Monday, 5 May 2008
|
| Re: login mount problem Ian Collins 10:47:59 |
| | Roger wrote:
Hi, I have setup a Solaris 9 machine. I mounted the /home directory from another server /home2 in /etc/vfstab. 135.252.134.58:/home2 - /home nfs - yes rw,bg,intr,hard,proto=tcp You really should not do that, /home is there for the automounter.
Then I can access my home directory /home/zhangl on my Solaris 9 machine. But I can not rlogin to the machine with my login zhangl and password of 135.252.134.58. What should I do? Which file need to be updated? Read up on NIS, it is very easy to set up on a small (or even large)
network.
--
Ian Collins.
|
| |  1 answer | Add comment |
|
| Interactive Unix password recovery Guest 09:40:34 |
| | I have an ancient Interactive Unix box with an application still
running on it. The passwords were lost long ago. I need to change
the IP address on it. Is there any way to recover the lost password?
|
| | 5 answers | Add comment |
|
| Re: Can't play DVD's in SNV-B87 Thommy M. 08:25:18 |
| | Bernie wrote:
[...]
I was going to put solaris on my daughters laptop but she'll get pcbsd instead now.
Nice. Does that mean that we won't see you here again?
|
| | Add comment |
Sunday, 4 May 2008
|
| IPMP and Zones Pavelj 23:54:47 |
| | Hi guys.
I have a v240 with 4 bge nics.
I would like to run 2 zones on it, so each zone has it's one IP.
What is also needed, is a nic level redundancy for each zone - 2 nics
per zone.
And a nic redundancy for the global zone - all 4 ports.
The question is, can you define several IPMP groups on one nic.
Smth like:
bge0 bge1 bge2 bge3
zonegr_1 zonegr_1 zonegr_2 zonegr_2
10.0.0.1 10.0.0.2
globgr globgr globgr globgr
10.0.0.3
All the nics are in the same vlan, but it's very important that i can
pull off 2 cables and only one zone will lose connectivity while the
second one and global zone remain reachable - application redundancy
testing.
If you can point me to any docs\examples or share your your ideas what
is the best way to achieve a config like this.
It seems that if one defines the zone's physical device to a nic which
is already a part of an IPMP group, you get the redundancy
automatically.
But found no mention wherever one can define so a nic will be a member
of several IPMP groups.
Thanks in advance.
|
| | 2 answer | Add comment |
|
| Re: What services can I disable on a laptop? Thommy M. 21:27:06 |
| | Dave wrote:
Rich Teer wrote: On Sat, 3 May 2008, Dave wrote: so I might as well disable it. But it dropped me out of Gnome and I use it again until I had enabled it. Is this a fresh install or upgrade? I ask because fresh installs now have SBD (secure by default), which means that most (network) services are disabled by default. I did a fresh install of the latest community edition. I noticed the wireless was disabled, which threw me a bit, as I'm sure it was enabled when I installed the developer edition a few months back. How much RAM and what CPU do you have? My Ferrari 3400 with 2GB of RAM is still quite snappy. Dual core 2.0 GHz Intel. 2 GB RAM. It's not slow, but if I can save a bit of time during boot I would like to. It was dreadfully slow at one point, but this was because is sat in a loop telling me the wireless was off, and it could not set the power level. That bug is now resolved, so it must have saved 2-3 minutes. The drive is quite new - its a 300 GB unit and they have not been around long. I don't know the speed of it off-hand, but being quite new, I assume it is not too bad.
Compared to boot Windows, SXCE is quite snappy. I've a Ferrari 4000/2GB
with nvb_76 and a Ferrari 5000/2GB with WinXP/SP2 and to test them I
boot both up in parallel, login and start Firefox to the same page. SXCE
always wins and then it has also started MySQL, Apache and GlassFish.
Haven't tested any Linux distro but I think it will beat WinXP too.
|
| | Add comment |
|
| Re: SUN T1000: Kernel panic during jumpstart Hume Spamfilter 20:53:25 |
| | In comp.unix.solaris Niels <write.niels@gmail.com> wrote:
kernel correctly via TFTP. The kernel also mounts the NFS root file system and starts the installation. After the installer probed bge3 to
What is the NFS server? I'm assuming you don't have one T1000 jumpstarting
the other.
--
Brandon Hume - hume -> BOFH.Ca, http://WWW.BOFH.Ca/
|
| | Add comment |
Friday, 25 April 2008
|
| Adding line in crontab from command line... Guest 10:44:19 |
| |
Hi:
Can you tell me if is there any way to make this task ?
00 1 * * * /monitor_file_system 2>/dev/null > crontab
Thank You !
|
| | 10 answers | Add comment |
|
| 280R slow disks? Henrik Goldman 04:12:07 |
| | Following my other Sun 280R questions I've had this week I reached a state
where the system is up and stable.
I installed it now with Solaris 8 (plain install, Generic_108528-29) but
comparing with our older Ultra5 I don't see the 3-4 times speed increase
that I had hoped for.
We mainly use this system for compilation of c++ code using g++ and the
specs being 2x 750 MHz with 2GB ram should outperform our Ultra 5 with 400
MHz and 512 MB easily.
While I've only done a bit of early testing I noticed that the disks are
pretty slow. The system comes with 2 Seagate 36GB FC-AL disks but currently
we only use one of them.
Maybe I'm missing some magic patches compared to a base install that would
help?
Thanks.
-- Henrik
|
| | 31 answer | Add comment |
|
| zfs list - size in KB/MB/GB Martin Paul 03:33:17 |
| | For a local script which generates disk usage statistics, I need to
collect data about e.g. the size of users' home directories. For UFS,
quot was very usable, but unfortunately this doesn't work on ZFS.
So I decided to use "zfs list -H -o name,used" instead, but it has the
bad habit of making the output user-friendly by showing KB, MB or GB
depending on the size of a filesystem. E.g.:
data/user1 24.5K
data/user2 316M
data/user3 1.45G
It's not that I don't know how to convert GB into KB, but the problem is
that I get different granularity depending on the size of the
filesystem. When comparing zfs output from yesterday with the one from
today, for user3 to actually see a diff his filesystem must grow by at
least 10MB. For user2, a change of 1MB is visible, while for user1 512b
make a diff.
Is there a way to make "zfs list" show sizes in KB (or at least MB) for
*all* filesystems, independent of the size? I think an implementation
like used in du (default block, -k KB, -h user-friendly) would be better
for scripts.
mp.
--
SysAdmin | Institute of Scientific Computing, University of Vienna
PCA | Analyze, download and install patches for Solaris
| http://www.par.univie.ac.at/solaris/pca/
|
| | 1 answer | Add comment |
|
| Is anyone else having trouble connecting to the Sun Update service? Gorkhan 01:33:15 |
| | I get a lot of traffic to my /var/adm/messages like this:
Apr 22 04:46:57 hqsun4 CNS Transport[7163]: [ID 617356 daemon.error]
Could not send message from queue: /var/tmp/cctransport/CCTQueueStore/
store/queues/REGISTRATION/registration
So I tried this:
(s) root@hqsun4 [~] 514# export PATH=/usr/jdk/instances/jdk1.5.0/bin:
$PATH
(s) root@hqsun4 [~] 515# /usr/sbin/sconadm register -a -r ~/
registration.properties.txt
sconadm is running
Authenticating user ...
finish registration!
(s) root@hqsun4 [~] 516# /usr/lib/cc-cfw/platform/transport/bin/
cctrunner -p
* About to connect() to cns-transport.sun.com port 443
* Trying 198.232.168.137... * Connection timed out
* couldn't connect to host
* Closing connection #0
* About to connect() to cns-transport.sun.com port 443
* Trying 198.232.168.137... * Connection timed out
* couldn't connect to host
* Closing connection #0
curl error code: 7
curl message: couldn't connect to host
Then I tried this:
(s) root@nvsun3 [~] 546# traceroute 198.232.168.137
traceroute: Warning: Multiple interfaces found; using 10.110.3.103 @
eri0
traceroute to 198.232.168.137 (198.232.168.137), 30 hops max, 40 byte
packets
1 10.110.3.2 (10.110.3.2) 0.618 ms 0.426 ms 0.332 ms
2 vl739.na21.b003393-1.iad03.atlas.cogentco.com (38.104.58.93)
2.036 ms 1.932 ms 1.918 ms
3 vl3515.mpd01.iad03.atlas.cogentco.com (38.20.35.249) 1.606 ms
2.003 ms 1.574 ms
4 vl3489.mpd01.dca02.atlas.cogentco.com (154.54.7.93) 2.279 ms
2.244 ms vl3490.mpd01.dca02.atlas.cogentco.com (154.54.7.97) 2.205 ms
5 vl3495.mpd01.iad01.atlas.cogentco.com (154.54.5.62) 2.690 ms
vl3497.mpd01.iad01.atlas.cogentco.com (154.54.5.66) 2.205 ms
vl3495.mpd01.iad01.atlas.cogentco.com (154.54.5.62) 2.718 ms
6 cox.iad01.atlas.cogentco.com (154.54.14.22) 2.425 ms 2.881 ms
2.502 ms
7 ip70-167-150-62.at.at.cox.net (70.167.150.62) 3.349 ms 3.705 ms
3.298 ms
8 border7.te1-1-bbnet1.wdc002.pnap.net (216.52.127.33) 4.597 ms
8.341 ms 7.243 ms
9 seven-6.border7.wdc002.pnap.net (216.52.125.250) 4.177 ms 4.296
ms 4.549 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * *^C
That doesn't really tell me anything useful, so I tried this:
(s) harriss@eusun0 [~] 267$ telnet 198.232.168.137 443
Trying 198.232.168.137...
telnet: Unable to connect to remote host: Connection timed out
Now I am thinking this problem exists outside of my machines and my
sphere of control. If true, then others must be seeing it also. Is
anyone else having the same issue?
Thanks in advance,
-smh
|
| | Add comment |
|
| Problems with new Solaris Developers Express x86 Chuck Forsberg 01:09:33 |
| | I downloaded the new sol-nv-b79b-x86-dvd.iso and
am having problems.
Mounting an NFS share elicits the complaint "not owner".
I tried the procedure to set a static IP address and then
the network didn't work at all.
I'm running a terminal right clicking from Gnome and /usr/X11/bin
isn't in PATH!
How to run vncserver with a Gnome desktop?
--
/u/caf/signature.txt
|
| | 2 answer | Add comment |
|
| problem with netra t 105 LOM Gorkhan 00:54:47 |
| | I have a Netra T 105 that I am unable to login thru ttya (the LOM).
My keystrokes are ignored. I cannot access the LOM, nor can I do
anything at the "ok" prompt. This machine is the only "T 105" at my
disposal, so I do not have anything to compare it to.
We are running Solaris-9 on the affected box.
I have checked and rechecked my serial cables & adapters. If I tip
from another machine to the bad machine's ("baddog") ttya, then reboot
baddog, I see the expected traffic across the serial port, including
it prompting for a login at the end of the boot.
I added this entry into /etc/inittab:
conb:1234:respawn:/usr/lib/saf/ttymon -g -h -p "`uname -n` login: " -T
vt100 -d /dev/term/b -l contty -m ldterm,ttcompat
Logins to ttyb work just fine (using the same cables & adapters).
Thinking I was on to something, I added a similar line for /dev/term/
a.
No change.
If I tip from baddog out ttya, as in:
# tip -9600 /dev/term/a
On the machine I am tip'ing into I can do this:
# echo "HELLO" >> /dev/term/b
And I will see HELLO on baddog. I take this to mean that physically
the hardware is fine - it is capable of both sending and receiving
characters.
I then thought I could just use ttyb as the console, using "consadm".
No such luck, consadm reports that no carrier is present on ttyb, even
when I am logged in thru the port. Also, consadm refuses to accept
/dev/term/b in it's permanent list:
(s) root@nvsun3 [~] 531# consadm -a /dev/term/b -p
consadm: invalid device /dev/term/b
(s) root@nvsun3 [~] 532# consadm
(s) root@nvsun3 [~] 533# consadm -p
We are running the most current firmware for this model of LOM:
---
(s) root@nvsun3 [~] 528# lom -a
PSUs:
1 OK
Fans:
1 OK speed 92%
2 OK speed 95%
3 OK speed 100%
4 OK speed 100%
LOM configuration settings:
serial escape sequence=#.
serial event reporting=on
Event reporting level=fatal, warning & information
Serial security=enabled
Disable watchdog on break=disabled
Automatic return to console=disabled
alarm3 mode=user controlled
firmware version=3.14
firmware checksum=1190
product revision=0.1
product ID=Netra T1 200
LOM Event Log:
+326d+23h58m52s host reset
+333d+23h55m2s host reset
+340d+23h51m11s host reset
+347d+23h47m26s host reset
+354d+23h43m28s host reset
+361d+23h39m32s host reset
4/22/2008 21:29:21 GMT LOM time reference
+0h0m0s LOM reset
+0h0m0s host power on
4/22/2008 21:33:48 GMT LOM time reference
LOM alarm states:
Alarm1=off
Alarm2=off
Alarm3=on
Fault LED=off
LOM watchdog (ASR) settings:
Watchdog=off
Hardware reset=off
Timeout=127 s
Supply voltages:
1 5V status=ok
2 3V3 status=ok
3 +12V status=ok
4 -12V status=ok
5 CPU core status=ok
6 +3VSB status=ok
System status flags:
1 SCSI-Term status=ok
2 USB0 status=ok
3 USB1 status=ok
4 SCC status=ok
System Temperature Sensors:
1 Enclosure 36 degC : warning 67 degC : shutdown 72 degC
System Over-temperature Sensors:
1 CPU status=ok
Console output prior to last reset:
an't find host ntp2.usno.navy.mil
Apr 21 18:29:57 nvsun3 ntpdate[178]: can't find host
tick.usno.navy.mil
Apr 21 18:30:25 nvsun3 ntpdate[178]: can't find host
tock.usno.navy.mil
Apr 21 18:30:25 nvsun3 ntpdate[178]: no servers can be used, exiting
---
So I am stuck for an answer. Thanks in advance for any suggestions!
-smh
|
| | 2 answer | Add comment |
Thursday, 24 April 2008
|
| Re: how many ways to disable telnet Oleksii Dzhulai 19:26:10 |
| | On Apr 20, 7:10 pm, solaris chat <admpra...@gmail.com> wrote:
can any body tell me howmany ways to disabletelnetin solaris 9 in particular in /etc/inet/inetd.conf how we do it thanks in advance
Also you can rename
/usr/sbin/in.telnetd to /usr/sbin/not.telnetd
|
| | Add comment |
|
| Reducing TCP timers Gev_d 19:01:51 |
| | Hi All,
In a test enviroment i'm interested in tweaking the system timers
( eg for a time out on a 1/2 open TCP connection). Not sure how this
can be done. Anyone with any tips, i'd appreciate your insights.
Regards,
Giovanni
|
| | 2 answer | Add comment |
|
| nfs problem Jim Smith 14:57:46 |
| | I am not sure if this is strictly a Solaris problem or not. I am
trying to mount a directory from an nfs server running linux on a
Solaris client.
The directory seems to mount fine and works fine for a minute or two
and then I start getting errors. All subdirectories will give
"permission denied" errors.
Any subdirectory that I open during the first minute or two while
things are working will remain accessible, but if I close it I will get
the same errors and not be able to re open it. This makes me think
the error has something to do with stale file handles, but I am not
sure how to proceed.
I can mount the directory from a linux client with no troubles.
|
| | 9 answers | Add comment |
|
| Graphical file system Monitor Guest 11:46:41 |
| | Hi
I'm looking for a Graphical file system Monitor, to monitorying 20
servers, and I'd like any suggestion from you.
Thanks in advance.
|
| | 2 answer | Add comment |
|
| Milestones Oleksii Dzhulai 11:05:31 |
| | Hi!
As you know 10 the recommended way of changing runlevels is through
SMF on Solaris 10.
For example:
#svcadm milestone single-user
It is a knew way.
The old way was through init command, for example:
#init 1
and it works also.
So a question is. Changing runlevel in a new way I don't see changes
with "who -r" command. It keeps to show old runlevel. Is it ok or may
be I do something wrong?
Thanks.
---
http://unixinmind.com
|
| | 2 answer | Add comment |
|
| Cant get the ok prompt (on Sunfire V240) Ale 04:36:05 |
| | Hello!
This is the problem that see on a SunFire V240:
break -y
(the bge0 IP correctly stop to ping)
console -f
no more output from serial (or net mngt ethernet) the ok prompt do not
start.
I usually go to ALOM console also on other servers (V240 and V245) to
install
different releases of Solaris starting for the "boot cdrom" at ok
prompt.
I cannot found a reason for this behavior on this server.
Any idea ?
Regards,
Alessandro
|
| | 6 answers | Add comment |
|
| Come join us at CommunityOne - May 5th in San Francisco Guest 03:43:28 |
| | Hi!
It is time for the second annual CommunityOne event, which is a FREE
and open developer conference taking place on May 5th in the Moscone
center in San Francisco. There will be a variety of tracks and
technical sessions from Java and scripting development to operating
systems and databases with developer content for everyone ranging from
from students to experts and startups to enterprises.
For you C, C++, or Fortran developers out there, there are a variety
of sessions to help get you started with OpenSolaris and Linux
development using Sun Studio software, including how to take advantage
of optimizing compilers and multicore development. Check out the
"Maximize Application Performance on the OpenSolaris™ and Linux
Operating Systems, Using Sun Studio Software" session to see how Sun
Studio Software simplifies development and increases application
performance for the newest multicore systems. Learn about
characteristics of the CMT processor that can be used to improve
application performance using Sun Studio Sowftare and the CMT
Developer Tools in the "Techniques for Ultilizing CMT" session.
There is also the exciting OpenSolaris track - filled with sessions on
the new Image Packaging System. Learn about everything from building
a software project to delivering and taking advantage of a great
distribution opportunity. Check out the "A Student and New User View
of the OpenSolaris Operating System" session and learn about
navigating OpenSolaris packages using SunStudio Sofware and the
NetBeans IDE.
Come join us and see how sun innovation and community innovation is
brought to market in integrated offerings. Meet key engineers and
view Sun Studio demos in our Sun Studio booth in the expo area.
For more information on tracks and sessions and for free registration
visit the CommunityOne website: developers.sun.com/events/communityone
See you there!
Ikroop Dhillon
Sun Studio Product Marketing Manager
|
| | Add comment |
|
| when zone boot, it ignores some entries in sysidcfg Jdh13 00:57:18 |
| | hello all,
i play with zones on T2000 S10 8/07. I create a new zone, and copy a
sysidcfg in ../etc before booting it.
Here is my sysidcfg:
system_locale=C
timezone=MET
timeserver=localhost
network_interface=PRIMARY {netmask=255.255.255.0
default_route=192.168.4.1
protocol_ipv6=no}
name_service=DNS {domain_name=moi.fr name_server=192.168.4.3}
root_password=xxxxxxxx
security_policy=NONE
terminal=vt100
nfs4_domain=dynamic
but i'm prompted to confirm the hostname, to give the dns, before it can
continue to boot.
I'm using the same sysidcfg to jumpstart real machines without this problem.
How can i find why it prompts me for hostname/dns?
thanks in advance,
gerard
|
| | 1 answer | Add comment |
Wednesday, 23 April 2008
|
| nvramrc Solaris Chat 22:44:58 |
| | what "rc" in nvram-rc?
|
| | 2 answer | Add comment |
|
| How to capture ssh authenticity changes in bash? J 22:08:32 |
| | Hi,
How do I capture if the ssh authenticity of a host has changed and
redirect it to a log file using bash? Some have people have suggested
to use "expect", however, would it be possible to just use bash?
$ ssh ssh-server.example.com
The authenticity of host 'ssh-server.example.com (1.2.3.4)' can't be
established.
RSA key fingerprint is 98:2e:e7:d0:ge:9f:ac:67:28:c2:42:2d:
37:19:58:4d.
Are you sure you want to continue connecting (yes/no)?
I've tried redirecting it to a log file but the redirection doesn't
seem to work; my log file is empty -
$ ssh ssh-server.example.com > output.log
The authenticity of host 'ssh-server.example.com (1.2.3.4)' can't be
established.
RSA key fingerprint is 98:2e:e7:d0:ge:9f:ac:67:28:c2:42:2d:
37:19:58:4d.
Are you sure you want to continue connecting (yes/no)?
$ cat output.log
$
The stdout doesn't seem to work either -
$ ssh ssh-server.example.com 1> output.log
The authenticity of host 'ssh-server.example.com (1.2.3.4)' can't be
established.
RSA key fingerprint is 98:2e:e7:d0:ge:9f:ac:67:28:c2:42:2d:
37:19:58:4d.
Are you sure you want to continue connecting (yes/no)?
$ cat output.log
$
If I use the verbose mode ($ ssh -v server.example.com > output.log),
the script captures and redirects everything in the log file EXCEPT
for "The authencity of host... Are you sure you want to continue
connecting (yes/part)?" part. Is the message not a standard output?
I'm able to redirect the stderror (i.e. when connection times out) to
the log file.
The case of authenticity change is not quite as easy to capture as I
thought it would be.
Thoughts?
TIA.
J
|
| | 2 answer | Add comment |
|
| alternate command Solaris Chat 18:05:46 |
| | is there any other command tosee the running processes except
1.prstat
2.ps
regards
|
| | 5 answers | Add comment |
|
