LinuxQuestions.org
Visit Jeremy's Blog.
Home Forums Tutorials Articles Register
Go Back   LinuxQuestions.org > Forums > Other *NIX Forums > Solaris / OpenSolaris
User Name
Password
Solaris / OpenSolaris This forum is for the discussion of Solaris, OpenSolaris, OpenIndiana, and illumos.
General Sun, SunOS and Sparc related questions also go here. Any Solaris fork or distribution is welcome.

Notices


Reply
  Search this Thread
Old 09-27-2008, 06:47 AM   #1
czezz
Member
 
Registered: Nov 2004
Distribution: Slackware/Solaris
Posts: 924

Rep: Reputation: 43
terribly slow tape writting


OS: Solaris 10 5/08 s10s_u5wos_10 SPARC
HARDWARE: sun4u Sun Fire V490
Memory: 64 Gb
CPU: 4x dual core

This machine has conected tape drive: Quantum Super DLT 600

Every weekend I need to start backup that size is 160 Gb. Its fine because tape has 600 Gb space.

What is not fine is that writting speed of tape is around 3.5 Mbyte/sec.
That is terribly slow when I need to record 160 Gb it takes about 12 - 14 hours.

Is there anything I can tune up with that tape drive ???

Below 'iostat' during tape work:
Code:
 27  2  0 70
                    extended device statistics
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d0
    0.0    1.0    0.0    4.5  0.0  0.0    0.0    6.7   0   1 md/d1
    0.0    1.0    0.0    8.0  0.0  0.0    0.0    5.5   0   1 md/d3
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d4
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d10
    0.0    1.0    0.0    4.5  0.0  0.0    0.0    6.7   0   1 md/d11
    0.0    1.0    0.0    8.0  0.0  0.0    0.0    5.5   0   1 md/d13
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d14
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d20
    0.0    1.0    0.0    4.5  0.0  0.0    0.0    4.3   0   0 md/d21
    0.0    1.0    0.0    8.0  0.0  0.0    0.0    4.2   0   0 md/d23
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 md/d24
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 c0t0d0
    0.0    3.0    0.0   13.0  0.0  0.0    0.0    6.3   0   1 c1t1d0
    0.0    2.0    0.0   12.5  0.0  0.0    0.0    6.1   0   1 c1t0d0
    0.0    0.0    0.0    0.0  0.0  0.0    0.0    0.0   0   0 ssd4
    0.5    0.5    4.0    4.0  0.0  0.0    0.0    3.6   0   0 c4t40d1
    0.0   16.5    0.0  132.0  0.0  0.0    0.0    0.4   0   1 c4t40d2
  464.5    0.0 3716.1    0.0  0.0  0.2    0.0    0.4   0   8 c4t40d3
    0.0   58.5    0.0 3685.6  0.0  1.0    0.0   17.0   0 100 rmt/0
Note: c4t40d is an array disk and during daily tasks works with read/write mode 60 - 90 Mbytes/sec (so it is not a matter of disk speed).

Last edited by czezz; 09-27-2008 at 06:57 AM.
 
Old 09-27-2008, 06:03 PM   #2
MensaWater
LQ Guru
 
Registered: May 2005
Location: Atlanta Georgia USA
Distribution: Redhat (RHEL), CentOS, Fedora, CoreOS, Debian, FreeBSD, HP-UX, Solaris, SCO
Posts: 7,831
Blog Entries: 15

Rep: Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669Reputation: 1669
Most tape drives (including DLT) have a "compressed" mode where they do hardware compression. Typically this is the fastest write the device will do. Different device files actually determine different characteristics (based on the major/minor of the device) which include things like compression, no compression, automatic rewind, no rewind.

You need to insure you're using the compression device for one thing.

There are various tuning things that might help but you'd have to check those in context of what else you're doing on the system.
 
Old 09-28-2008, 03:36 AM   #3
czezz
Member
 
Registered: Nov 2004
Distribution: Slackware/Solaris
Posts: 924

Original Poster
Rep: Reputation: 43
So far device file I use is: /dev/rmt/0n
As I understand I should try also other devices in /dev/rmt and then compare data transfer rate.
I guess it is only way to ensure which one is the fastest.

Dont have access now to serwer so I will check it on monday.

Anyway Im still worried because vendor specification says that data transfer should be like:
Native: 36 MBps
Compressed: 72 MBps
My current is 3.5 MBps
I was wondering about some "magic" settings in /etc/system which I dont know about ?
 
Old 09-28-2008, 05:02 AM   #4
jlliagre
Moderator
 
Registered: Feb 2004
Location: Outside Paris
Distribution: Solaris 11.4, Oracle Linux, Mint, Debian/WSL
Posts: 9,789

Rep: Reputation: 492Reputation: 492Reputation: 492Reputation: 492Reputation: 492
The tape drive is 100% busy so the bottleneck is here however the drive is given for ten times the throughput you observe ...

What command are you using to write on it (I guess ufsdump given the 63KB block size) ?

Are there any errors (iostat -E, dmesg) ?

Did it use to run faster or is this a new configuration ?

Did you check the hardware ? cables, terminations, head cleaning ?
 
Old 09-30-2008, 02:30 AM   #5
czezz
Member
 
Registered: Nov 2004
Distribution: Slackware/Solaris
Posts: 924

Original Poster
Rep: Reputation: 43
Here is what I do with tape:
Code:
ufsdump 0ucf /dev/rmt/0n /opt/backup/
ufsdump 0ucf /dev/rmt/0n /opt/appz/
These 2 commands. As you can see /dev/rmt/0n is used.
I guess I should try some other device listed below.
Here are 2 that looks promising for me:
Code:
/dev/rmt/0h   # Sounds like "hardware-something- acceleration..."
or
/dev/rmt/0c   # Sounds like "compressing-on-the-fly... or so"
Here is a list of device-files I have for tape device:
Code:
[root@Etna /]# ls -al dev/rmt
total 58
drwxr-xr-x   2 root     root         512 Jun 17 17:04 .
drwxr-xr-x  19 root     sys         4096 Sep 10 11:39 ..
lrwxrwxrwx   1 root     root          43 Jun 17 17:04 0 -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0b -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:b
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0bn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:bn
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0c -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:c
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0cb -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:cb
lrwxrwxrwx   1 root     root          46 Jun 17 17:04 0cbn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:cbn
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0cn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:cn
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0h -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:h
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0hb -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:hb
lrwxrwxrwx   1 root     root          46 Jun 17 17:04 0hbn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:hbn
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0hn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:hn
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0l -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:l
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0lb -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:lb
lrwxrwxrwx   1 root     root          46 Jun 17 17:04 0lbn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:lbn
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0ln -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:ln
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0m -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:m
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0mb -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:mb
lrwxrwxrwx   1 root     root          46 Jun 17 17:04 0mbn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:mbn
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0mn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:mn
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0n -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:n
lrwxrwxrwx   1 root     root          44 Jun 17 17:04 0u -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:u
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0ub -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:ub
lrwxrwxrwx   1 root     root          46 Jun 17 17:04 0ubn -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:ubn
lrwxrwxrwx   1 root     root          45 Jun 17 17:04 0un -> ../../devices/pci@8,600000/scsi@2,1/st@5,0:un
I will test these 2 I mentioned above. Help me up to choose the best one.
That tape is connected via SCSI cable. Will check hardwer healthy - should be good because it is newnhardwer.
Also next time I execute iostat -E to see errors.

Last edited by czezz; 09-30-2008 at 02:52 AM.
 
Old 10-03-2008, 05:43 AM   #6
czezz
Member
 
Registered: Nov 2004
Distribution: Slackware/Solaris
Posts: 924

Original Poster
Rep: Reputation: 43
Hi,
here is a list of all my tries with all devices I have in /dev/rmt.
As you can see "iostat" shows no differences at all.
Take a look:
Code:
#ufsdump 0ucf /dev/rmt/0cn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  464.0    0.0 3711.9    0.0  0.0  0.2    0.0    0.4   0   8 c4t40d3
    0.0   59.0    0.0 3716.9  0.0  1.0    0.0   16.9   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0hn /oradata/u03/cold_backup/
     r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  460.5    0.0 3684.0    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   59.0    0.0 3717.0  0.0  1.0    0.0   16.9   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0hbn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  464.0    0.0 3712.2    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   59.0    0.0 3717.2  0.0  1.0    0.0   16.9   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0ln /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  460.0    0.0 3680.0    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   58.5    0.0 3685.5  0.0  1.0    0.0   17.0   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0lbn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  460.0    0.0 3680.3    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   58.5    0.0 3685.8  0.0  1.0    0.0   17.0   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0mbn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  460.0    0.0 3679.8    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   58.5    0.0 3685.3  0.0  1.0    0.0   17.0   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0mn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  460.0    0.0 3679.8    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   58.5    0.0 3685.3  0.0  1.0    0.0   17.0   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0ubn /oradata/u03/cold_backup/
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  459.4    0.0 3675.3    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   58.5    0.0 3684.8  0.0  1.0    0.0   17.0   0 100 rmt/0

#ufsdump 0ucf /dev/rmt/0un /oradata/u03/cold_backup/    
    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device
  459.5    0.0 3675.8    0.0  0.0  0.2    0.0    0.3   0   7 c4t40d3
    0.0   59.0    0.0 3716.8  0.0  1.0    0.0   16.9   0 100 rmt/0

jlliagre:
Yes, ufsdump uses 63 Kb bocks:
Code:
DUMP: Writing 63 Kilobyte records
I have also preapared iostat -E for errors during ufsdump is recording on a tape. Please note that I executed 'iostat -E 2' eche 2 seconds new output were displayed and it has shown exectly this same results. No increasing errors.


Code:
sd3       Soft Errors: 2 Hard Errors: 0 Transport Errors: 2
Vendor: TSSTcorp Product: CD/DVDW TS-L632D Revision: SR02 Serial No:
Size: 0.00GB <0 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 2 Predictive Failure Analysis: 0
ssd0      Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
Vendor: SEAGATE  Product: ST314655FSUN146G Revision: 3092 Serial No: 0807S47FK3
Size: 146.80GB <146800115712 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 0 Predictive Failure Analysis: 0
ssd2      Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
Vendor: SEAGATE  Product: ST314655FSUN146G Revision: 3092 Serial No: 0807S463ZY
Size: 146.80GB <146800115712 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 0 Predictive Failure Analysis: 0
ssd4      Soft Errors: 1 Hard Errors: 0 Transport Errors: 0
Vendor: SUN      Product: StorEdge 3510    Revision: 421F Serial No:
Size: 586.17GB <586166566912 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 1 Predictive Failure Analysis: 0
ssd5      Soft Errors: 2 Hard Errors: 4 Transport Errors: 9
Vendor: SUN      Product: StorEdge 3510    Revision: 421F Serial No:
Size: 586.17GB <586166566912 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 4 Recoverable: 0
Illegal Request: 2 Predictive Failure Analysis: 0
ssd6      Soft Errors: 1 Hard Errors: 3 Transport Errors: 0
Vendor: SUN      Product: StorEdge 3510    Revision: 421F Serial No:
Size: 146.54GB <146541641728 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 3 Recoverable: 0
Illegal Request: 1 Predictive Failure Analysis: 0
ssd7      Soft Errors: 3 Hard Errors: 0 Transport Errors: 0
Vendor: SUN      Product: StorEdge 3510    Revision: 421F Serial No:
Size: 299.73GB <299731517440 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 3 Predictive Failure Analysis: 0
st1       Soft Errors: 0 Hard Errors: 2 Transport Errors: 0
Vendor: QUANTUM  Product: SDLT600          Revision: 2B2B Serial No: Á+

Last edited by czezz; 10-03-2008 at 05:46 AM.
 
Old 01-02-2009, 07:31 AM   #7
czezz
Member
 
Registered: Nov 2004
Distribution: Slackware/Solaris
Posts: 924

Original Poster
Rep: Reputation: 43
I have plugged "terminator dongle" into tape drive - still no changes - 3.5 MB/w, 100% busy
 
  


Reply



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
Fresh Slack installation terribly slow! atriya Slackware 11 09-21-2008 02:12 AM
SUSE 11 RC1 is terribly slow riba43 SUSE / openSUSE 7 06-10-2008 01:03 AM
Full duplex NIC running terribly slow pdeman2 Linux - Hardware 5 11-13-2006 11:35 AM
Konqueror terribly slow while loading web pages Thora Linux - Software 2 09-09-2006 02:25 AM
computer terribly slow Atmchicago Linux - Hardware 2 08-02-2005 12:38 AM

LinuxQuestions.org > Forums > Other *NIX Forums > Solaris / OpenSolaris

All times are GMT -5. The time now is 03:15 AM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Open Source Consulting | Domain Registration