I installed the lmdb-utils package and ran:

mv ~/.local/share/nheko/nheko/db-$NAME ~/.local/share/nheko/nheko/db-$NAME.old

mkdir ~/.local/share/nheko/nheko/db-$NAME

lmdb_copy -c $BACKUP/.local/share/nheko/nheko/db-$NAME \
 ~/.local/share/nheko/nheko/db-$NAME

Then I ran nheko and it worked.

https://github.com/bmatsuo/lmdb-go/issues/131
https://github.com/pantsbuild/pants/issues/18726

This is a known issue with lmdb, this seems to be clearly a bug in the lmdb 
library as corrupted data should be expected on occasion and not lead to a 
SEGV.  It's OK for a library to abort a load because of corrupted data and 
it's OK for an application to abort because the data couldn't be loaded.  But 
it's not OK to SEGV.

Could Nheko handle this without a SEGV?

If you decide to just reassign this but to liblmdb0 then that would be 
acceptable to me.  If you decide to create a new bug report about liblmdb0 and 
leave this bug for the nheko reaction to it then that would be acceptable too.

-- 
My Main Blog         http://etbe.coker.com.au/
My Documents Blog    http://doc.coker.com.au/

Reply via email to