-
Notifications
You must be signed in to change notification settings - Fork 871
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
Exception in thread 'OrientDB WAL Flush Task' #9270
Comments
Hi, @quanghaisoft what is the WAL page size in your environment? You can find size of the WAL page in the log. Do you set any non-default settings? Thank you for the bug report. |
thanks you for your answer , this is logs, recent day, this error not appear 2020-06-03 17:01:01:421 INFO Allocation of 972800 pages. [OEngineLocalPaginated] |
We have written a lot of data simultaneously |
Hi @quanghaisoft . I am not sure that I understood you. Do you mean that now you can not reproduce this error on your tests? |
This error does not show again. We are running in production, so I cannot test it. If it appears I will open this issue |
this bug in my production so we copy data and rebuild cluster immediately.if it appears again,we will send it to you |
I see thank you, I will try to reproduce with settings which you sent to me. |
Hi @quanghaisoft what amount of RAM do you have on the server ? |
we using more than 150G ram , CPU 24 core on node , with 6 node in cluster |
when we using multi write upsert to one vertex in 3 master node, it slow down , can you help me a recommend Config to active fast upsert with bigdata, avoid lock record when confict happen,avoid not response .etc |
orient-server.log.1_master_node.txt |
@quanghaisoft issue which you had with WAL was caused by integer overflow. It is already fixed. I can not help you with the issue with distributed storage because I am working on another set of components. Could you please create a separate issue for your problem? |
I have just see this error again, i you give me a snapshot of 3.0.31 version Jun 19 18:45:00 ograph06-76 server.sh[29054]: Exception in thread 'OrientDB WAL Flush Task' |
in others node have this error: Error during WAL writing |
I see, interesting. I will fix another issue and then will be back to yours. |
@quanghaisoft about directory |
I'm sorry, my server is running in production, so I copied the dir database in a master node to all nodes and restarted the distributed cluster, all I had was the log. dir |
one server full of log file with text: |
I did not get it. Did you copy file system data under live server and put it under the replicas or did you shutdown replicas? The correct way to do that is to remove broken databases from nodes and start "empty" replicas, the system will copy all the data correctly. |
I have 5 servers 71,72,73,74,75,76, i writing data to a master node 71 |
I''m sorry this error is old,it is last 6 day ago |
@Laa Please help me with this error , can you give me a .jar at 3.0.31 thanks a lot Jun 19 18:45:00 ograph06-76 server.sh[29054]: Exception in thread 'OrientDB WAL Flush Task' |
@quanghaisoft you can download https://orientdb.org/download 3.0.32 changes are minor only small fixes in few components. |
@quanghaisoft could you send me log directory from the 76 node of the cluster ? |
log.zip |
OK thank you |
I have upgraded to version 3.0.32 and am watching, if I run a problem again I will notify you, thanks a lot |
@quanghaisoft what version of JDK do you use? |
Hi, @quanghaisoft I have introduced a small fix, which should resolve your issue. Is it possible for you to apply new build which I will send to you? |
Please send me new build you have fixed, thanks you very much we using java : java version "1.8.0_211" |
@quanghaisoft I am going to add more precautions, it will take couple of days, then I will provide you with new distribution. |
@quanghaisoft could you try this build https://drive.google.com/file/d/1wtQpyzbfeD3w07NpurG1wmGIxScggGXI/view?usp=sharing |
Thanks you very much, please give me access to file with email bumha... or xuanhuong... |
I have installed and when start service it throw an exception below 2020-07-17 22:22:43:223 INFO Limit of open files for disk cache will be set to 64000. [OEngineLocalPaginated] |
I see it was built under higher JDK, the same API but now returns the current instance of the buffer. I will rebuild it next week. |
OK thanks you very much . |
OK i will try it , thanks a lot |
I have upgraded to this build and it works fine, thanks you very much . |
Hi @quanghaisoft it could be one more reason why you experience this issue. It is inside the code of the distributed database. We are planning to port it soon to all the versions. I will let you know. |
Yes, thanks a lot, we'll waiting for it. |
Fixed. Should be delivered in next hotfix. Please reopen if the issue will still persist on your side. |
OrientDB Version: 3.0.31
Java Version: java version "1.8.0_211"
OS: ubuntu
Expected behavior
Upsert success, not raise any error
Actual behavior
Can not update/insert data any more
Steps to reproduce
My OrientDB system has over 100 classes with over 50GB data, setup in cluster mode with 6 node, on 6 different IP.
When I upsert big data into 1 master node orientdb in distributed ,
Error appear in logs
The text was updated successfully, but these errors were encountered: