Dear:

Wechat group "Apache Linkis(incubating)  community Development group "chat 
records are as follows:  微信群"Apache Linkis(incubating)  社区开发群"的聊天记录如下:



 

—————  2022-11-9  —————

Jason  18:05




Ladies and gentlemen, please ask a question: What causes the frequent 
occurrence of spark engine RequestEngineLock time out? No relevant issue was 
received







peacewong@WDS  18:49




I need to check spark ec's log for this







Jason  19:06




ec log, every time when tryLock can not go







Jack  19:09




Delete the database execution information and try again







Jack  19:09




Back up First







Jason  19:12




What execution information do you mean?







Jack  19:13




linkis_cg_manager_lock







Jason  19:16
















Jason  19:16




The database has no lock record







peacewong@WDS  19:17




Not this lock, this should have been fixed, you can look at jstack, it should 
be stuck syncing.







Jason  19:17




good I think it's also because I haven't typed this log







Jason  19:18




How many issues are fixed? I can't find it.







peacewong@WDS  19:18




I'll send it later







Jason  19:18




Ok, thanks a lot







peacewong@WDS  19:42




You can view the specific jstack of spark ec







Jason  19:43




Ok, hold on







Jason  19:51




You can look at that







peacewong@WDS  19:54




This one doesn't have a line break. It's not very nice







Jason  19:55




Take a look at this







peacewong@WDS  20:06




It looks like this one has an exception, the executor is still executing, and 
the lock should not have been requested.







peacewong@WDS  20:06
















Jason  20:14




good




Jason  18:05

各位大佬,请教个问题:spark引擎经常出现 RequestEngineLock time out  是什么原因导致的呢? 没有收到相关的issue




peacewong@WDS  18:49

这个要看看spark ec那边的日志下




Jason  19:06

ec 的日志,每次都是tryLock的时候走不动了




Jack  19:09

数据库的执行信息 删掉 再试试 




Jack  19:09

先备份




Jason  19:12

你是指什么执行信息?




Jack  19:13

linkis_cg_manager_lock




Jason  19:16







Jason  19:16

数据库没锁记录




peacewong@WDS  19:17

不是这个锁,这个应该已经修复了,你可以看看jstack,应该是同步卡住了。




Jason  19:17

好的。我感觉也是因为我这个日志都没打




Jason  19:18

修复的issue 是多少呀?我没找到




peacewong@WDS  19:18

晚点我发下




Jason  19:18

好的,多谢




peacewong@WDS  19:42

可以给下spark ec具体的jstack看看下




Jason  19:43

好的,稍等




Jason  19:51

可以看下




peacewong@WDS  19:54

这个没有换行,不太好看




Jason  19:55

看下这个




peacewong@WDS  20:06

看起来是这个有异常,executor还在执行,不应该被请求锁来着。




peacewong@WDS  20:06







Jason  20:14

好的



















--

Best Regards
------
康悦 ritakang 
GitHub:Ritakang0451
E-mail:rita0...@163.com

Reply via email to