Merge commit 'refs/top-bases/linux-debug-zerocopy' into linux-debug-zerocopy
This commit is contained in:
commit
4cee20b087
|
@ -0,0 +1,29 @@
|
||||||
|
SUMMARY OF MAJOR KNOWN PROBLEMS IN v0.4.0 (Development Release)
|
||||||
|
|
||||||
|
- 'zpool create' hangs in the create ioctl() when initializing a new pool
|
||||||
|
backed by loopback devices. The lo-raid0 configuration easily recreates
|
||||||
|
this issue. I currently suspect a problem in vdev_disk.c implementation
|
||||||
|
is causing this, but I have not yet looked to closely. On the surface
|
||||||
|
things appear to be fine when creating the pool with real device or files.
|
||||||
|
|
||||||
|
./zpios.sh -c lo-raid0 -t tiny -v
|
||||||
|
|
||||||
|
- SPLError: 10167:1968:(spl-kmem.c:1286:spl_kmem_cache_reap_now())
|
||||||
|
ASSERTION(skc->skc_magic == SKC_MAGIC) failed
|
||||||
|
|
||||||
|
The above assertion is overserved when perform more IO than can be fully
|
||||||
|
cached by the ARC. The Linux VM applies back pressure to the slab which
|
||||||
|
in turn detect what appears to be memory corruption. I've seen a few
|
||||||
|
flavors of this so far, so I'm not yet convinced this is actually an
|
||||||
|
issue with the SPL slab. It may just be the most common victim, more
|
||||||
|
investigation is needed. It is also possible the new untested vdev_disk.c
|
||||||
|
is to blame. A lot of work is needed here.
|
||||||
|
|
||||||
|
- SPLError: 7324:1224:(dnode.c:304:dnode_create()) VERIFY3(0` >= 1`)
|
||||||
|
|
||||||
|
When enabling debugging in ZFS with the --enable-debug configure option
|
||||||
|
we always trip the following VERIFY. This issue was present in the
|
||||||
|
previous 0.3.3 release and was avoided simply by leaving debugging
|
||||||
|
disabled until it could be explained. Well it has not just gone
|
||||||
|
always with the update to b105 so we need to run it to ground and
|
||||||
|
explain what is going on.
|
Loading…
Reference in New Issue