms, type=MAP, distributedJoin=false, enforceJoinOrder=true, lazy=false, schema=PUBLIC, node=TcpDiscoveryNode [id=a3f4gthv, consistentId=a4deb8, addrs=ArrayList [127.0.0.1, 172.19.0.5], sockAddrs=HashSet [/127.0.0.1:47500, 97ba4d812948/172.19.0.5:47500], discPort=47500, order=1, intOrder=1, lastExchangeTime=1603362363483, loc=true, ver=2.8.1#20200521-sha1:86422096, isClient=false], reqId=6, segment=31, sql='SELECT
ignite_1 | __Z0.ID __C0_0,
ignite_1 | __Z1.CITY __C0_1
ignite_1 | FROM PUBLIC.TABLE_C __Z1
ignite_1 | INNER JOIN PUBLIC.TABLE_A __Z0
ignite_1 | ON TRUE
ignite_1 | WHERE __Z0.ID = __Z1.ID', plan=SELECT
ignite_1 | __Z0.ID AS __C0_0,
ignite_1 | __Z1.CITY AS __C0_1
ignite_1 | FROM PUBLIC.TABLE_C __Z1
ignite_1 | /* PUBLIC."IDX_TABLE_C_proxy" */
ignite_1 | INNER JOIN PUBLIC.TABLE_A __Z0
ignite_1 | /* PUBLIC."_key_PK_proxy": ID = __Z1.ID */
ignite_1 | ON 1=1
ignite_1 | WHERE __Z0.ID = __Z1.ID]
ignite_1 | [10:29:34,890][WARNING][long-qry-#89][LongRunningQueryManager] Query execution is too long [time=3363ms, type=REDUCE, distributedJoin=false, enforceJoinOrder=false, lazy=false, schema=PUBLIC, reqId=6, sql='SELECT COUNT(DISTINCT `TABLE_A`.`id`) `num_people`,`TABLE_C`.`city` FROM `TABLE_A` JOIN `TABLE_C` ON `TABLE_A`.`id`=`TABLE_C`.`id` GROUP BY `TABLE_C`.`city` ORDER BY `num_people` DESC LIMIT 10000]
ignite_1 | [10:29:35,923][WARNING][client-connector-#211][IgniteH2Indexing] Long running query is finished [time=4400ms, type=REDUCE, distributedJoin=false, enforceJoinOrder=false, lazy=false, schema=PUBLIC, reqId=6, sql='SELECT COUNT(DISTINCT `TABLE_A`.`id`) `num_people`,`TABLE_C`.`city` FROM `TABLE_A` JOIN `TABLE_C` ON `TABLE_A`.`id`=`TABLE_C`.`id` GROUP BY `TABLE_C`.`city` ORDER BY `num_people` DESC LIMIT 10000]
ignite_1 | [10:30:03,730][INFO][grid-timeout-worker-#71][IgniteKernal]
ignite_1 | Metrics for local node (to disable set 'metricsLogFrequency' to 0)
ignite_1 | ^-- Node [id=a3f4gthv, uptime=00:04:00.021]
ignite_1 | ^-- H/N/C [hosts=1, nodes=1, CPUs=32]
ignite_1 | ^-- CPU [cur=0.03%, avg=7.18%, GC=0%]
ignite_1 | ^-- PageMemory [pages=3141162]
ignite_1 | ^-- Heap [used=5288MB, free=67.72%, comm=16384MB]
ignite_1 | ^-- Off-heap [used=12413MB, free=74.9%, comm=49352MB]
ignite_1 | ^-- sysMemPlc region [used=0MB, free=99.99%, comm=100MB]
ignite_1 | ^-- default region [used=12413MB, free=74.74%, comm=49152MB]
ignite_1 | ^-- metastoreMemPlc region [used=0MB, free=99.92%, comm=0MB]
ignite_1 | ^-- TxLog region [used=0MB, free=100%, comm=100MB]
ignite_1 | ^-- Ignite persistence [used=29178MB]
ignite_1 | ^-- sysMemPlc region [used=0MB]
ignite_1 | ^-- default region [used=29177MB]
ignite_1 | ^-- metastoreMemPlc region [used=0MB]
ignite_1 | ^-- TxLog region [used=0MB]
ignite_1 | ^-- Outbound messages queue [size=0]
ignite_1 | ^-- Public thread pool [active=0, idle=0, qSize=0]
ignite_1 | ^-- System thread pool [active=0, idle=5, qSize=0]
ignite_1 | [10:31:03,733][INFO][grid-timeout-worker-#71][IgniteKernal]
此后，Ignite Metrics 不断出现，节点不再 returns 任何输出或接受新任务。没有记录 JVM 暂停。
我做错了什么，如果同时查询 运行，为什么 Ignite 冻结并且不响应？
如果您一次 运行 一个查询 运行 就可以正确查询。
PS: 我在 Nabble 论坛上阅读了大量 Ignite 文档和故障排除。 任何帮助将不胜感激。
您很可能 运行 堆满了。使用该查询，特别是如果您没有正确的索引，您将使用大量堆，并且当您 运行 退出时，Java 几乎会冻结试图在循环。
你有什么指标？另外，如果您选择 sum(column1)，“按 column1 排序”对您意味着什么？
你能显示来自 Ignite 节点的日志吗？我会特别寻找“JVM 暂停”消息。
我能够解决问题！ （犯了一个愚蠢的错误）
我使用 python 瘦客户端连接到 Ignite，我没有为每个查询创建一个新客户端，而是重复使用了一个全局客户端。
我想事情是这样的： 由于客户端在发送一个查询执行后是 运行 第二个查询，因此未检索到第一个查询的输出。这阻止了 Ignite 运行 第二个查询和检索第一个查询的输出。
如果对 Ignite 有更多了解的人可以纠正我或告诉我我是否正确，将不胜感激。