2006-08-31 10:08:47

by Akinobu Mita

[permalink] [raw]
Subject: [patch 0/6] RFC: fault-injection capabilities (v2)

Changes from last version

- use lightweight random simulator instead of get_random_int()
- added per queue filter for disk IO failures
(/sys/blocks/sda/sda1/make-it-fail, /sys/blocks/sda/make-it-fail)
- added process filter
(/debug/{failslab,fail_page_alloc,fail_make_request}/process-filter,
/proc/<pid>/make-it-fail)

---
This patch set provides some fault-injection capabilities.

- kmalloc failures

- alloc_pages() failures

- disk IO errors

We can see what really happens if those failures happen.

In order to enable these fault-injection capabilities:

1. Enable relevant config options (CONFIG_FAILSLAB, CONFIG_PAGE_ALLOC,
CONFIG_MAKE_REQUEST) and runtime configuration kernel module
(CONFIG_SHOULD_FAIL_KNOBS)

2. build and boot with this kernel

3. modprobe should_fail_knob

4. configure fault-injection capabilities behavior by debugfs

For example about kmalloc failures:

/debug/failslab/probability

specifies how often it should fail in percent.

/debug/failslab/interval

specifies the interval of failures.

/debug/failslab/times

specifies how many times failures may happen at most.

/debug/failslab/space

specifies the size of free space where memory can be allocated
safely in bytes.

/debug/failslab/process-filter

enable process filter.

5. see what really happens.


2006-08-31 17:29:15

by Josh Triplett

[permalink] [raw]
Subject: Re: [patch 0/6] RFC: fault-injection capabilities (v2)

On Thu, 2006-08-31 at 19:07 +0900, Akinobu Mita wrote:
> This patch set provides some fault-injection capabilities.
>
> - kmalloc failures
>
> - alloc_pages() failures
>
> - disk IO errors
>
> We can see what really happens if those failures happen.

Looks very useful for testing error paths; nice work.

Should this perhaps taint the kernel when used?

- Josh Triplett


2006-08-31 18:31:52

by Kyle Moffett

[permalink] [raw]
Subject: Re: [patch 0/6] RFC: fault-injection capabilities (v2)

On Aug 31, 2006, at 13:29:20, Josh Triplett wrote:
> On Thu, 2006-08-31 at 19:07 +0900, Akinobu Mita wrote:
>> This patch set provides some fault-injection capabilities.
>>
>> - kmalloc failures
>>
>> - alloc_pages() failures
>>
>> - disk IO errors
>>
>> We can see what really happens if those failures happen.
>
> Looks very useful for testing error paths; nice work.
>
> Should this perhaps taint the kernel when used?

It shouldn't; these are all failures that could quite possibly happen
during normal operation even without this enabled, they're just a few
orders of magnitude less likely (in most situations).

Cheers,
Kyle moffett