Atin Mukherjee

peer state enablement API

2

Peer can be enabled/disabled for maintenance, if Peer state is disabled then that peer should not be used for Volume creation or for adding new entities. This seems to be...

gluster-block integration with GD2

4

With GD2 we need to come up with an integration method on how gluster-block can be setup.

Capture all incoming commands in a logfile

10

We need something similar to what GD1 provides in cmd_history.log where all incoming CLI commands on that node are logged.

Document the etcd and other configuration cleanup steps

2

I was trying to spend some time in (re)setting up a GD2 cluster of 3 nodes in my docker containers and peer probe failed 3/3 times randomly. I suspect this...

test node scalability support limit with GD2

2

This is a test effort to check with current form of GD2 what's the initial support limit we can claim w.r.t node scaling. This test effort should also focus on...

Mechanism to suppress repetitive logs

1

As gluster clients have the logic of attempting to reconnect to the server after every 3 seconds, there's a critical need to have a wrapper log API over logrus to...

glustershd needs to be restarted for volume start/stop operations

1

Based on code reading, it was found that glustershd wasn't restarted on volume start/stop operations which was resulting shd process to not having the latest volfile and it was continuing...

Fix spurious socket connect failure issue in brick multiplexing stop code path

1

While sending parallel volume requests some of the volume stop operations were failing with the brick socket file reporting EOF.

stale brick process when volume stop operations are done in parallel in brick multiplexing mode

8

During stopBricks () , if we happen to send the volume stop requests in parallel, GD2 might fail to determine the last brick in the proc entry because - Given...

parallel volume deletion requests fail with device or resource busy for some of the PVs

1

On a GCS deployment (with brick multiplexing enabled), when PV deletion requests are sent on parallel, some of volume delete requests fail with device or resource busy error which means...