Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

task_result is null #140

Closed
yuyang0 opened this issue Oct 29, 2018 · 11 comments
Closed

task_result is null #140

yuyang0 opened this issue Oct 29, 2018 · 11 comments

Comments

@yuyang0
Copy link

yuyang0 commented Oct 29, 2018

task completes successfully, but the task_result is null, my gremlin command is like this g.V().has(xxx).outE().toList, I think the result should not be null, even in case of empty result, it should return [], I am not sure if it is a bug in hugegraph.

@javeme
Copy link
Contributor

javeme commented Oct 29, 2018

Yes, the result should not be null. Is it possible to provide information about the task(response of task-get api) or steps to reproduce? thank you

@yuyang0
Copy link
Author

yuyang0 commented Oct 29, 2018

@javeme thanks for your reply, I can't reproduce the problem stably, I need some time

@yuyang0
Copy link
Author

yuyang0 commented Nov 2, 2018

@javeme 找不到原因,没法稳定的复现,我的代码如下

    String cmd = "g.V().hasLabel('user').hasId(within(['xxxx', ...]).outE().as('e').label().as('label').select('e').values('count').as('count').select('e').outV().id().as('source').select('e').inV().id().as('target').select('label', 'source', 'target', 'count').toList()"
    HugeClient hugeClient = new HugeClient("xxx", "hugegraph", 600);
    long taskId = hugeClient.gremlin().gremlin(cmd).executeAsTask();
    hugeClient.task().waitUntilTaskCompleted(taskId, 300);
    Task t = hugeClient.task().get(taskId);
    $.msg("+++++++++++++++++");
    $.msgf("%d: %s\n", t.id(), t.result());

within中是一个id列表,大概1500个id, 这个代码大概跑三到四处就会出现一次null,但是如果用curl对应的task发现有result,而且接下来使用java的客户端获取这个task,发现也有result,总之就是用这个代码跑得时候会出现null,服务器日志没发现异常

@javeme
Copy link
Contributor

javeme commented Nov 5, 2018

@yuyang0 Thanks for your information, it looks like there is race condition when getting task status, we will find and fix it, thank you!

@yuyang0 yuyang0 closed this as completed Nov 5, 2018
@yuyang0 yuyang0 reopened this Dec 6, 2018
@yuyang0
Copy link
Author

yuyang0 commented Dec 6, 2018

0.8并没有修复这个问题 @javeme

@javeme
Copy link
Contributor

javeme commented Dec 6, 2018

@yuyang0 请问你是使用的hbase后端吗?如果是的话可能与hbase的时间戳问题有关,请参考:https://issues.apache.org/jira/browse/HBASE-2256

类似问题 #230

作为一个规避方案,试试直接拿到task是否可以:
Task t = hugeClient.task().waitUntilTaskCompleted(taskId, 300);

@yuyang0
Copy link
Author

yuyang0 commented Dec 6, 2018

我就用的rocksdb @javeme

@javeme
Copy link
Contributor

javeme commented Dec 7, 2018

@yuyang0 如果是rocksdb的话那可能是hg本身的bug,看看能不能找到一个稳定的复现过程。非常感谢反馈

@yuyang0
Copy link
Author

yuyang0 commented Dec 10, 2018

@javeme 我如果能稳定的复现,我就自己修复了,弄得真心累,还遇到了过好几个问题,比如升级0.8,数据库server直接卡死,内存占用暴增(128G的服务器,hugegraph用掉了124G,如果服务器有更多空闲内存肯定还会用掉更多),结果这样卡了半个小时,直接挂了,删掉0.7创建的数据似乎又好了,还有其它小问题一堆,这个开源版本确实不是一个成熟可用的数据库

@yuyang0 yuyang0 closed this as completed Dec 10, 2018
@javeme
Copy link
Contributor

javeme commented Dec 10, 2018

@yuyang0 建议在遇到问题的时候给我们反馈错误日志及操作流程,我们希望能帮助你解决这“一堆”问题。

@yuyang0
Copy link
Author

yuyang0 commented Dec 11, 2018

出问题查日志我还是懂得,关键是日志没什么有用的信息,我说很多小问题真的不是乱说,比如说我折腾接近一个月,至少遇到过三次server莫名其妙的挂掉,日志没什么异常,我甚至都没做什么操作,但是就是挂掉了,你像这种问题我真的没法反馈,当然我这里说的只是这个开源版本,我最开始其实对hugegraph期望挺高的,但是现在看来确实不成熟,不管怎么说,还是希望hugegraph越做越好,至少可以多一个选择 @javeme

VGalaxies pushed a commit that referenced this issue Aug 3, 2024
- add builderCallback param for client to add custom config
- add params connectTimeout and readTimeout to instead the param time
- update version to 1.3.0

---------

Co-authored-by: imbajin <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants