You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running ipfs-cluster-service daemon on a Raspberry Pi 3 B with CRDT consensus fails due to badger not being able to mmap memory.
$ ipfs-cluster-service daemon --consensus=crdt
06:07:40.790 INFO service: Initializing. For verbose output run with "-l debug". Please wait... daemon.go:48
error creating datastore: Mmap value log file. Path=/home/pi/.ipfs-cluster/badger/000000.vlog. Error=cannot allocate memory
I found this badgerdb issue which is resolved and if I ran that main.go on the RPi, it worked but if I changed the options.ValueLogLoadingMode back to the default, which is to use options.MemoryMap, that example program fails.
@hsanjuan when you get the chance could you please confirm this?
The fix, if this is a problem, is to allow passing through the configuration of value for options.ValueLogLoadingMode to badger.
The text was updated successfully, but these errors were encountered:
@lanzafameValueLogFileSize is set to 1GB by default. Can you try reducing it? How much RAM is there in these Raspberry Pis ? Ideally we would automatically handle this in systems with lower memory somehow. We can always expose [needed] badger options though.
Additional information:
Running ipfs-cluster-service daemon on a Raspberry Pi 3 B with CRDT consensus fails due to badger not being able to
mmap
memory.I found this badgerdb issue which is resolved and if I ran that main.go on the RPi, it worked but if I changed the
options.ValueLogLoadingMode
back to the default, which is to useoptions.MemoryMap
, that example program fails.@hsanjuan when you get the chance could you please confirm this?
The fix, if this is a problem, is to allow passing through the configuration of value for
options.ValueLogLoadingMode
to badger.The text was updated successfully, but these errors were encountered: