Hi,
I'm confused here about device modes, and we don't appear to be getting
proper compression on actual backups:
Can someone tell me what dev/nst0a, nst0l and nst0m do? I've read the
man page for stinit and mt, and
I can't see where they define what these device modes accomplish and how
the differ from, say, nst0. I'm
concerned that we're not getting proper compression on our backups. We
have four SDLT 600 drives (nst0-3).
These are the non-rewind devices I've been using. We're running a
Quantum M1800 tape library with
four SDLT 600 drives, attached via SCSI to a Linux storage node server.
When I ran jbconfig, I specified
/dev/nst# where # was 0-3 for each device. I'm wondering what these
other modes do, and if maybe I should
be using one of those instead. I'm able to backup and recover fine to
these drives, but compression rates
seem low. Here are some of the volumes we've written to:
volume (%used) written
AR0058 full 284 GB
AR0059 51% 151 GB
ARC026 95% 283 GB
FL1394 54% 163 GB
FL1396 full 300 GB
FL1397 full 291 GB
FL1398 100% 308 GB
FLN026 68% 202 GB
FLN027 65% 195 GB
INC001 full 329 GB
INC002 full 346 GB
INC003 full 292 GB
INC004 full 349 GB
INC005 100% 355 GB
I thought SDLT II tapes were 300/600. With our old ATL P1000 library
that has SDLT 220 drives it was not uncommon to get
135-160+ GB on a SDLT 1 tape. That's nearly 1.5 times the 110 native
capacity on average. I'm not seeing anything near 1.5 times
(let alone twice) the capacity of the SDLT II tapes, however. That would
be 450 GB. Granted, mileage will vary, but I would
have expected to see some higher rates, especially given that we've run
fulls and incrementals on all our hosts. Perhaps, I'm reaching
this inference prematurely, but these volumes comprise a wide variety of
data from across the network, including incrementals and
fulls. I don't see any tapes that have even come close to 400 GB, never
mind 5-600. I've not seen any operational problems, though.
Everything seems to be running just dandy, and I've done many recovers -
all OK, but these low sizes have me concerned.
I've seen write speeds anywhere from 7 MB/s to 60+ MB/s. Last night,
during some fulls I had all 4 drives cranking (5 target
sessions per device), and one drive was at 19 MB/s, another drive at 24
MB/s, another at 50 and the other at 60. The numbers changed around, but
these
speeds seem OK., I guess. We're not using software compression. We're
running NetWorker 7.2.2. I selected Media type: sdlt600 when I first used
jbconfig to configure the library.
I copied the stinit.def file template from Quantum's site, and copied it
to /etc/stinit.def. It has the
following entry for the SDLT600:
# QUANTUM SDLT600
manufacturer=QUANTUM model="SDLT600" {
timeout=3600 # 1 hour timeout
long-timeout=14400 # 4 hour long timeout
can-partitions=0
mode1 blocksize=0 density=0x4A compression=1 # SDLT600 density,
compression on
mode2 blocksize=0 density=0x4A compression=0 # SDLT600 density,
compression off
mode3 blocksize=0 density=0x49 compression=1 # SDLT320 density,
compression on
mode4 blocksize=0 density=0x48 compression=1 # SDLT220 density,
compression on
}
Here's what mt -f /dev/nst0 status shows with no tape in the drive:
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 0 bytes. Density code 0x0 (default).
Soft error count since last status=0
General status bits on (50000):
DR_OPEN IM_REP_EN
Here's what it shows with a NetWorker labeled tape in the drive:
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 0 bytes. Density code 0x4a (no translation).
Soft error count since last status=0
General status bits on (1010000):
ONLINE IM_REP_EN
Here's what inquire shows for drive 1 (same on all the drives):
[email protected]:QUANTUM SDLT600 2929|Tape, /dev/nst0
S/N: RB0502AMC05153
ATNN:QUANTUM SDLT600
RB0502AMC05153
WWNN:500E09E000146337
PORT:00000001
Here's what /proc/scsi/scsi shows:
Host: scsi1 Channel: 00 Id: 01 Lun: 00
Vendor: QUANTUM Model: SDLT600 Rev: 2929
Type: Sequential-Access ANSI SCSI revision: 04
So am I to assume that nst0, 1, 2 and 3 would use mode 1, and nst0a
would used mode 2 and
nst0l would use mode 3 and then nst0m would use mode 4? How do you know
which mode
is assigned to which device: nst0, a, l or m? Are we correct to be using
nst0, 1, 2 and 3 then for hardware compression?
Does anyone have any suggestions?
Thanks
George
--
George Sinclair - NOAA/NESDIS/National Oceanographic Data Center
SSMC3 4th Floor Rm 4145 | Voice: (301) 713-3284 x210
1315 East West Highway | Fax: (301) 713-3301
Silver Spring, MD 20910-3282 | Web Site: http://www.nodc.noaa.gov/
- Any opinions expressed in this message are NOT those of the US Govt. -
To sign off this list, send email to listserv AT listserv.temple DOT edu and type
"signoff networker" in the body of the email. Please write to networker-request
AT listserv.temple DOT edu if you have any problems with this list. You can access the
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
|