Masahiro Fujiwara
Masahiro Fujiwara
we have supports for multiple OS images on Wakame-vdc. but there was no standard way to build them.
Add feature to resize volume. API Change: ``` PUT /volumes/vol-xxxx/resize?size=99999 ``` - `size` is absolute byte size to extend the volume. - only supports to grow operation. Target Hypervisor: -...
`PUT /insta/ces/i-xxxx/backup?all=true` performs to take backup for all associated disks. but it fails with 500 and the instance can not be terminated because the validation for those incomplete backup objects...
The moved instance can not continue log monitoring. hva forgets to update the contents of /var/lib/wakame-vdc/fluent.conf on the destination side.
hva.log ``` 2015-03-10 16:26:56 JobContext thr=JobWorker[0/1] [INFO]: Job start 78f281bc72e9ce4b46cdf262c4dc7354650492c2 (Local ID: 78f281bc72e9ce4b46cdf262c4dc7354650492c2)[ delete_alarm ] 2015-03-10 16:26:56 JobContext thr=JobWorker[0/1] [ERROR]: Job failed 78f281bc72e9ce4b46cdf262c4dc7354650492c2 [ delete_alarm ]: {:message=>"undefined method `instances' for...
Assume we have two instances, instance A and instance B, and we tried to swap a volume on shared disk from instance A to B. The attach succeeds on instance...
disk backup operation is expected to run for `halted` instance to freeze the disk image. The instance state validation is missed on API so the request is forwarded to agent...
API should check the associated volumes of the instance that all they are shared volume. then the request will return error for the local volume instance.
/data/instances holds the information files for the respective instances. Once HA triggers the instance move due to the hardware failure, the failure side still keeps those information files. It makes...