If you suspect this could be a bug, follow the template.
What version of Dgraph are you using?
v1.0.14
Have you tried reproducing the issue with latest release?
Since v1.0.14 is the latest verified release, yes.
What is the hardware spec (RAM, OS)?
64 GB DDR4, 2 x 4 TB SATA Enterprise hard drive 7200 rpm , Intel® Core™ i7-6700 Quad-Core - Ubuntu 18.04
Steps to reproduce the issue (command/config used to run Dgraph).
Followed the 'Getting Started' guide. Set up 20480 MB's of RAM for the alpha. Then I start inserting thousands of data concurrently. After some point, insertions stack up and alpha crashes with the error :
'Error=Server overloaded with pending proposals. Please retry later. Retrying...
Killed'
The thing is I am not even inserting anymore and it acts like this for over 5 days, I think it is still trying to insert the data. Also when alpha is on, it keeps giving this error : 'A tick missed to fire. Node blocks too long!'
Expected behaviour and actual result.
I am not able to even run a basic query, it timeouts with the error : '": context deadline exceeded"'
Are we able to abort all the pending proposals?
Have you tried using the Dgraph Live Loader? It will run mutations by batching 1000 triples per mutation, with 10 concurrent mutations at a time. You could also do something similar in your own loading.
I have the same problem,How can I resolve them. I used Spark and dgraph4j to import data
I used v1.1.1, but the problem is still existed.the data from kafka,and I use java client to mutate(1000 node per mutation)
This sounds like a deadlock. We fixed this issue via #4453. Closing this issue. Feel free to reopen.
I have the same problem when using version v20.07.0.