headers
SIBLEY, Ken R. - ACCOR-NA | 29 Nov 17:38

SUMMARY: T2000 RAID creation

Thanks to those who responded to my question:
The original question was:
Can I retain the contents of c0t0d0 if I run
raidctl -c c0t0d0 c0t1d0

The answer seems to be consistenly: NO

There is an off chance that creating the mirror
at the ok prompt using:
ok 0 1 create-im-volume

might allow the data to be retained, but it
does not look promising.

Thanks again,
Ken
_______________________________________________
sunmanagers mailing list
sunmanagers <at> sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Permalink | Reply | 0 comments |
headers
Plant, Dean | 29 Nov 13:50
Picon

SUMMARY: /var out of disk space

As covered in 5.10 of the FAQ, space is not freed up until the process
exits or closes the file.

In my case it turned out that clearcase needed to be restarted to free
up the space. Lsof was the tool suggested by most and what I used to
find the problem process.

Thanks to all the replies, there were far too many to list them all but
Ill list the ones that helped me locate the problem process and a few
others which may help in other circumstances.

Most people suggested restarting syslogd (pkill -HUP syslogd) first to
make sure none of the log files were the problem as this is an quick and
easy check.

Rob McMahon:
This is FAQ 5.10.  Although it mentions lsof in there, it doesn't give
you a cookbook recipe.  The best way to handle this is to get lsof and
do
lsof -a +L1 /var

Robert Collishe
A process probbly has a file lock (flock) open on the file you have
deleted.
Use the fuser -c command (I think it writes the flags to standard error
so 'fuser -c /var 2> /dev/null' works nicely to get just the procs
themselves) on /var to see what's running - you can then run 'pfiles'
against processes in the output to see what's opened - kill the process
and hey presto all your space will return.  If this is a common issue,
scripting the above stuff is pretty straightforward.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Steaphany Jean Waelder | 29 Nov 18:18

SUMMARY: Third Party PCI EIDE Interface Cards In SPARC systems

I only received one reply:

- - -
If you've already got the drives, then this isn't so helpful, but if you
haven't, I've seen more discussions on SATA pci card support on sparc.
SATA drives are cheaper than PATA these days, mostly, and in fact you
might find it cheaper per GB to buy 300/320/400GB drives.  Similarly, a
2-channel card will be cheaper (though only buy a few dollars) than a 4
channel if you fit it in two drives.

the hatter
- - -

Seems like, despite Sun's designs employing standard interfaces, little third 
party technology really exists.

Here is my  original post:

>Does anyone have any experience installing and using a third party EIDE 
>interface card supporting drives greater than 137 GB into SPARC based systems ?
>
>My goal is to build a NFS server serving 3 Western Digital 250GB drives but 
this 
>exceeds the limit of the onboard interface.

Ms Steaphany Jean Waelder
Owner
Kitsune
1028 Monument Road
Loving, Texas 76460-9724
(Continue reading)

Permalink | Reply | 0 comments |
headers
Andrew Bergman | 28 Nov 04:07
Picon

Summary: can't detect USB tape drive on Solaris 9 x86

My thanks go out to Ric Anderson and Mohammed Tanvir for their replies 
in dealing with my USB tape drive issue.

In the end I resolved it by going onsite, shutting down the machine 
which thankfully is not live yet anyway.  I checked and the internal DAT 
72 drive was powered and plugged in to the USB socket.  To be sure I 
switched sockets and then I restarted the computer and ensured it did a 
boot reconfig.  The drive reappeared on the system and /dev/rmt was 
recreated with all it's devices.

mt -f /dev/rmt/0cn status showed the correct output and the tape 
responded to commands to offline and eject the tape etc.  Backups are 
now running ok so it's all good.

Thanks again

Regards

--

-- 

Andrew Bergman
Systems Administrator
Microlistics Pty Ltd
Direct: +61 3 8329 2050
Mobile: +61 411 706 713
Fax:  +61 3 9326 7588
e-mail:  andrew.bergman <at> microlistics.com.au
--------------------------------------
"It is the way of the world, Baldrick - the abused always kick downwards. I am annoyed, and so I kick the cat...
the cat pounces on the mouse, and finally, the mouse bites you on the behind."
(Continue reading)

Permalink | Reply | 0 comments |
headers
Pandey, Abhimanyu | 24 Nov 18:46
Favicon

SUMMARY: security

Thanks to all for the response.

I have managed to keep the system quite secure until now help from
sun.com and all of you.

As I prepare to move to a new job (sun admin also) my main concern is
that since there will be a time lag before a new sys admin gets in I
feel morally and realistically obligated to secure our current servers
as much as I can.  Thank you all for those responses.  It will serve my
friends here well.

SUMMARY:

Ric Anderson [ric <at> Opus1.COM]

* Go to sunsolve.sun.com and subscribe to Sun Alerts so you get

   email from Sun about security issues.

* Make sure you download and install the recommended OS patch cluster

   regularly.

* Disable any network services you don't absolutely know you need.

* Install a package that lets you monitor the MD5 checksums of all
system

   files and directories, and run the monitor daily (or more often if
your
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jarkko Airaksinen | 24 Nov 08:23

SUMMARY: hba name anomaly

Hello,

Thanks to everyone who replied. The most helpful comments came from Mr.
Koef and Mr. Andrey Borzenkov (their comments are included in this
email).

Shortly, I had problems with fcode load for the other of two seemingly
identical adapters but with different names.

Looks like the other HBA is an original QLogic HBA and the other is Sun
branded QLogic HBA. The adapters have identical hardware but different
NVRAM & FCode.

Br,
Jarkko

-----Original Message-----
From: Koef [mailto:koef <at> notsupported.org]
Sent: jueves, 23 de noviembre de 2006 23:45
To: Jarkko Airaksinen
Subject: Re: Another Partial Summary: hba name anomaly

On Thu, Nov 23, 2006 at 08:41:37PM +0100, Jarkko Airaksinen wrote:

>         SUNW,qlc, instance #0
>
>         fibre-channel, instance #4

It appears you have one original Qlogic HBA (fibre-channel, instance #4)
and
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jarkko Airaksinen | 23 Nov 20:41

Another Partial Summary: hba name anomaly

Hello once more,

Almost there!

For some reason scli shows the adapters' models to be different:

        Enter Selection: 1

------------------------------------------------------------------------
------

Host Name                  : superserver2

HBA Model                  : QLA2310

Port                       : 1

Node Name                  : 20-00-00-E0-8C-80-9F-5A

Port Name                  : 21-00-00-E0-8C-80-9F-5A

Port ID                    : A1-0F-00

Serial Number              : A32159

Driver Version             : qlc-20060630-2.16

FCode Version              : 1.14.15

Firmware Version           : 3.03.117 IP
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jarkko Airaksinen | 23 Nov 18:28

Partial Summary: hba name anomaly

Hello again,

Well, after upgrading the QLA driver & the SAN, doing a reconfigure boot
and refreshing the fabric_WWN_map the server finally found adapter and
it works. I even downed and upped the ports in the two fc switches and
the failover worked like a charm.

 The only part that still fails is the fcode load: luxadm just doesn't
see the adapter.

Prtdiag shows:

PCI#00    fibre-channel-pcif77,fe05                             66Mhz

PCI#01    SUNW,qlc-pci1077,2300.10+                             66Mhz

Luxadm still finds only the PCI#01 adapter (I think?):

  Found Path to 1 FC100/P, ISP2200, ISP23xx Devices

  Opening Device: /devices/pci <at> 80,2000/SUNW,qlc <at> 2/fp <at> 0,0:devctl

  Detected FCode Version:       ISP2300 Host Adapter fcode version
1.14.15 05/16/06

One of responses I got suggested that the adapters are not identical
(like the name anomaly suggests). However what makes this weird is that
both adapters work with the latest qla driver.

Does anyone know how luxadm searches for the adapters? How come the
(Continue reading)

Permalink | Reply | 0 comments |
headers
Warren Liang | 21 Nov 17:51
Picon

SUMMARY: Live Upgrade from 8 to 10

Morning,

Thanks to all the responses, all points to one thing: Need to install SUNWluu and SUNWlur from Solaris 10's
release. Then, apply requested patches.

Have a wonderful holiday.
Warren
_______________________________________________
sunmanagers mailing list
sunmanagers <at> sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Permalink | Reply | 3 comments |
headers
Robert K | 20 Nov 21:03
Picon
Favicon

SUMMARY: Adding entries to /etc/hosts on different servers

Hello Managers
  Thanks for all the help. By playing with the following terrific clues, I managed my problem.

  Denes, Daniel
  cat /etc/hosts | nawk '{for (i = 2; i<= NF; i++) print $i;}' | grep 
'ss[0-9]*$'

  Hutin Bertrand
  bash-2.05#  awk '/ts/{print $2}' /etc/hosts |uniq  
 bash-2.05#  awk '/ss/{print $2}'/etc/hosts |uniq  

  Matthew Stier
  awk '$0 !~ /^[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+/ { next; } $2 ~ 
/[ts]s[0-9]+$/ { print $2 }' /etc/hosts | uniq

Sandwich Maker
  #!/usr/bin/ksh -p
# quick kornshell hack to extract hostnames ending in ts or ss
# optionally followed by digits

while read IP HOST JUNK
do
 case $HOST in
  *[ts]s*([0-9])) print $HOST;;
 esac
done < /etc/hosts | sort -u

  Crist Clark
  $ awk '$2 ~ /[st]s/ {printf "%s\t%s\n" $1 $2}' /etc/hosts | uniq
(Continue reading)

Permalink | Reply | 0 comments |
headers
aams listastecnicas | 19 Nov 17:57
Picon

SUMMARY issue with sdlt in solaris 8 and 9

The issue was fixed in the converter scsi-fiber, the problem was permissions
in an ACL.

The status actually in, the luxadm command it continues reporting the device
as a disk, however solaris now view the correct device in iostat -En as SDLT
drive, and create the special file in /dev/rmt.

The mt command also report correctly this device and tar and ufsdump
commands work fine.

According to which the manufacturer is normal that luxadm report a disk
because  it is reported at FC  evel it is the converter and not directly the
SLDT drive

thanks to all the interested ones
_______________________________________________
sunmanagers mailing list
sunmanagers <at> sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Permalink | Reply | 0 comments |

Gmane