it fixes the issue The Ceph bufferlist structure is not actually NULL-terminated, even if you invoke the c_str() function. (Sorry about that bad design!) So when you invoke bufferlist::c_str() and give it to a string constructor, it is going to be of arbitrary length until the first random NULL byte following it. You can fix this by getting the length and setting it explicitly when constructing the string.
I wish this helpful for you It has been happened because in the ceph.conf you must set mon ip in the public network not in the private. And I had mon ip : 192.168.57.101 (which is the private) but public network was: 10.0.2.0/24.
Consisten results with Multiple runs of h2o deeplearning
To fix this issue Deeplearning with H2O will not be reproducible if it is run on more than a single core. The results and performance metrics may vary slightly from what you see each time you train the deep learning model. The implementation in H2O uses a technique called "Hogwild!" which increases the speed of training at the cost of reproducibility on multiple cores. So if you want reproducible results you will need to restrict H2O to run on a single core and make sure to use a seed in the h2o.deeplearning call.
Ceph-deploy is not creating ceph.bootstrap-rbd.keyring file