Takashi Menjo

@fuku-ys: Thank you for your reporting. I can reproduce this issue on 1.0_65_gb6d64b2 and Ubuntu 16.04.

I think this is because of lack of **atomicity**. In brief, 'dog vdi snapshot -s _tag_' procedure consists of three steps below: 1. Get working VDI's ID (now it's _V_)...

I got an idea that sheep simply rejects SD_OP_NEW_VDI in case that it is to create snapshot and a given _vid_ is also snapshot. However, it is useless because this...

## Problems: 1. atomicity 2. telling intent of SD_OP_NEW_VDI to sheep ## TODO: ### Protocol: - For [1], request body of SD_OP_NEW_VDI should contain of both name and tag -...

Thanks for your report. I try to reproduce this on Ubuntu 16.04 (because I have no Debian 7 environment), but I cannot. ### script ```sh #!/bin/bash export LANG=C LC_ALL=C set...

I walked through open issues and pull requests. I think the features (or, at least, their current implementation) listed up below should be removed: - object cache (#195, #74, #64)...

Using 1.0_65_gb6d64b2, I cannot reproduce this issue. I run a testing script below but it continued looping 300 times. ### testing script ``` sh #!/bin/bash export LANG=C LC_ALL=C set -ueo...

> In a case of single drive node, [...] Using plain_store, the node leave_cluster() on failure of the disk, then goes down [1]. So the node will not continue running...

**Please don't merge this PR yet**. This is not tested yet. However, comments for log message are very welcome.

Some functional tests failed with diff like below: ```patch -Cluster status: running, auto-recovery enabled +Cluster status: running, auto-recovery disabled ``` I suspect that this patch causes the failure. I will...