Cheng Pan

I tried m5.large using CoreOS-stable-1632.3.0-hvm (ami-692faf11) with kernel 4.14.67. The test is keep attaching and detaching a volume to an existing instance repetitively in a loop. But I couldn't reproduce...

I am able the reproduce the issue by following steps [here](https://bugzilla.redhat.com/show_bug.cgi?id=1658237#c46) My setups are: * kernel - 4.14.88-88.76.amzn2.x86_64 * amazon linux 2 * m5.2xlarge * gp2 200GiB Also verified `nvme_core.timeout`...

Any plan to merge this fix? as it has been outstanding for more than 4 years, and ppl are wasting tons of time in tracing down such issues.

> The way that we implemented it in the GCP PD driver is to encode zone/region information in the volume id. Yep. that's something could be achieved at driver side....

Any timeline for fix this issue? invoking pre-exit after `cancel-grace-period` is necessary for the correctness of the system

Extending grace period is what we do now. However, we cannot ensure all the job are handling signals correctly, as we are running a multi-tenant environment

Any updates to this PR? And which bazel version could potentially have this merged in? For some context, I’ve been evaluating —remote_download_symlink_template to reduce disk consumption on the client machine...