Hello,
I am seeing an issue with libnfs V2.0 where nfs_lutimes_async is giving us the same callback twice when backend is not reachable.
lutimes(/path) failed: ret=110, Connection timed out [Socket failed with POLLHUP]
We call nfs_lutimes_async(nfs, path, times, callback, private_data)
As soon as the connection is killed and we try to access a file, we get the callback where we freeup the private_data. Then we get the callback again and here we have no clue how to handle this.
I had a suspicion on the autoreconnect code but I have disabling it on my setup by setting rpc->auto_connect = 0
Has anyone seen this issue, any suggestions ?
Thanks,
Bharat