John Sully
John Sully
I'm a bit suspicious of the CI failures will need a deeper dive >{12008:12023:S 26 Mar 2024 19:57:42.694 # Internal error in RDB reading offset 9610080, function at rdb.c:3529 ->...
main is the branch we should target
I cannot automatically retarget the correct branch, can you update on a branch based off of main.
Do you have repro steps to trigger? We will need to add a new test since I finally have all tests passing in the async_flash branch
That line above, “failed to evict keys, falling back to flushing keys” is likely the problem or closely related to it. Make sure we are setting m_numexpires to 0 when...
That emergency deleting all cached values is pretty rare (we don't run in situations that would hit that). So my money is on the bug being related to ir.
@jianjun126 How come you disable mmap? This results in large buffers getting created and free'd which can exacerbate this problem
I've been told this reproduces still even with your other fix @MalavanEQAlpha
Thanks @keithchew for confirming its fixed there. Sorry we worked too independly on this. I have one remaining show stopper with FLASH + FASTSYNC we are hitting which has stopped...
For those who are diving into the code on this thread would it be helpful to have a community meeting to share notes?