-
Notifications
You must be signed in to change notification settings - Fork 1.5k
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
Dgraph Crashes Randomly While Running In GCE #1020
Comments
@tzdybal : Can you look into it immediately? |
It seems that bug is caused by a race condition, when multiple filters are applied concurrently to the same subgraph. I'll try to reproduce this. @willcj33 could you provide the problematic query? |
Detailed problem description:
Partial results (UID lists) are shared by all filters. At each level filters are executed concurrently. The problem occurs when To eliminate this data race, algorithms should work on separate copies of pointers to UID list (fortunately copying entire lists is not required). |
In #1020 there was data race that resulted in the crash (where `and` and `not` filters was executed simultaneously). This commit removes data race between two `not` filters running in parallel. It makes sure that results of each `algo.Difference` execution are stored in separate (newly allocated) slice.
In #1020 there was data race that resulted in the crash (where `and` and `not` filters was executed simultaneously). This commit removes data race between two `not` filters running in parallel. It makes sure that results of each `algo.Difference` execution are stored in separate (newly allocated) slice.
In #1020 there was data race that resulted in the crash (where `and` and `not` filters was executed simultaneously). This commit removes data race between two `not` filters running in parallel. It makes sure that results of each `algo.Difference` execution are stored in separate (newly allocated) slice.
The text was updated successfully, but these errors were encountered: