Not an really good subject, I know but that''s kind of what happend.
I''m trying to build an backup-solution server, Windows users using
OSCAR (which uses rsync) to sync their files to an folder and when complete
takes a snapshot. It has worked before but then I turned on the -R switch to
rsync and when I then removed the folder with rm -rf it crashed. I
didn''t save what the error that the shell gave me but I know is
mentioned something about an colon (:).
I have no idea how to debug the Solaris kernel, I am both an ZFS and an Solaris
newbie, but I''ve managed to gather some information that I''ve
seen others pasting:
bash-3.00# mdb 0
Loading modules: [ unix krtld genunix specfs dtrace uppc pcplusmp scsi_vhci ufs
ip sctp arp usba uhci nca lofs zfs random sppp ptm nfs crypto ipc md cpc fcip
fctl fcp logindmux ]> ::status
debugging crash dump vmcore.0 (32-bit) from unknown
operating system: 5.11 snv_48 (i86pc)
panic message:
ZFS: bad checksum (read on <unknown> off 0: zio dbbf8880 [L1 ZFS plain
file] 4000L/400P DVA[0]=<0:6aa000:800> DVA[1]=<0:b80a7000:800>
fletcher4 lzjb LE contiguous birth=12316 fill=2 cksum=601eba85e8:
dump content: kernel pages only> ::spa
ADDR STATE NAME
da9f5ac0 ACTIVE backup> ::memstat
Page Summary Pages MB %Tot
------------ ---------------- ---------------- ----
Kernel 41647 162 32%
Anon 56673 221 44%
Exec and libs 11331 44 9%
Page cache 2963 11 2%
Free (cachelist) 11554 45 9%
Free (freelist) 4742 18 4%
Total 128910 503
Physical 128909 503> $C
db95bae4 vpanic(f9f95858, d7a91388, d7a913b8, f9f94660, 0, 0)
db95bc70 zio_done+0x122(dbbf8880)
db95bc8c zio_next_stage+0x66(dbbf8880)
db95bcac zio_wait_for_children+0x46(dbbf8880, 11, dbbf8a70)
db95bcc0 zio_wait_children_done+0x18(dbbf8880)
db95bcd8 zio_next_stage+0x66(dbbf8880)
db95bd10 zio_vdev_io_assess+0x11a(dbbf8880)
db95bd24 zio_next_stage+0x66(dbbf8880)
db95bd64 vdev_mirror_io_done+0x289(dbbf8880)
db95bd78 zio_vdev_io_done+0x25()
db95bdc8 taskq_thread+0x176(d430a290, 0)
db95bdd8 thread_start+8()
bash-3.00# zpool status -v
pool: backup
state: FAULTED
status: One or more devices could not be used because the the label is missing
or invalid. There are insufficient replicas for the pool to continue
functioning.
action: Destroy and re-create the pool from a backup source.
see: http://www.sun.com/msg/ZFS-8000-5E
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
backup UNAVAIL 0 0 0 insufficient replicas
raidz1 UNAVAIL 0 0 0 insufficient replicas
c0d1 FAULTED 0 0 0 corrupted data
c1d0 FAULTED 0 0 0 corrupted data
c1d1 ONLINE 0 0 0
The most important stuff from /var/adm/messages:
Oct 5 14:02:39 unknown sshd[5421]: [ID 800047 auth.crit] fatal: Read from
socket failed: Connection reset by peer
Oct 5 14:04:40 unknown unix: [ID 836849 kern.notice]
Oct 5 14:04:40 unknown ^Mpanic[cpu0]/thread=db95bde0:
Oct 5 14:04:40 unknown genunix: [ID 809409 kern.notice] ZFS: bad checksum (read
on <unknown> off 0: zio dbbf8880 [L1 ZFS plain file] 4000L/400P
DVA[0]=<0:6aa000:800> DVA[1]=<0:b80a7000:800> fletcher4 lzjb LE
contiguous birth=12316 fill=2
cksum=601eba85e8:408ce39c2ee8:172f0d844a20fa:5dfa8a1d4d38722): error 50
Oct 5 14:04:40 unknown unix: [ID 100000 kern.notice]
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bc70
zfs:zio_done+122 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bc8c
zfs:zio_next_stage+66 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bcac
zfs:zio_wait_for_children+46 (dbbf8880, 11, dbbf8)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bcc0
zfs:zio_wait_children_done+18 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bcd8
zfs:zio_next_stage+66 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bd10
zfs:zio_vdev_io_assess+11a (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bd24
zfs:zio_next_stage+66 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bd64
zfs:vdev_mirror_io_done+289 (dbbf8880)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bd78
zfs:zio_vdev_io_done+25 (dbbf8880, 0, 0, 0, )
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bdc8
genunix:taskq_thread+176 (d430a290, 0)
Oct 5 14:04:40 unknown genunix: [ID 353471 kern.notice] db95bdd8
unix:thread_start+8 ()
Oct 5 14:04:40 unknown unix: [ID 100000 kern.notice]
Oct 5 14:04:40 unknown genunix: [ID 672855 kern.notice] syncing file
systems...Oct 5 14:04:40 unknown genunix: [ID 733762 kern.notice] 2
Oct 5 14:04:41 unknown genunix: [ID 733762 kern.notice] 1
Oct 5 14:04:42 unknown genunix: [ID 904073 kern.notice] done
Oct 5 14:04:43 unknown genunix: [ID 111219 kern.notice] dumping to
/dev/dsk/c0d0s3, offset 144769024, content: kernel
Oct 5 14:04:50 unknown genunix: [ID 409368 kern.notice] ^M100% done: 44041
pages dumped, compression ratio 2.85,
Oct 5 14:04:50 unknown genunix: [ID 851671 kern.notice] dump succeeded
The machine is an Intel P4 laptop (1GB of RAM which Solaris is using 512MB of)
running Window SP2 with Solaris 10 Nevada b48 inside VMWare Player. I noticed
that almost right away there were some (5-10) CKSUM errors but those might be
caused by me pausing the machine turning off the computer and starting over the
next day?
I''m not on the mailinglist, though I am subscribed to the RSS.
Thanks,
- Simon
This message posted from opensolaris.org
