Erlend Faxvaag

Still get `ExtraData: unpack(b) received extra data.`, but after trying @KirannBhavaraju suggestion it worked! Only thing I did was to remove the`chunk_size=xxxx` argument. ``` client = InfluxDBClient(blah blah) result =...

I'm having the same problem here. What do you mean by root data object in the query string @rose-a?

The error is gone, but I couldn't get this working properly. Not sure if this is an MMS-related problem or a user error somewhere, let me know if that was...

Tried the code in a working environment now and `watch` works fine there. Still can't get this working with tests and MMS. MMS works fine with the current configuration for...

Hmm, we start the instance using the address in another package, maybe that has something to do with it? Below is how we setup stuff (renamed some vars and functions),...

Thanks for taking a look. Unfortunately, changing the `storageEngine` did not help. When using `wiredTiger` it also hangs when calling watch. The default storageEngine hangs when writing or reading from...

I ended up just disabling watch in the tests to bypass the problems I had. I tried some sleeps here and there, but no luck. I tried changing the storageEngine,...

Thank you for your response @avolkov1 and @bzamecnik! Highly appreciated. `nvidia-smi topo -m` gives: ``` GPU0 GPU1 GPU2 GPU3 CPU Affinity GPU0 X PHB SOC SOC 0-15 GPU1 PHB X...

@avolkov1 I did as you said, checked if `mnist_tfrecord_mgpu.py` was up to date and it was. When running both of the two `CUDA_VISIBLE_DEVICES=0,1 python ./examples/cifar/cifar10_cnn_mgpu.py --mgpu --epochs=5` programs I get...

@avolkov1 I ran `cifar10_multi_gpu_train.py ` and I think it worked. Examples/Sec increased with numbers of GPUs. Sec/Batch increased from 1 GPU to multiple, but did not increase from two to...