Zhenyu Guo
Zhenyu Guo
That will definitely help~
Thanks, @linmajia. @shengofsun this may be related to the io-looper bug issue, would you please handle this issue together when you debug that problem? thx.
There is a currently a `flush` method in `aio_provider`. Can you guys do a survey to see whether flush and fsync are the same thing or not? `aio_fsync` is great...
Thanks, @shengofsun. There are two ways to ensure the data is really pushed to disk: `fsync` and use direct io. I guess in our case using direct IO is easier?...
Seems O_DIRECT + O_SYNC is good enough. But unfortunately, the buffer should be the same as the situation under Windows that it needs to be aligned. For your questions: -...
Let's try and see.
what is the original number (local rocksdb)?
@shengofsun Please also help look into this issue while I'm also trying to figure it out why.
The last fix is Commit 01eec618
This surfaces again last night. Last time when we fix it, it is because app_info is not appropriately set and is_stateful becomes false which causes incorrect update of the configuration.