[Libguestfs] Re: nbdkit crashes on attempt to retrieve extent with len of 2^32-1
On Tue, Apr 22, 2025 at 03:41:43PM -0500, Eric Blake via Libguestfs wrote:
On Tue, Apr 22, 2025 at 10:08:01PM +0300, Nikolay Ivanets wrote:
> nbdkit crashes when the client is trying to get extents with len=2^32-1.
> Here is client code (nbdsh):
>
> h.add_meta_context("base:allocation")
> h.connect_uri("nbd://localhost:10809/disk0-flat.vmdk")
Can you post the disk0-flag.vmdk image (or better, instructions for
how to create it?)
>
> def f(metacontext, offset, e, status):
> print(e)
>
> h.block_status(2**32-2, 0, f)
> [4294967295, 0] <--- OK
At any rate, it looks like you have an export that is all data up to this length...
> Server prints:
>
> nbdkit: file.8: debug: file: extents count=4294967295 offset=0 req_one=0
...and that you were using the file plugin to read it. So let's see
what happens if I create a file larger than 4G to try and reproduce:
$ dd if=/dev/urandom of=myfile bs=1G count=5
5+0 records in
5+0 records out
5368709120 bytes (5.4 GB, 5.0 GiB) copied, 15.5435 s, 345 MB/s
$ ./nbdkit -fv file myfile &
$ nbdsh --base-allocation --uri nbd://localhost
<nbd> def f(m, o, e, s):
... print(e)
...
<nbd> h.block_status(2**32-2, 0, f)
[4294967295, 0]
<nbd> h.block_status(2**32-1, 0, f)
server crashed
Okay, I can reproduce it with the file plugin, but not with the memory
plugin. Now working on root cause.
--
Eric Blake, Principal Software Engineer
Red Hat, Inc.
Virtualization: qemu.org | libguestfs.org