<div dir="ltr">Ok I will try this right away. <div><br></div><div>It's a typo. I meant hundreds. </div><div><br></div><div>I did a couple of different tests. But in the end, I noticed that the error happens event with a simple</div>
<div>dd if=/dev/zero of=/dev/vda</div><div><br></div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/12/13 Liu Yuan <span dir="ltr"><<a href="mailto:namei.unix@gmail.com" target="_blank">namei.unix@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">On Fri, Dec 13, 2013 at 04:33:18PM +0100, Hadrien KOHL wrote:<br>
> I sure can, but how can I print those information?<br>
><br>
> I didn't mention that earlier, but the problem does not arise before a<br>
> couple of hundreds of MB (even though it's a bit random).<br>
<br>
</div>you can try following patch and tell the output. what did you mean by 'hunreads<br>
of MB', I notice you just mkfs.ext4 and said then it would cause problem.<br>
<br>
Thanks<br>
Yuan<br>
<br>
diff --git a/block/sheepdog.c b/block/sheepdog.c<br>
index b4ae50f..2036fbf 100644<br>
--- a/block/sheepdog.c<br>
+++ b/block/sheepdog.c<br>
@@ -1774,6 +1774,7 @@ static int sd_truncate(BlockDriverState *bs, int64_t offset)<br>
<div class="im"> unsigned int datalen;<br>
<br>
if (offset < s->inode.vdi_size) {<br>
</div>+ printf("%ld, %ld\n", offset, s->inode.vdi_size);<br>
<div class="HOEnZb"><div class="h5"> error_report("shrinking is not supported");<br>
return -EINVAL;<br>
} else if (offset > SD_MAX_VDI_SIZE) {<br>
</div></div></blockquote></div><br></div>