the spool directory also contains log files from rocksdb, those might also tell you something interesting
the rocksdb logs are quite long and verbose, and I’m doubtful that they can be effectively read via screenshot.
The size of the rocksdb seems unusual. Did you experience unusual circumstances leading up to this? eg: running out of disk space? Crazy load?
Rocksdb will generally manage and compact the sst files down as needed and I wouldn’t expect to see one so large when there are no messages (or just 78 that it found according to the screenshot)
If you restart the instance, does it compact the spool? And does it report that there are still some messages?
I have .sst logs for the date of 24 October
I hope you are referring to 78 message that of July
the count of .sst logs files are around 865
there should be a file named LOG in the spool directory. It is a human readable (but verbose!) log
I have
read through it to see if it is yelling about something bad
any specific keywords we are looking for?
no idea TBH
it’s mostly meaningful to the rocksdb developers
but if it is complaining about something, that might be useful to see. I’d suggest looking at the bottom
It has information on compaction stats, which may be insightful in this sort of situation
please have look
hoping you will find something, as I am unable understand
So, I’m not going to dig into a screenshot of this file right now
if you want me to dig in deeper, we can talk about purchasing support