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
Guarantee failed: [get_safety_boundary() >= num_elements] #7005
Comments
The bug is most likely caused by there being corrupt data on disk. This could have been caused by a hardware bitflip or possibly a RethinkDB bug that wrote incorrect data, either because of a bug in code that manages blocks on disk, or the code which serialized a ReQL object into a byte-stream. It turns into a crash when a particular object gets referenced. Maybe it only occurs if a specific replica is used to retrieve the data. For your sake, it would be good to figure out what document has this problem. I guess the first step would be to figure out what query caused the crash. It seems to be a group-by query. If this theory is correct, enumerating every object in the database would likely trigger the crash. If can do this without crashing, in such a manner that the data gets retrieved from every replica, then this explanation is probably wrong. Reformatting the backtrace:
|
Thank you @srh for your reply ! I will review our source code in order to check for a group-by query and see if I can do as you said. I'll back to you. |
We have a single instance of a rethinkdb 2.4.1~0bionic (CLANG 6.0.0 (tags/RELEASE_600/final)) on a Ubuntu 18.04.6 LTS which crashes randomly like 3 times in January, 2 times in December, never in Novembre, 2 times in October and so forth.
Here is the crash logs:
Restarting the DB solves the problem.
Is there anything I could do in order to help you track that bug ?
The text was updated successfully, but these errors were encountered: