On an embedded system.
Two partitions are dedicated for the rootfs. Partition A and B.
When the system is running on partition A I use casync to update the partition B with the latest firmware and reboot on partition B. The partition used for the running system is mounted in read-only mode.
To limit the downloaded data I use the current partition / as a seed.
While performing some tests I noticed that if I remove one file on the partition B, a file which is still in partition A, then 2 chunks are always downloaded. I thought the file would be copied from the partition A.
On an embedded system. Two partitions are dedicated for the rootfs. Partition
A
andB
.When the system is running on partition
A
I use casync to update the partitionB
with the latest firmware and reboot on partitionB
. The partition used for the running system is mounted in read-only mode.To limit the downloaded data I use the current partition
/
as a seed.While performing some tests I noticed that if I remove one file on the partition
B
, a file which is still in partitionA
, then 2 chunks are always downloaded. I thought the file would be copied from the partitionA
.If I try the same but the system is running from an sdcard so I can mount and use the partition A as a seed, using the command:
then no chunk is downloaded.
The store has been created with the command run in a fakeroot environment so the UIG:GID have been changed to 0.
The options
--verbose
and--level debug
are not helpful is this case. Is there a way to know what files are stored in the downloaded chuncks ?