Re: sparc64 bridging

2004-05-02 Thread Dan Vande More
On Sun, 2004-05-02 at 02:55, Axel Scheepers wrote:
> Thanks for the pointers, I've just installed the bridge-utils-64 package and
> I'm able to get somewhat further. But some funny things are happening;
> 
> plog0:/home/axel# brctl addbr br0
> plog0:/home/axel# brctl addif br0 eth0
> bridge br0 doesn't exist!
> plog0:/home/axel# brctl show
> bridge name bridge id   STP enabled interfaces
> plog0:/home/axel# brctl addbr br0
> device br0 already exists; can't create bridge with the same name
> plog0:/home/axel# brctl delbr br0
> plog0:/home/axel#
> 
> Kind of 'it doesn't exist! -It does! it doesn't!' ;-)
> (I guess this means compiling it myself with -g and do some debugging)
> 
> The machine in question is a netra x1. It has no floppy/cdrom so I have to
> netboot it to install, with FreeBSD 5.2.1 I get a 'Fast Data MMU miss' every
> time the loader tries to reallocate the kernel. I've already posted this to
> relevant lists. We've got about 6 of these, all with 1gb mem, in our
> basement doing nothing, I figured they'd do nice as a pop/mailhub cluster
> (the remaining 5) and for firewalling (this one).
> I've also tried OpenBSD, which in fact runs pretty fine on them. We do have
> some enterprise 450 machines also which I can probably get my hands on (4way
> smp las time I checked :-) OpenBSD doesn't have decent smp support, and my
> experiences with NetBSD is somewhat dissapointing under heavy loads. So
> that's why I'm trying debian now. (I really like the way things are handled
> compared to the other distros available)
> I don't like keeping a mix of more then 4 operating systems at our site, in
> fact, I'm trying to bring it all down a bit to ease configuration and
> migration.
> We don't have that many people who can manage solaris decently (only 2) so
> I'd be giving myself some extra work if we keep using that.. ;-)

You're in the same situation I am. Most people here CAN manage solaris,
but with the ridiculous install time, patching time, and buggy
utilities(awk, sed, tar), it's nice to have something that works.

I revised my initial reply to the link above, which I found using
google. I was originially going to state that I simply don't think linux
or *bsd supports bridging on the sparc64 arch yet. I don't really know
why, but I think it has more to do with 64 bit vs 32 bit. This is
probably what I should revert back to, I just don't think it's working
yet. Sun is famous for their screwed up hardware, though they are more
famous for their buggy CPUs/Memory, which they claim are transient
errors that a reboot will fix.

Personally, I've tried it quite few times, and haven't been able to get
it to work, but I've only tried on Netra t105s and quad hme cards. 






Re: sparc64 bridging

2004-05-02 Thread Dan Vande More
On Sun, 2004-05-02 at 02:55, Axel Scheepers wrote:
> Thanks for the pointers, I've just installed the bridge-utils-64 package and
> I'm able to get somewhat further. But some funny things are happening;
> 
> plog0:/home/axel# brctl addbr br0
> plog0:/home/axel# brctl addif br0 eth0
> bridge br0 doesn't exist!
> plog0:/home/axel# brctl show
> bridge name bridge id   STP enabled interfaces
> plog0:/home/axel# brctl addbr br0
> device br0 already exists; can't create bridge with the same name
> plog0:/home/axel# brctl delbr br0
> plog0:/home/axel#
> 
> Kind of 'it doesn't exist! -It does! it doesn't!' ;-)
> (I guess this means compiling it myself with -g and do some debugging)
> 
> The machine in question is a netra x1. It has no floppy/cdrom so I have to
> netboot it to install, with FreeBSD 5.2.1 I get a 'Fast Data MMU miss' every
> time the loader tries to reallocate the kernel. I've already posted this to
> relevant lists. We've got about 6 of these, all with 1gb mem, in our
> basement doing nothing, I figured they'd do nice as a pop/mailhub cluster
> (the remaining 5) and for firewalling (this one).
> I've also tried OpenBSD, which in fact runs pretty fine on them. We do have
> some enterprise 450 machines also which I can probably get my hands on (4way
> smp las time I checked :-) OpenBSD doesn't have decent smp support, and my
> experiences with NetBSD is somewhat dissapointing under heavy loads. So
> that's why I'm trying debian now. (I really like the way things are handled
> compared to the other distros available)
> I don't like keeping a mix of more then 4 operating systems at our site, in
> fact, I'm trying to bring it all down a bit to ease configuration and
> migration.
> We don't have that many people who can manage solaris decently (only 2) so
> I'd be giving myself some extra work if we keep using that.. ;-)

You're in the same situation I am. Most people here CAN manage solaris,
but with the ridiculous install time, patching time, and buggy
utilities(awk, sed, tar), it's nice to have something that works.

I revised my initial reply to the link above, which I found using
google. I was originially going to state that I simply don't think linux
or *bsd supports bridging on the sparc64 arch yet. I don't really know
why, but I think it has more to do with 64 bit vs 32 bit. This is
probably what I should revert back to, I just don't think it's working
yet. Sun is famous for their screwed up hardware, though they are more
famous for their buggy CPUs/Memory, which they claim are transient
errors that a reboot will fix.

Personally, I've tried it quite few times, and haven't been able to get
it to work, but I've only tried on Netra t105s and quad hme cards. 




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: sparc64 bridging

2004-05-01 Thread Dan Vande More
On Sat, 2004-05-01 at 12:17, Axel Scheepers wrote:
> Hi,
> 
> I'm not sure if I should post it here, if it's not appropiate please excuse
> me for doing so.
> I'm currently looking into alternatives for solaris to put some ultrasparc
> hardware in use at our isp. We mainly use freebsd, but I'm having major
> problems running the latest releases on this hardware.
> Anyway, I just installed debian on a sparc64 machine and upgraded it to
> testing to be able to run the 2.6.5 kernel. I wanted to use this machine as
> a bridging firewall for some cobalt machines we run, but I can't create a
> bridge.
> I have the bridge module loaded, and 'brctl addbr br0' doesn't give an error
> nor error exit code. When I run 'brctl show' it doesn't show my newly
> created bridge, and I can't create any further configs with it or delete it,
> it just isn't there.
> 
> I'm wondering what I'm doing wrong here, as far as I can tell the commands I
> used are correct. Was it a bad idea to upgrade to 2.6.5? (I really like some
> features of it which imho brings it way closer to a bsd kernel)
> Since I needed to recompile the kernel anyway it seemed like a good choice
> to build 2.6.5, I never tried it with a 2.4.24 kernel.
> 
> To be sure about things I've already tried to
> - enable ip forwarding
> - build it into the kernel instead of a module
> - ifconfig'd ethN promisc up all interfaces
> but all have no effect.
> 
> I'm not (yet) subscribed to this list, could you please cc me in a response?
> 
> Thanks,
> Kind regards,
> 
> Axel Scheepers

Have you already tried these:

http://www.phunnypharm.org/pub/for/sparc-folks/
Ref:
http://lists.debian.org/debian-sparc/2004/debian-sparc-200402/msg00220.html

What hardware are you running on that freebsd isn't compatible with?
Seeing as how it doesn't, it'd be nice of you to post it to the freebsd
list, so others know what's up.

Dan Vande More





Re: sparc64 bridging

2004-05-01 Thread Dan Vande More
On Sat, 2004-05-01 at 12:17, Axel Scheepers wrote:
> Hi,
> 
> I'm not sure if I should post it here, if it's not appropiate please excuse
> me for doing so.
> I'm currently looking into alternatives for solaris to put some ultrasparc
> hardware in use at our isp. We mainly use freebsd, but I'm having major
> problems running the latest releases on this hardware.
> Anyway, I just installed debian on a sparc64 machine and upgraded it to
> testing to be able to run the 2.6.5 kernel. I wanted to use this machine as
> a bridging firewall for some cobalt machines we run, but I can't create a
> bridge.
> I have the bridge module loaded, and 'brctl addbr br0' doesn't give an error
> nor error exit code. When I run 'brctl show' it doesn't show my newly
> created bridge, and I can't create any further configs with it or delete it,
> it just isn't there.
> 
> I'm wondering what I'm doing wrong here, as far as I can tell the commands I
> used are correct. Was it a bad idea to upgrade to 2.6.5? (I really like some
> features of it which imho brings it way closer to a bsd kernel)
> Since I needed to recompile the kernel anyway it seemed like a good choice
> to build 2.6.5, I never tried it with a 2.4.24 kernel.
> 
> To be sure about things I've already tried to
> - enable ip forwarding
> - build it into the kernel instead of a module
> - ifconfig'd ethN promisc up all interfaces
> but all have no effect.
> 
> I'm not (yet) subscribed to this list, could you please cc me in a response?
> 
> Thanks,
> Kind regards,
> 
> Axel Scheepers

Have you already tried these:

http://www.phunnypharm.org/pub/for/sparc-folks/
Ref:
http://lists.debian.org/debian-sparc/2004/debian-sparc-200402/msg00220.html

What hardware are you running on that freebsd isn't compatible with?
Seeing as how it doesn't, it'd be nice of you to post it to the freebsd
list, so others know what's up.

Dan Vande More



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Monitoring software

2004-04-30 Thread Dan Vande More
On Mon, 2004-04-26 at 05:17, Michelle Konzack wrote:
> Am 2004-04-26 12:04:18, schrieb Andreas John:
> >Hi!
> >
> >I guess you will try out several others and end up with nagios :)
> >Try "apt-cache search monit" and you will find some.
> >
> >rgds,
> >j.
> 
> Hello, 
> 
> I think, he will come back to 'nagios'. ;-)
> 
> Greetings
> Michelle

I wonder, has nagios evolved since I've tried it last? Personally, I use
Big Brother. 
IIRC, nagios could check for open ports, and general stuff but nothing
like this:

The reds page me, yellow pages if it's yellow for a while, green leaves
me alone. 


**BB Example***
green Fri Apr 30 20:26:05 MDT 2004 
MySQL Replication is OK.
Client reports Read Master Log Position as 267748466 .

**BB Example***
green Fri Apr 30 20:50:23 MDT 2004 All processes are OK
green clamd >=1 - 1 instance running 
green amavisd >=1 - 11 instances running 
green mysql >=1 - 134 instances running 
green named >=1 - 1 instance running 
green cron >=1 - 1 instance running 
green bbrun >=1 - 3 instances running 
**BB Example***
yellow Fri Apr 30 20:52:02 MDT 2004 up: 3 days, 0 users, 49 procs, load=759
LOAD AVG on backup is 759
**BB Example***
yellow /var/adm/messages (scanned at Fri Apr 30 20:37:59 MDT 2004)
Apr 30 18:48:11 db unix: WARNING: [AFT1] Uncorrectable Memory Error on CPU0 
Data access at TL=0, errID 0x002890d1.a047e8ef
Apr 30 18:48:11 db unix: WARNING: [AFT1] Uncorrectable Memory Error on CPU0 
Data access at TL=0, errID 0x002890d1.a047e8ef
Apr 30 18:48:11 db unix: WARNING: [AFT1] errID 0x002890d1.a047e8ef Syndrome 0x3 
indicates that this may not be a memory module problem
Apr 30 18:48:11 db unix: WARNING: [AFT1] errID 0x002890d1.a047e8ef Syndrome 0x3 
indicates that this may not be a memory module problem
Apr 30 18:48:11 db unix: WARNING: [AFT1] initiating reboot due to above error 
in pid 419 (oracle)
Apr 30 18:48:11 db unix: WARNING: [AFT1] initiating reboot due to above error 
in pid 419 (oracle

**BB Example***
=== Oracle Instance Check 

green Instances specified in ORACLE_SIDS (prod) match those found in 
/var/opt/oracle/oratab

==

green Listener UP

=== Oracle Listener ===


LSNRCTL for Solaris: Version  - Production on 30-APR-2004 20:44:58

(c) Copyright 1998 Oracle Corporation.  All rights reserved.

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROC)))
STATUS of the LISTENER

Alias LISTENER
Version   TNSLSNR for Solaris: - Production
Start Date30-APR-2004 20:39:10
Uptime0 days 0 hr. 5 min. 47 sec
Trace Level   off
Security  OFF
SNMP  OFF
Listener Parameter File   /oracle/product/network/admin/listener.ora
Listener Log File /oracle/product/network/log/listener.log
Services Summary...
  PLSExtProchas 1 service handler(s)
  prod  has 1 service handler(s)
  prod  has 2 service handler(s)
The command completed successfully


===

green Database prod processes UP

== Oracle Instance Processes ==
**BB Example***

HTH

Dan Vande More




Re: Monitoring software

2004-04-30 Thread Dan Vande More
On Mon, 2004-04-26 at 05:17, Michelle Konzack wrote:
> Am 2004-04-26 12:04:18, schrieb Andreas John:
> >Hi!
> >
> >I guess you will try out several others and end up with nagios :)
> >Try "apt-cache search monit" and you will find some.
> >
> >rgds,
> >j.
> 
> Hello, 
> 
> I think, he will come back to 'nagios'. ;-)
> 
> Greetings
> Michelle

I wonder, has nagios evolved since I've tried it last? Personally, I use
Big Brother. 
IIRC, nagios could check for open ports, and general stuff but nothing
like this:

The reds page me, yellow pages if it's yellow for a while, green leaves
me alone. 


**BB Example***
green Fri Apr 30 20:26:05 MDT 2004 
MySQL Replication is OK.
Client reports Read Master Log Position as 267748466 .

**BB Example***
green Fri Apr 30 20:50:23 MDT 2004 All processes are OK
green clamd >=1 - 1 instance running 
green amavisd >=1 - 11 instances running 
green mysql >=1 - 134 instances running 
green named >=1 - 1 instance running 
green cron >=1 - 1 instance running 
green bbrun >=1 - 3 instances running 
**BB Example***
yellow Fri Apr 30 20:52:02 MDT 2004 up: 3 days, 0 users, 49 procs, load=759
LOAD AVG on backup is 759
**BB Example***
yellow /var/adm/messages (scanned at Fri Apr 30 20:37:59 MDT 2004)
Apr 30 18:48:11 db unix: WARNING: [AFT1] Uncorrectable Memory Error on CPU0 Data 
access at TL=0, errID 0x002890d1.a047e8ef
Apr 30 18:48:11 db unix: WARNING: [AFT1] Uncorrectable Memory Error on CPU0 Data 
access at TL=0, errID 0x002890d1.a047e8ef
Apr 30 18:48:11 db unix: WARNING: [AFT1] errID 0x002890d1.a047e8ef Syndrome 0x3 
indicates that this may not be a memory module problem
Apr 30 18:48:11 db unix: WARNING: [AFT1] errID 0x002890d1.a047e8ef Syndrome 0x3 
indicates that this may not be a memory module problem
Apr 30 18:48:11 db unix: WARNING: [AFT1] initiating reboot due to above error in pid 
419 (oracle)
Apr 30 18:48:11 db unix: WARNING: [AFT1] initiating reboot due to above error in pid 
419 (oracle

**BB Example***
=== Oracle Instance Check 

green Instances specified in ORACLE_SIDS (prod) match those found in 
/var/opt/oracle/oratab

==

green Listener UP

=== Oracle Listener ===


LSNRCTL for Solaris: Version  - Production on 30-APR-2004 20:44:58

(c) Copyright 1998 Oracle Corporation.  All rights reserved.

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROC)))
STATUS of the LISTENER

Alias LISTENER
Version   TNSLSNR for Solaris: - Production
Start Date30-APR-2004 20:39:10
Uptime0 days 0 hr. 5 min. 47 sec
Trace Level   off
Security  OFF
SNMP  OFF
Listener Parameter File   /oracle/product/network/admin/listener.ora
Listener Log File /oracle/product/network/log/listener.log
Services Summary...
  PLSExtProchas 1 service handler(s)
  prod  has 1 service handler(s)
  prod  has 2 service handler(s)
The command completed successfully


===

green Database prod processes UP

== Oracle Instance Processes ==
**BB Example***

HTH

Dan Vande More


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]