Re: Effing HAL

[Freminlins]

2009/10/30 Adam Vande More 


> Where's your config files, your errors logs, your stats?
>

I am not going to reply to you after this because you are blinkered. I'm not
going to waste my time with you. I've given you more than enough to feed on,
but you just can't smell the coffee.



> Adam Vande More
>


MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

2009/10/30 Neil Short 

>
> This is comforting for many reasons: notably, that I'm not alone in this
> experience.
>


Do you know something, it's not comforting. Not really. It just means you
are another user for whom a program doesn't work properly.

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

You are arrogant because you won't accept that people have problems with
your beloved HAL, the Half Arsed Luser program. Some or many in the Linux
crowd are planning to move away from it because it doesn't work properly.

You don't accept that a 3k config file now needs 18MB of RAM. And doesn't
work properly into the bargain anyway.

You blame the users.

Let's leave it at that. I have done what is right - I have moved away from
Xorg + HAL which didn't work to Xorg - HAL which words properly. And I've
saved 18MB of RAM.

Goodbye.

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

2009/10/30 Adam Vande More 

> I was part of this and the x11 mailing list during the period in which most
> made the switch to hal.  I am fully aware of all the complaining which
> occurred.  There was a bug which made the issue difficult.  I experienced
> it, the workaround was available nearly immediately and fixed soon after.
> Nearly of the complaints were due to that bug, or misconfiguration just as
> you are experiencing.  The bug was basically moused and hal fighting over
> who was polling the mouse while X was running.  top is a horrible method of
> measuring memory usage by a process.
>

But people are STILL having problems, like me last night. These
"misconfigurations", isn't that what HAL is meant to stop people from doing,
by configuring things itself? I tried with no xorg.conf file at all, I tried
all sorts of things, it was very very ugly. Did you actually read what I
wrote, or did you make up your mind that I had misconfigured something? Let
me repeat it - I was given no keyboard or mouse.




> procstat(1) will give you a much better picture, I suggest you challenge
> your assumptions and explore that path.
>
> However since you asked here is the diff.



You have just proved that you didn't read what I wrote. Let me repeat it:
"Show us the output of top from your box with the hal processes". You have
just showed the headings, which doesn't mean a lot.

I think you have closed your eyes to the problems that people experience
with hal. That is a pity, because it won't improve anything.

-- 
Adam Vande More


MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

2009/10/30 Adam Vande More 

>
> I am unable to replicate this.


YMMV. But I did replicate it. I measured "before and after".

Show us the output of top from your box with the hal processes, as I did for
a start.


> Nowhere have you demonstrated HAL is not working as it's meant to.  This is
> pointless to argue about since it's so easy to debug.  Simply post the X log
> from your original state, and the reason it didn't work will be clearly
> shown.
>

I disagree, and frankly I think you are ignoring facts. My old X worked
fine, X with HAL did not. What can be plainer than that?

Go and Google for hal problems - you will see that this is not an uncommon
occurrence. It seems that the purpose of hal is to make things easier. Well
it didn't for me - it made them harder. I dunno, perhaps I'm too stuck in my
ways. But I've been configuring X for about 12+ years. I had the odd
nightmare about 10 years ago, when I was still a noob, but not since. That
is until last night, when I tried a new spoinky X+hal. Let's face it, a 3k
config file which works perfectly shouldn't need to be replaced with 18MB of
continuously running programs which still needs configuring.

Instead of trying to say that hal works and I haven't demonstrated
otherwise, actually go and look, as I did. Just Google "xorg hal" and you
will see all kinds of probs. Google for "xorg without hal" for some other
people's choice words on problems with hal.

Reading up a little more on HAL today, it makes me laugh and cry. Here's a
few bits with my take:

1. "HAL .fdi files--the new Xorg configure"
2.  "Xorg/hal works but no mouse wheel"
3. "Reclaim your sanity from Xorg and HAL"
4. "Fed up with Xorg + hal mess"
5. "X.Org is well on its way to getting rid of lots of xorg.conf magic and
moving it into obscure elements of HAL"

My interpretation:

1. Great, Xorg doesn't need a config file, but to make it do what you want
you will need config FILES.
2. HAL detection doesn't work properly. Wheel mice are very common.
3. HAL is insane.
4. It's not just me.
5. Why make something obscure? Is obscure better?

Look, I appreciate the good intentions behind HAL, to make X setup very easy
and automatic for as many people as possible. But when it doesn't work
properly, it's no good saying that it does. If people still have problems
with X/HAL/mice/keyboards/keymaps then it should not be called success. We
haven't really gained anything. We have just from learning how to configure
X in Xorg.conf to HAL in other places. Some people would call that
re-inventing the wheel.

-- 
> Adam Vande More
>


MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

2009/10/30 Adam Vande More 

>
> No my point was top is not accurate measure of HAL's memory usage.  HAL has
> shared library's just like many other applications.
>

Yep, I know all about that. But it is indicative. And indeed born out by the
fact that when HAL is not running I get 18MB more memory free.

This is only because of your misinterpretation of data and failure to RTFM.
>

Not entirely true. I didn't misinterpret the data - it was accurate. I
didn't read the FM, but then again if HAL worked as it is meant to, I
shouldn't need to. Isn't that the whole point of HAL? Starting X and finding
no keyboard or mouse working is hardly what I would call success.


> Adam Vande More
>


MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

I've read the responses and comments here, so don't think I'm ignoring
anyone because I haven't responded directly.

I rebuilt xorg-server without HAL. I killed hal stone dead and started up
the new (i.e. old-skool) xorg. It all works fine.

My mouse and keyboard work as specified in the xorg.conf file, rather than
in the new-fangled xml way of doing things or adding setxkbmap to my xinitrc
file. I am also 18MB of  RAM better off. Specifically for Adam, who asks a
rhetorical question about HAL, memory usage and top. The answer for me is
18MB too much.

My advice to anyone who has problems with X and HAL - rebuild xorg-server
without HAL (it doesn't take long),  then start from that base.

I have to say this HAL way of doing things is using a sledgehammer to crack
a nut. Sure X can be a bit horrible to configure, but HAL itself is ugly,
resource hungry and doesn't work 100%. It seems to be an example of
supposedly making things easier, except when it doesn't work.

Life is a calm blue ocean once again.

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Effing HAL

[Freminlins]

2009/10/29 Paul Schmehl 

> Far be it from me to pile on when you're already so frustrated, but I run
> into these sorts of problems myself from time to time.  It's usually because
> I didn't bother to read /usr/ports/UPDATING first, which in this case might
> have warned you.
>


Yeah, thanks for that. I knew about that file, but don't often read it.
There's even more to the saga - Xkblayout doesn't work. This whole HAL thing
stinks horribly. IF X is built with HAL basically certain options specified
in xorg.conf no longer work. HAL thinks it knows best. But it doesn't, cos
it's broken.

What really gets my goat about this is that things that used to work, and
people understand how they worked and how they were configured, no longer
work. And I'm 18MB of RAM worse off into the bargain.

There's a thread about other people's experience here:
http://forums.freebsd.org/showthread.php?p=10924#post10924. One of the posts
contains these words "I've been fighting this one for two days now, and
still don't have a fully working system." That's seriously nasty for anyone.

The whole Xorg thing, at least on FreeBSD, is just a minefield. I like to
remove unnecessary packages, to save space for when I do backups. I don't
have an Nvidia card on this box so:

pkg_delete xf86-video-nv-2.1.13
pkg_delete: package 'xf86-video-nv-2.1.13' is required by these other
packages
and may not be deinstalled:
xorg-drivers-7.4_1
xorg-7.4_1

Great. So what is the point in having a separate package if I can't remove
that damn thing? I know I can pkg_delete -f, but why make it hard?


> Paul Schmehl, Senior Infosec Analyst


MF
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Effing HAL

[Freminlins]

For Christ's sake. I have an IBM X41 laptop, which was happily running
FreeBSD 7.1. Having a little free time this evening I decided to update it
to 7.2. The upgrade failed miserably so I had to install from scratch. But
that's OK, because I had backed up the machine beforehand.

The install went through fine (via NFS after PXE boot - no CDROM on this
box). And then I get to X. As this is a minor update I go to use the old
Xorg config which worked fine. No joy. I get the dreaded "No screens found"
error, although the screen is there of course. After trying X -configure
several times, tweaking, etc., I get nowhere slowly. i810 is now just intel,
but whatever. The screen comes up, but no mouse or keyboard. After yet much
more fiddling and tweaking I finally get to the crux of the problem:
fricking HAL.

Having installed this from scratch using the X User defaults I might have
expected that something that is now required to get the flipping keyboard
and mouse working would be enabled by the installer. But no, I have to waste
time on this crap just to get back to where I was before. Oh, and HAL uses
6MB or so of RAM. Not a lot. But as I already specify the hardware in the
Xorg conf file it is, as far as I can see, an unnecessary waste.

I thought it would be easy, but no, it's just a pain. It's wasn't
unsolvable, as I have years of fiddling with UNIX and FreeBSD in particular.
But for effing Christ's sake.

I know this isn't specifically a FreeBSD problem, HAL being needed by X. But
the flipping installer should enable it when I selected X flipping User from
the install options.

My little upgrade has now turned from a bit of fun into a saga that I don't
want to go through again.

I had to get this off my chest. It was, as they say, doing my head in.

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: flashplugin

[Freminlins]

2009/10/25 Matthew Seaman 


>% pkg_info -r linux-f10-flashplugin-10.0r32
>Information for linux-f10-flashplugin-10.0r32:
>
>Depends on:
>Dependency: linux_base-f10-10_2
>Dependency: linux-f10-openssl-0.9.8g
>Dependency: linux-f10-openldap-2.4.12_1
>Dependency: linux-f10-libssh2-0.18
>Dependency: linux-f10-cyrus-sasl2-2.1.22
>Dependency: linux-f10-curl-7.19.4_4
>Dependency: linux-f10-nspr-4.7.4
>Dependency: linux-f10-sqlite3-3.5.9_1
>Dependency: linux-f10-nss-3.12.2.0
>

Why the hell the Flash plugin (for Linux) needs openldap and sqlite I do not
know. SASL too for that matter.

I must admit I gave up ever getting Flash to work RELIABLY on FreeBSD a long
time ago. It's just too hard, too much work, and not worth the misery of
installing heaps of crud just to get a flipping browser plugin working
unreliably.



> Cheers,
>
>Matthew
>

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Non-root user and accept() or listen()

[Freminlins]

2009/9/14 Chris Rees 

>
> Isn't this a bit drastic? Listening sockets are opened by very many
> types of processes, as well as remembering that sendmail, BIND, and
> others don't actually run as root... I suppose it'd be possible, but
>  would it actually be useful?
>

Sure, those open listening sockets. But those are things I want to listen.

Now suppose a user account was hacked, and "Bob" sets up a web server
listening on some random port above 1024. If "Bob" couldn't use listen() he
wouldn't be able to do that.

Of course, user accounts should be made secure, but what I am getting at is
making the hack much less useful.


> BTW, there may be an ipfw rule for this, I'll have to look it up when
> my servers are back online!
>
> Chris
>

Frem. (Apologies for Gmail quoting, which is horrible).
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Non-root user and accept() or listen()

[Freminlins]

Hi,

I am not sure if this exists (but don't think so), so I am asking.

Is there a sysctl type thing to disallow non-root users, or indeed any
specified user or group, from running a program with listen() ?

What I am looking at is improving network security, such that if a user
account is compromised it can then not be used to run a dodgy web
server/whatever on a non-privileged port. Although I can firewall off any
port I wish, it seems like an obvious thing to disallow any user from
opening a listening socket in the first place. I am suggesting something
like "sysctl user.socket_listen" with enable or disable.

Am I being really daft? Or does this exist already?


Cheers,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Opera 10.00 (native) & flash

[Freminlins]

2009/9/7 Jerry 

> I don't think that readily addresses the OP's question. I personally
> have never gotten 'flash', or most other add-ons to work Opera. It is
> one of the main reasons that I discourage others from using it. It
> suffers even worse on a Windows machine. RoboForm does not work with
> the Opera browser. Opera has a closed architecture that does not allow
>  third party browser extensions
>

Sorry, but that is just bollocks.

Opera and Flash work perfectly well on Windows and Solaris (yep, there's a
native plugin). And Flash is a third party plugin, i.e. it is not provided
by Opera. If Adobe doesn't provide a native Flash plugin for FreeBSD that is
not Opera's fault.


>  I would strongly advocate the use of another browser.


And still there is no native FreeBSD plugin. If you can't get it to work you
should ask for some help.

Jerry
> ges...@yahoo.com
>

MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: 7.2 RELEASE ? Buggy as hell

[Freminlins]

2009/7/30 Mel Flynn

>

>
> You should really be using PCBSD if you want a packaged desktop system, for
> which the developers claim responsibility and for which much (if not all)
> of
> the configuration has been done for you.

I disagree with that. It even says on the FreeBSD web site "FreeBSD® is an
advanced operating system for modern server, desktop...".
I have used FreeBSD on the desktop for about 6 years (but not yet running
7.2). It has mostly been a pleasure. I didn't like it when X was changed to
individual packages, as it now takes considerably longer to install. And the
output from pkg_info takes correspondingly longer to search through. It also
installs two scripting languages (Perl and Python). I haven't had a problem
configuring X for years.
If something has changed which then causes problems to end users, then that
is not good. And it's no good telling people "use PCBSD" or something else.
That's not what we want. We want to use FreeBSD on the desktop.
Don't try and put people off using FreeBSD. It would be much better to help
them resolves the problems they are having.


>
> --
> Mel

MF.

>
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: 10gb network interface suggestions

[Freminlins]

2009/2/7 Wojciech Puchar 

> But also I am open to other brands. Especially If anyone has
>> experience on that this matter would be very very helpful. I have seen
>> that Sun has quad 10gbE adapters:
>> http://www.sun.com/products/networking/ethernet/10gigethernet/index.xml
>> I am not sure if FreeBSD supports that?
>>
>
> could you please explain ANY CASE when 4*10GbE/s could be utilized.
>
> it's 5 GIGABYTES/s each direction. computer memory subsystem is rarely
> faster
>
To be fair, the adaptor on the link is a dual port device (I looked with
interest!), and Sun states the maximum is 16Gb/s aggregrate.

A machine that is multi homed on 10G networks could use all ports. It
doesn't mean it has to do 10G of traffic all the time.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: NFS, how to find out which files are used

[Freminlins]

2009/2/3 Dan Nelson 

> In the last episode (Feb 03), Sandra Kachelmann said:
> > I have an NFS fileserver and would like to figure out which files are
> > being read/written to. Is there something to find that out? Something
> > similar to samba's 'smbstatus' command.
>
> The best you can do currently is run tcpdump/wireshark and watch the remote
> file operations as they happen...  NFS doesn't access files by filename,
> but
> by NFS filehandle (basically device+inode number), so a remote client first
> looks up the filename to get the filehandle, and all accesses are done via
> the filehandle at that point.  Theoretically, one could write a dtrace
> script that watches calls to nfs_namei, nfsrv_read, and nfsrv_write, and
> then matches read/write ops with the filenames that were looked up
> beforehand.

Solaris NFS has a logging option, which does exactly what Sandra is asking
for. It's al reason why I prefer to use Solaris for NFS servers.
F.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: opera 9.63 installation and Qt version

[Freminlins]

Sorry, my bad. I misread you wanted the static version for 7.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: opera 9.63 installation and Qt version

[Freminlins]

2009/1/24 Paul B. Mahol 

>
> Well for 9.63 there is only static one for FreeBSD 5.
> After all this is not freebsd problem, ask opera where is
> static version for FreeBSD 7

ftp://ftp.opera.com/pub/opera/unix/freebsd/963/en/intel/static/
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: sysctl machdep.independent_wallclock

[Freminlins]

2009/1/14 Mister Olli 

> hi...
>
> what is the exact function of this sysctl setting?

I'm guessing it's something to do with Xen, having seen a few references in
Linux for xen.machdep.independent_wallclock.

Have a look here:
http://docs.xensource.com/XenServer/4.0.1/guest/ch04s06.html


>
> I couldn't find any documentation on it.
>
> greetz
> olli
>
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Sysinstall

[Freminlins]

2009/1/8 Jerry McAllister 

>
> Beyond that, it is just the pretty pictures
> that are missing.


sysinstall also works over serial console. No use for pretty pictures
there...


> jerry


MF.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Kernel SMB performance

[Freminlins]

2008/11/21 Ansar Mohammed <[EMAIL PROTECTED]>

> I think you are. We should care about a KERNEL module being bug free and
> high performance.

Agreed.
We did a migration from a Windows email server a while back (about 40,000
mail boxes). As customers logged into the FreeBSD boxes, a process was
kicked off to copy email from the Windows box to the FreeBSD boxes, this
being done using SMB + some parsing in the script. This worked flawlessly
for about 10 million files.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: 7.0 installation, and Xorg in particular

[Freminlins]

On 29/11/2007, Dominic Fandrey <[EMAIL PROTECTED]> wrote:
>
> > So I do a pkg_add -r xorg. After about 70 packages I give up. I only
> used to
> > have about 65 packages in total on my old desktop, now I need more than
> 70
> > and I haven't even got x windows up yet. So I go off and have a look and
> > discover that x.org 7.x is modular - "
>
> Which is a blessing for maintenance.


But not for an end user. Who really can keep track of 300 packages? Who has
some ports installed, changes one of them which is a dependency and finds
something else breaks? Now this will happen to X.org.

It's not like there's more stuff being installed. Only you can be more
> selective about which parts you need and don't need. And you don't have to
> rebuild all of xorg for a little update.


But it is FAR more complicated. I stopped installing it after about 70
packages. I noticed it installed Cyrillic and Ethopic (didn't even know
there was one) fonts. I didn't want either of them. In the "old days" I
simply ensured that Cyrillic fonts were unchecked. Now I am supposed to go
through 300 packages. It might be more selective but it is not easier.

That's what the x11/xorg meta-port is for.


Well the reasoning behind it is broken, and so is its implementation (see
above about the unwanted fonts).


> I suggest you take a look at what depends on them (pkg_info). You will
> have
> the answer, then. None of them is required to run FreeBSD.


Err, yeah. Look through hundreds of packages to see which dependencies they
have. Helpful. Not.

This way of doing X11 is seriously unhelpful to end users. If having
individual packages for everything is so good, please tell me why everything
in /bin, /usr/bin and so on is not an individual package. It's because the
idea of doing so is dumb.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

7.0 installation, and Xorg in particular

[Freminlins]

I used to find FreeBSD easy. What has happened? I have a couple of machines
I usually install new versions on, one is headless the other is a desktop
machine (which was a 100% reliable 5.4 installation). I boot the headless
machine using floppies, then install across the net. But something has
happened such that I now need five floppies, and I have to put the boot one
in at least twice. This wasn't the case previously. It now reminds me of an
OS/2 installation with its floppy shuffling.

Then for my desktop machine. sysinstall crashes if I try to install x.org.
So I do a pkg_add -r xorg. After about 70 packages I give up. I only used to
have about 65 packages in total on my old desktop, now I need more than 70
and I haven't even got x windows up yet. So I go off and have a look and
discover that x.org 7.x is modular - "
http://blogs.ittoolbox.com/unix/bsd/archives/xorg-72-on-freebsd-13661";. This
fellow is talking about 300 packages just for x.org! This is nuts. No two
ways about it. Whoever decided to do this needs their head (or heads)
examined. It used to be so simple. Now it's not. If x.org didn't work for
some reason I wouldn't want to track down which of hundreds of packages is
missing. Who would? Also, I noticed that python as well as perl was being
installed. Is not one scripting language enough for x.org? Why are two
needed?

I am really frustrated. I don't understand how installing X* this way is
supposed to be an improvement. What does it actually give me that I didn't
have before? Note my old system was reliable, as is my desktop at work (a
6.2 machine). I was so frustrated that I gave up installing 7 on my home
desktop and am now in Windows land. It just seems so pointless. It reminds
me of the nastiness of Gnome, which has bazillions of packages, and Gnome
needs nearly all of them so why make them separate?

I've done enough head banging tonight. Maybe Xfree86 is still available. I
haven't looked yet.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: How to use dump?

[Freminlins]

Hi,

On 29/08/2007, Wojciech Puchar <[EMAIL PROTECTED]> wrote:
>
> dump doesn't copy files to files, but files to raw device (partition,
> tape, DVD) or to one/few big files.


Dump is used to back up a file system and can write that data to a file. It
doesn't have to write to a raw device.

To dump /home to /backup, do something like "dump 0auf /backup/home.dump
/home".

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

ipfw - limit other networks

[Freminlins]

Hi,

I am trying to limit the number of connections from "foreign" networks to a
server. I don't want to limit bandwidth, just the number of connections.
Let's say I have a network 192.168.1.0/24. I want to allow 192.168.2.0/24 to
have at most 50 connections. I want to allow 192.168.3.0/24 to have 20
connections. And so on. Is this even possible? Some applications can do this
but I would prefer to do this at the network level.

I can limit connections on a per IP basis easily, but that isn't what I am
looking for.


Cheers,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: serious performance problems with 6.2 Release

[Freminlins]

Ted,

On 16/02/07, Ted Mittelstaedt <[EMAIL PROTECTED]> wrote:



I don't know where your getting the impression that I said this was a
hardware bug.



Umm, quoted from you above: "Defects that are specific to hardware that are
not documented in the PR database generally do not get fixed. "

If I didn't know this is simply the way you are at times I would think you
have gone mad.

Ted


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: serious performance problems with 6.2 Release

[Freminlins]

Ted,

On 15/02/07, Ted Mittelstaedt <[EMAIL PROTECTED]> wrote:



We are all ears for your suggestions to help him fix this, Frem.  I'm sure
we
all expect to see some kernel patches from you any day now.



Please sort out your formatting. It looks horrible.

You didn't offer any help whatsoever. And who says this is a kernel problem
anyway?

Please review the charter of this list.  If this was supposed to be fixed on

a
mailling list, freebsd-bugs would be at least a bit closer to the mark..



Please sort out your formatting. It still looks horrible.

The original post is completely on topic. I suggest you go and read it again
like I did.

I've snipped your assumption that this is a hardware problem because it is
misleading at this stage. It could well be a configuraiton issue.


Ted


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: serious performance problems with 6.2 Release

[Freminlins]

On 15/02/07, Ted Mittelstaedt <[EMAIL PROTECTED]> wrote:


please use send-pr and include a dmesg output with debugging turned on,
and exact model of motherboard and bios revision.

questions isn't for bugs.  I don't mean to be rude but you won't get the
problem fixed by bitching about it on this mailing list.



Ignore Ted.

There is nothing wrong with your post to questions. There wasn't any
bitching. Your post was very appropriate. Indeed, all you askedin the end
was "please help". You won't get that from Ted.

Ted


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: mknod, devfs and FreeBSD

[Freminlins]

Ruben,

On 31/01/07, Ruben de Groot <[EMAIL PROTECTED]> wrote:


Not exactly. Solaris, especially Solaris 10 is relying more and more on
pseudo filesystems.

# uname -srpi
SunOS 5.10 sparc SUNW,Sun-Fire-V210
# mount | grep '^/devices'
/devices on /devices read/write/setuid/devices/dev=47c on Thu Nov  2
16:14:25 2006

Everything in /dev is just a symlink to /devices.




That's true BUT I can still use mknod to create a device node elsewhere and
it works.

I'm not complaining about devfs, just that I would be forced to use devfs on
FreeBSD when IMHO mknod would suffice and used to suffice.

--

Ruben




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: mknod, devfs and FreeBSD

[Freminlins]

Kris,

On 29/01/07, Kris Kennaway <[EMAIL PROTECTED]> wrote:



To put it bluntly, it's something you're just going to have to get
over :-)



That's unhelpful. It is, in my opinion, a bad idea to have to mount up 1400
instances of devfs just to get a few device nodes. It just doesn't seem
right. It's a kludge. What I will do instread is migrate the box to Solaris
where I can do what I want to do.

It's a poor argument to say basically "that's the way it is". I have always
found FreeBSD to be flexible, not restrictive.

If devfs is the only way to go, why does mknod still exist? Why does it
allow me to create device nodes that don't work?

Kris


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: mknod, devfs and FreeBSD

[Freminlins]

Kris,

On 28/01/07, Kris Kennaway <[EMAIL PROTECTED]> wrote:



I not understand this no sentence :)



Sorry, I didn't read what I typed. I meant to type "Was the effect of this
considered at all?"

What reasons, other than cosmetic, do you have for not wanting to do

this?



Well, I am sure you would agree it is simpler to mknod for a small subset of
/dev than to mount a devfs. Also, it means I have to migrate my existing set
up which works perfectly as it is.

It isn't just cosmetic, it really is more awkward than running mknod. I take
your point that there's no technical reason not to do this, but it isn't
pretty.

Kris


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: mknod, devfs and FreeBSD

[Freminlins]

Kris,

On 26/01/07, Kris Kennaway <[EMAIL PROTECTED]> wrote:



Sorry, it's the only way.



Was the considered at all? There's simply no way that I would mount up 1400
devfs. It is a backward step.

Kris





Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: mknod, devfs and FreeBSD

[Freminlins]

Kris,

On 26/01/07, Kris Kennaway <[EMAIL PROTECTED]> wrote:


Set up and mount numerous devfs file systems ;)



That is exactly what I am trying to avoid. One of the servers has 1400 sites
on it, and I really don't want 1400 devfs mounts. If the only way to do this
now is by having so many devfs mounts I am better off not upgrading, and it
is very arugable that FreeBSD has lost some functionality by forcing such a
scheme.

Really it's not hard, you just specify the devices you want with a

simple devfs(8) ruleset.



It's not how hard it is, it's how untidy it is.

Kris


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

mknod, devfs and FreeBSD

[Freminlins]

Hello,

I have a web server still running FreeBSD 4.7 which I want to update to
FreeBSD 6.2. There are quite a few sites on this machine, and each of them
has a chroot containing their own /dev. In their /dev are things like null,
zero, random and so on.

I don't really want to set up or mount numerous devfs file systems. I tried
creating the the relevent files using mknod but they don't work. What is the
best way to proceed?

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: is THIS why the 6.2 release seems stalled ?

[Freminlins]

On 11/01/07, Jeff Mohler <[EMAIL PROTECTED]> wrote:



The basic reason is that a ../.. walk invalidates cached metadata, and
you end up with a pipe full of getattr's all of the time.  Freebsd-fs
has discussed this a bit, but no fixing is coming soon.  We use linux
to compile builds, we'd like to use Freebsd, but linux on Filers via
NFS is about 3x faster than the same builds on Fbsd to the same filer.
  ../.. baby.




Did you try different mount options on the FreeBSD clients. I have no idea,
but Linux may have different defaults.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: is THIS why the 6.2 release seems stalled ?

[Freminlins]

On 10/01/07, Chris <[EMAIL PROTECTED]> wrote:



A reason why you have less problems is I expect you using premium
hardware such as scsi, currently I am lucky enough to not be using
realtek lan cards although I am still having problems with intel nics.



I wouldn't term SCSI as premium. Maybe it used to be, but these days
machines are so cheap anyway. Nearly all our x86 boxes have Intel NICs. I
haven't had problems with them.

the specific nfs issues are related to mounting linux filesystems, I

am not the only one there is dozens of posts on these mailing lists
from users with the same problems, usually livelocks or panics caused
by mounting nfs filesystems on freebsd most seem to have no
resolution.



Funnily enough, I thought you were going to say you were mounting a Linux
NFS server. It is not surprising that Linux client to Linux server goes
together better than FreeBSD client to Linux server. It could of course be
the Linux NFS server implementation that is buggy, rather than the FreeBSD
client. As I've said, mounting NetApp filers I have no problems at all.

realtek isnt great hardware but is that a good reason for realtek

performing significantly worse then on linux, shouldnt it be on par?



I don't know. I haven't compared them. They are simply not high performance
network cards. I

issues of performance been worse, the biggest example is probably

mysql and uniprocessor performance, I understand with ule 2.0 mysql
performance is signficantly better so there is hope there, I would
like to see more performance from uniprocessor and the mp safe support
on nics set to disabled by default to put stability first.



Well, I agree with you on this. MySQL performance on FreeBSD is acceptable
for my purposes (not usually intensive), but it is not as good as Linux.
I've read as much about this as possible, and tried using options and thread
libraries. But this has not fixed the problem.

But, on many other things I don't see a performance problem at all. I think
it's important to give exact examples rather than saying "performance has
been worse". If you said "MySQL performance is worse". I've seen a
performance problem with ClamAV. Funnily enough, both ClamAV and MySQL are
threaded applications, so I'm guessing that the FreeBSD threading is the
source (cause) of the performance problem for these apps.

The installer well that comes down to using a variety of datacentres,

quite often datacentre staff are not too well trained and mainly used
to redhat and windows gui installers, so when it comes to freebsd
there is many datacentres who dont even support freebsd when I ask is
because they say it wont install, the ones that do support freebsd the
feedback I get off them is often related to both the installer been a
pain for them and hardware compatibility.



One of our Windows techies learned how to do a FreeBSD install in fifteen
minutes. If someone really cannot learn it, they shouldn't be anywhere near
datacentres. If they can't handle the FreeBSD install, I have no idea how
they would handle Solaris, which is much less friendly and definitely
belongs in  datacentres. And Solaris on non-Sun hardware has less
compatibiltiy than FreeBSD.

How much testing goes into heavy workloads such as heavy apache loads

and DDOS attacks?



I don't know. But you are free to vounteer to do this!

I expect my server to not livelock and come back to

responsiveness after such loads without having to reboot it.  Freebsd
4.x was incredibly stable under heavy ddos attacks, freebsd 5.x held
out but of course was very slow on UP, freebsd 6.x is faster but has
suffered stability problems.



I agree that the 4.x series was (is) very stable - we still have some in
use.


On the performance side get the sata and raid problems sorted for

improved hd performance tagged ququeing etc.

Is there a sort of hire a dev button on the freebsd website?



I guess you could make a contribution to the foundation.

Chris




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: is THIS why the 6.2 release seems stalled ?

[Freminlins]

On 10/01/07, Chris <[EMAIL PROTECTED]> wrote:



What I think freebsd needs.

1 - To fix stuff that works in linux but goes to crap in freebsd, one
such example is NFS.



I don't actually have a problem with FreeBSD and NFS. This is using about
20+ clients and 2 NetApp filers. What problem are you having, rather than
just "goes to crap"?

2 - A better installer, this is probably the biggest single thing that

puts people of freebsd, the less people using freebsd the less funds
likely to be recieved.



Do you really think this puts people off? How do you know this? I find that
FreeBSD has the fastest installer for a complete OS. It may not look very
pretty, but it's logical and I can use it over serial console.

3 - Better hardware compatability, freebsd has poor hardware support,

worst then both linux and windows, one such example is realtek cards
have no problems in windows and linux but do in freebsd, sata support
is very poor as well.  Often when people say anything the response is
go out and buy premium hardware.



Well, everything has worse hardware support than Windows, so that's a
pointless comparison. Linux has more hardware support, this is true. But
some of this "support" is very bad. I have only had one box out of about 100
over the years which I couldn't make work to my liking with FreeBSD. Then I
went to Linux. I use Linux on a couple of large storage systems deliberately
because of the lack of a journalling FS in FreeBSD.

Realtek network cards, I am afraid to say, are cheap and nasty. Some, at
least, do work in FreeBSD. But I wouldn't consider them professional cards
for use with any OS. They are OK for home users with low end network needs.
Nothing mor than that.

SATA support was definitely a bid dodgy the last time I looked (about a
year), but I nearly always use SCSI so it's not usually a problem.

4 - Better attitude to bug fixing, not always possible to provide

backtraces as such from remote servers but can tell the devs how to
repeat problem so they can do on their own local machines.



That's asking a tiny bit too much IMHO. To get the best help you have to
offer the best information yourself. You are expecting someone else to
repeat your problems on hardware which they not have.

5 - journaling filesystem.


Yes, definitely this is needed and long overdue.

Ultimately I think freebsd is in real danger of losing its stable tag,

more and more things are not stable on freebsd as they get ignored,
some of the network drivers appear to be poorly maintained, and its
lagging behind in the performance charts.  All this considering it
used to be ahead in the game is a sad state of affairs.

Would I pay for freebsd? yes but in the right circumstances I fully
understand its voluntary work in many cases but the worst it gets the
less inclined people will be to pay.



The simple truth is, of course, that Linux has a large number of fully paid
developers. RedHat, for example, employs numerous programmers who fix bugs,
and these fixes end up in free distros of Linux.

Chris





Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: solaris

[Freminlins]

On 06/09/06, White Hat <[EMAIL PROTECTED]> wrote:



That is a totally unqualified evaluation.



No it's not. It's in response to YOUR comment that "A very large majority of
users simply want to use their PCs for email, occasional word processing and
possible game playing". And OpenOffice fits that bill.



While it may
be totally suitable for one individual, that in no way
infers that it meets the requirements of another.
There is no way you can define an end users
requirements based solely on your own usage.



It's based on YOUR assertion, not mine.

You are kidding right. I can find vastly more

documentation available for a win32 machine than for
FBSD.



Where?

In fact, the lact of documentation is one of the

reasons that support groups like this evolved.



And they exist for Windows users too. And they exist for the same reasons.




It is above average, I will agree. However, if it were
really perfect then this forum would not exist.



I didn't say it was perfect. But even if it was that doesn't wollow that
this group wouldn't exist.


If that were true, MS would not rule 90+ percent of

the PCs in use today. Why do you think users in third
rate countries pirate MS when they could get FBSD for
free?



Because Windows is pre-installed for one.



I would not want to insult anyone; however, if
you cannot install an MS operating system then perhaps
you should consider another hobby. Even my wife's
sister can handle that project, and that is a woman
who considers a can opener a high tech device.



I didn't say anything about installing - you keep going off topic - I said
using. And plenty of people do have problems using Windows. If they didn't,
there wouldn't be a need for all the call centre armies whose job is
basically helping Windows users.



You have users here with 10+ years experience who run
int problems. It is just the nature of the beast. It
comes with the territory.



That's right - it's because Windows is not as easy to use (or rather fix
when broken) as you are making out.


Obviously it required installation. Before you can

install, it is again obvious that you must secure the
item. One size definitely does not fit all. What is
your point?



I am contradicting your point. That is my point.


Norton is pathetic, that I will agree with you on that

one.



I didn't say it was pathetic, so please don't suggest that I did. I said it
had problems. It is one of the most popular products in its field.


That is why I switched three years ago to ZA. It

has never given me a moment of trouble, although the
CA AV it uses by default is not RFC 2595 compliant
which was causing my network problems. One I corrected
it though, everything was back to normal.

BTW, 'time consuming for your techies'? Ah gee, like
what are they paid for? To stand around and kiss each
others butt. I am sick of over paid techies who have
no working knowledge of what they are doing. If they
find their job to stressful, quit!



You clearly are way out of your depth and have no understanding of what
techies do. They should be spending time helping customers use our services,
NOT doing free tech support for Microsoft. And that is basically what they
spend a lot of time doing.


Please do me one favor, do not CC me. I am continually

getting two copies of these. I subscribe to the list.
I don't send you duplicate copies and therefore would
appreciate the same cutesy. Perhaps my address was
already inserted by a previous poster. If so, please
do remove it.



Removed.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: solaris

[Freminlins]

On 06/09/06, White Hat <[EMAIL PROTECTED]> wrote:



Immaterial. the singularly most important feature is
suitability to task. If it is free and it does not
work, what good is it?



It depends what you are using it for. You made a comment about "occaisonal
word processing" (pasted below). For such use OpenOffice is perfectly good
enough.



Yes, the lack of documentation is a shame.



In Windows, yes. In FreeBSD I can't see a lack.



The same lack of documentation
plagues every facet of software today.



No it doesn't. FreeBSD is well documented.


However, you have made my point.


No I haven't. I have contradicted your point. You said " A very large
majority of users simply want to use their PCs for email, occasional word
processing and possible game playing." I am saying that using XP as you
suggested is not as easy as you suggest for a very large number of people.


If a user cannot

decipher how to configure a simple thing like Outlook
Express, and there are programs available that will do
it for them, then how are they suppose to be capable
of handling a CLI OS like FreeBSD? It boggles the mind
-- at least mine. Worse, the configuration of OE is
handled by a wizard. It is truly sad when a user
cannot configure something when it is simplified down
to that level.



It's not so much the wizards, but third party applications like virus
scanners which change those settings which is a part of the problem. But you
are not quite comparing apples with apples. Configuring Thunderbird on
FreeBSD is near enough identical to doing the same on Windows. I wouldn't
however expect a complete computer novice to be able to set up a FreeBSD box
without some help.


How? Drop in two CDs or download the programs, run

them and case closed. Neither one requires any
significant configuration. The defaults work just fine
for most users. You could eliminate the Counter Spy
since ZA has its own proprietary SpyWare program, but
I just happen to prefer Counter Spy.



Your statement is simply wrong. AV and anti-spyware DO require
configuration. And they do require installing, and maybe downloading, and
being kept up to date. The defaults certainly don't work all the time in all
cases. Have a look here: "
http://www.theregister.co.uk/2006/09/06/faulty_ca_update/";. I have heard of
broken installations for Norton numerous times. And trying to help these
customers is time-consuming for our techies.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: solaris

[Freminlins]

On 06/09/06, White Hat <[EMAIL PROTECTED]> wrote:


 I have
tried Open Office. No matter what anyone says, it is
just not as full featured as Word 2003. It is not even
close.



True, but also compare the cost. Not even close...

He/she does

not want to read tons of manuals and spend hours in a
frustrating attempt to get it to run.



This is where you are completely wrong. I work for an ISP. I'm not
responsible for tech support but I keep my "ear to the ground". A VERY large
number of callers have problems configuring Outlook Express, for example. No
matter what the polls say, the experience is often very different. They may
not read the manuals (because they are no longer supplied), they just ring a
call centre instead.

The average user

does not care about configuring firewall, AV or
Spyware, etc. Just drop in a copy of ZA with perhaps
Sunbelt's Counter Spy and they are on their way.



That's one statement contradicting the other.



White Hat




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Large File System?

[Freminlins]

On 08/08/06, Chad Leigh -- Shire.Net LLC <[EMAIL PROTECTED]> wrote:


Right now, if no fsck is really really important to you for your data
store, then get an OpenSolaris system and put ZFS on it.  Never fsck
again as it is ALWAYS (they claim) in a coherent state.  Or wait for
ZFS to show up on FreeBSD



Indeed. However as most of my platform is running FreeBSD the problem
doesn't go away.

Not just for the above reasons, I am implementing a Solaris server

with 1.7TB on ZFS and sharing it to a bunch of FreeBSD machines over
nfs on dedicated gigabit with jumbo frames on separate interfaces
from the standard default interface.  (My main reason was to not have
storage tied to an individual worker server)



I would have used Solaris for this a while ago, but there were no drivers
for the RAID card :-( Hence, Linux

Chad



Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Large File System?

[Freminlins]

On 08/08/06, Martin Hepworth <[EMAIL PROTECTED]> wrote:


Softupdates are the FreeBSD equivalent. From my point of view they perform
better than a traditional journaling FS (do a google search for the original
usenix papers on these).



Journalling means not having to fsck the file system in the event of an
unclean shutdown. So it's wrong to describe softupdates as equivalent. It's
not.

I also find they speed up I/O quite alot, esp for fast changing filesystems

like mail spools.



Certainly I have found using softupdates to be considerably faster than
without.

martin




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Large File System?

[Freminlins]

On 08/08/06, Atom Powers <[EMAIL PROTECTED]> wrote:


What exactly does a journaling file system give you? As I understand
it, it doesn't prevent corruption and it doesn't help you fix the
corruption when it occurs.



As answered by Dan Nelson. It saves time (sometimes a lot) in the event of
an unclean shutdown/

File server. It's a Promise VTrak M300p, SCSI attached storage.

Frankly I'm more worried about the system crashing than the storage
device ( UPS, battery pack for the RAID controller, redundant power
supplies ).



Yes, I had all that. It is of absolutely no use in the event of an unclean
shutdown (on FreeBSD). If the file system itself is dirty, it will need to
fsckd. The bigger the file system, the longer it takes (generall). That is
what journalling saves you.

To give you some indication of what this means in real life, I'll refer
(again, sorry) to a power outage we suffered in our colo. This is FreeBSD on
modern hardware:

Jul 23 17:52:05 weeble kernel: WARNING: /var was not properly dismounted
Jul 23 17:55:52 weeble fsck: /dev/aacd0s1f: 1352 files, 956469 used,
13988364 free (1484 frags, 1748360
blocks, 0.0% fragmentation)

I've snipped out the logs in between. But that's nearly 4 minutes to get
itself sorted out. That file system has only 1.9GB of data. Our Solaris
boxes came up straight away.


2.5TB file system, used mostly for archival storage. Of the initial

2.0TB of data, I expect only about 2-3% to change on a weekly basis.

I have another 1.5TB Fiber Channel cabinet that I plan on using to
store roaming profiles ( for MS Win uers) and home drives (for *nix
users).



Depending on the size of the files, you may wish to use a different block
size.  Also I've found that on large file systems you may wish to change is
the "minfree" setting. You can do this either when running newfs, or tunefs.
The default setting is 8%. On 2.5TB file systems that's a lot. If you don't
know what this means check out the man pages.

That is about the only "management" you will need to do.

Perfection is just a word I use occasionally with mustard.

--Atom Powers--




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Large File System?

[Freminlins]

On 08/08/06, Atom Powers <[EMAIL PROTECTED]> wrote:



Thanks. I found my problem. (Sysinstall, aka fdisk, won't do more that
1.2TB.)
BTW, anybody have any good advice on how to manage a large file system?



Unfortunately I have to say "consider Solaris or Linux as they have
journalling file systems."

Although I have a couple of big file systems on FreeBSD, it is not a pretty
sight if there is some sort of problem. Recently our colo lost power. The
two boxes that needed manually fixing were the two big file system boxes.
Background fsck did not fix them. To compare, we have one almost identical
box running Linux. It came straight back up courtesy of ext3.

Ignoring all the suggestions to get UPS (the colo had generator backed UPS
which failed), etc, problems can/do happen. And when they do, journalling
for big file systems is very useful.

The single most important thing missing for me in FreeBSD is a journalling
file system as I would use it on every box.

You don't need to do anything more to manage big file systems per se. How
big a file system are you going to create? What are you going to use it for?
That might help with suggestions.

--Atom Powers--





Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: 4TB filesystem

[Freminlins]

On 05/08/06, Martin Hepworth <[EMAIL PROTECTED]> wrote:


Softupdates removes the issue if havinh to fsck filesystems after and
unclean umount.



No it doesn't. Absolutely not.

After an unclean shutdown fsck runs in the background. And sometimes it
can't do that. Here's an example:

Jul 23 17:54:01 zoe fsck: /dev/amrd0s1g: CANNOT CREATE SNAPSHOT
/d1/.snap/fsck_snapshot: File too large
Jul 23 17:54:01 zoe fsck:
Jul 23 17:54:01 zoe fsck: /dev/amrd0s1g: UNEXPECTED INCONSISTENCY; RUN fsck
MANUALLY.


The /etc/defaults/rc.conf file has a comment on this too:
background_fsck="YES"   # Attempt to run fsck in the background where
possible.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Midnight Commander in base distribution set

[Freminlins]

On 04/08/06, Joseph Le-Phan <[EMAIL PROTECTED]> wrote:


Personally, it'd be fantastic if mc was slated for inclusion. It's an
absolute necessary install once I get a system up and running.



The thing is, once you go down this route it won't stop. Other people regard
bash, or lsof, or vim, or wget as an "absolute necessary install". bash gets
put on all our boxes because three of my colleagues prefer it. For them it
is essential - they won't use csh. I regard Tcl as an "absolute necessary
install" as that is my chosen scripting language. I put it on every box I
use (even Windows). And that's the problem. As I personally don't use mc
it's inclusion would be wasted on me. And if Tcl was included it would be
wasted on many (most!) other people.

I have to say I absolutely enjoy using FreeBSD, more so than any other OS.
And part of that joy comes from the few minutes installation for the base
OS.

I used to use Redhat. My first outing with that was version 4.2. Then as
time went on it no longer fitted on a single CD (after 6.2 IIRC.) And the
clutter and junk it installed (presumably because someone thought it was
essential or useful) was just appalling.

IMHO the base install is best left lean like it is now - the only additions
should provide additional OS functionality, not user functionality.

Joseph Le-Phan  [GPG key: 292E09A0]



My 2 pence.
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Need Advice for Tuning NFS to place nice with a Netapp

[Freminlins]

On 02/08/06, N. Harrington <[EMAIL PROTECTED]> wrote:

Cool! Can you share with me what sort of settings you

use on your boxes? sysctl/kerneltunes/mount options?



This may be a disappointment to you but... I didn't have to do anything :-(
All I have is rw on the client.

It has taken me a over a month to even get to speak

to someone high enough up he food chain at Netapp to
not say "FreeBSD - that's a version of Linux right?"



It depends who you speak to. There are people at NetApp who know about
FreeBSD.

The web server replies (using either Apache and

Lighthttpd) seem to max out at about 17mb/s. Response
time for the web server will rise gradually, then
suddenly become 10-20seconds for a reply. Much like a
backup on a highway. They claim that the netapp unit
is spending too much time dealing with file
information IOPS than actual transfer of files.
However even on a non in-use server, if I make a
request for a file, that "heavy file access" seems
normal.
IE:
GtAttr Lookup Rdlink   Read  Write Rename Access
Rddir
0  0  0  0  0  0  0  0
248  160  0  4  0  0236  0
0  0  0  0  0  0  0  0



I would dispute NetApp's claim. NetApp filers are very capable at doing NFS
operations. Static files tend to sit usefully in the buffer cache on web
servers.  So unless you are doing something really odd with your web servers
I would tend to disagree with NetApp.

I've just looked on one of our POP3 servers (mounting NetApps). POP3 causes
far more random access than our web servers. As such it doesn't sit in the
buffer cache very long.

We have much higher figures than yours and absolutely no performance
problems.

Yes, 4X GigE from the filer via a Vif and trunking on

the switch. A nice 10Gb ready HP unit. I have asked if
using the Vif and trunking could have any effects but
been assured it should not. It does mean I cannot use
jumbo frames. But since web pages and images are
small, I don't think there would be any benefit.



There is nothing wrong in theory with that setup. But is may not be what you
want. Try it with just one GigE interface.

You're right - you probably don't need jumbo frames.

Which interface does the HP unit have? Also, have a look at netstat -in. Are
there any IErrs or OErrs or Coll? Paste the results here!

9907187 bytes/sec for a 16M file.

It will transfer in nanotime. So, I believe that
eliminates network performance as an issue.



Well, not really. The figure above is showing < 10MB a second. That's not
quite Fast Ethernet speed. If you are pushing 17mb (I guess that's megabits)
that's not really a problem though.

I've just tested this on the same POP3 server above, using dd to write a
file onto a NetApp and I get 10889359 a second. And this machine is busy.
Also, it is mounting the NetApp over Fast Ethernet.

Hmm. Drat. We have some web servers that do nothing

but send out data, but some that are used for
uploading and file manipulation. I will have to make
sure that global of an option will not effect what
they do.



It is a per volume option. And frankly I've never seen much use for atime.
It's useful sometimes, but not a lot.

Can you also put in the output of nfsstat -W -c 2. Maybe it's best to put
this up on the web somewhere as it's wide, and it's not easy to read in
email. Let it run for a minute or so, and if possible do two runs. One
during the OK time, the other during the problem time.

I would go back to basics. One GigE interface. Just rw mount options, and
start testing. By testing I mean measuring. NFS tuning is fiddly. I've been
using NetApps with FreeBSD for 5 years. It is a good combination.

Can you also post the output of sysctl -a|grep nfs. But don't start fiddling
with them yet!



One last thing - are you female?! In a UNIX
> newsgroup?!

Yup :)
Oh, and yes, I do play the drums :)



Oh gawd.  Whatever next? :-)

Thanks for your assistance!!


  Nicole




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Need Advice for Tuning NFS to place nice with a Netapp

[Freminlins]

Nicole,

On 02/08/06, N. Harrington <[EMAIL PROTECTED]> wrote:


Hi
I have several web servers that are attached to a
Netapp (network appliance) unit via NFS-3. A few
servers are 5.5 and a few are 6.1 for comparison
testing. All seem to have lousy performance.



We have a similar setup and it runs smoothly.

Can you define "lousy performance" ?

Can you give more details on your network? Are you using Gig ethernet? And
over what medium?

Can you also try just copying a 100MB file from the filer to one of the web
servers and record the time?

Are you running nfsiod?

When

going through the issues with Netapp, the reasons
given were that we have too many GettAdr/Lookup
requests compared to actual reads. So all the NFS IOPS
are being used up by these requests. As soon as the
webservers get busy,  requests pile up.

I have tried everything I can think of. The web
servers are even mounted read only with no help.

My current mount options are:
filer:/vol/fvol31/home/13/13  nfs
ro,noatime,-r=32768,-T,-b,-R0,-i,-D2,-L 0  0



Mounting noatime for web servers is a good idea but... your "noatime" option
has no effect on NFS mounts (check out the mount man page). You need "vol
options no_atime_update" on the NetApp.

Any advice for sysctl tunes or anything else would be

much appreciatted!

Thanks

  Nicole



One last thing - are you female?! In a UNIX newsgroup?!

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: switching from linux to freebsd

[Freminlins]

On 01/08/06, Erik Nørgaard <[EMAIL PROTECTED]> wrote:

If you configure your server using LDAP or NIS for user management then

you only need to mount the root file system rw when updating the base
system or changing root password. Add the MAC and you will likely be
able to protect further against the attack you mention.




Or when you want to patch or install other software, unless you put
/usr/local on its own partition. And put /usr/ports somewhere else. And
don't tinker with anything in /etc/mail. I think we're just going to
disagree on this.

I have never yet seen a situation where mounting the OS disk ro proved to be
useful. I have seen it hinder perfectly normal sysadmin work.

I have seen one instance in 10 years where it would have stopped a silly
mistake (someone moved libc on Solaris). But as that person was doing
something they were supposed to be doing and just made a mistake, they would
have made the same mistake after mounting the disk rw if it had been mounted
ro.

Cheers, Erik


Cheers,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: switching from linux to freebsd

[Freminlins]

On 01/08/06, Robert Huff <[EMAIL PROTECTED]> wrote:


On my system, "additional software" goes under the separate
partition /usr.  Or are we using different definitions of
"additional"?




/usr includes a large part of the base installation. /usr/local is the usual
place for additional software, though you can of course install stuff where
you want to.


   Robert Huff



Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: switching from linux to freebsd

[Freminlins]

On 01/08/06, Erik Nørgaard <[EMAIL PROTECTED]> wrote:

You usually don't patch up your system everyday. Remount rw do the

patching and remount ro. The problem is more that some 3rd party
applications assume that /usr is writeable. I found the problem more
annoying with / whenever I need to change some system file.




I still disagree. The base OS files which need protecting are already
protected sufficiently. If you don't agree with this then simply remounting
ro is not sufficient. Only with elevated securelevels would this be useful.
Else, anyone who gets root on the box can simply remount rw and do what they
will.

However, most important is to have /tmp on a separate partition. Then

there will only be few writes on /.



Except for useful things like installing additional software. That is
something I do do regularly.

I think it is very valuable to get the system up so I can rescue my

data. Having base system go down along with my data doesn't seem to have
any clear advantages




Mounting / and/or /usr ro will get your systems up faster and that

seemed to be the issue.



You made the point with reference to security, not system recovery. That is
what I am contradicting.

Cheers, Erik




Cheers,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: switching from linux to freebsd

[Freminlins]

On 01/08/06, Erik Norgaard <[EMAIL PROTECTED]> wrote:


you may

even want to mount it read-only for security. (I think these are good
advises on any system).



I used to agree with this (specifically the mantra was "mount /usr read
only") - until I tried to patch anything! Then it's useless.

What you end up with is a machine which in which the base install is more
secure, but all your data isn't. The base install is the one thing I know I
can get back (i.e. reinstall) in 5 minutes. The data I cannot.


Cheers, Erik




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

ntpd on FreeBSD

[Freminlins]

Hello,

After a power outaga (Level 3 at Goswell Road, London), all our FreeBSD
machines came up OK.

Nearly all of them had a problem though with ntpd. I'm guessing that most of
these machines booted before the the ntp servers came up. What happens is
that the machine runs two copies of ntpd:

root   337  0.0  0.3  2964  1772  ??  Ss   Sun04PM
0:24.75/usr/sbin/ntpd -c /etc/ntp.conf -p /var/run/ntpd.pid
root   427  0.0  0.3  2964  1788  ??  SSun04PM
0:00.76/usr/sbin/ntpd -c /etc/ntp.conf -p /var/run/ntpd.pid

... and they never sync. The only way to fix this is to kill both ntpds,
then restart ntpd.

Is there a tidy way round this? It's not much fun logging into 40+ machines
and killing a restarting a key process. I have no idea why two ntpds are
running in the first place. The machines that are correct have an identical
config.


Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Are hardware vendors starting to bail on FreeBSD ... ?

[Freminlins]

Nikolas,

On 24/07/06, Nikolas Britton <[EMAIL PROTECTED]> wrote:



This would be like running Windows 3.1 on a brand new Xeon 5100
dual-core CPU... sure it will run fast* but what the hell are you
going to do with it? Play solitaire?



You have this the wrong way round. The correct allusion would surely be
something like "imagine running  XP on a 80386", not an old OS on new
hardware. Old OSs don't always run at all on new hardware.

Anyway, I am sure that Ted can speak for himself.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Are hardware vendors starting to bail on FreeBSD ... ?

[Freminlins]

Ted,

On 24/07/06, Ted Mittelstaedt <[EMAIL PROTECTED]> wrote:




All you have to do to see this is try booting FBSD 6 on a 80386
and compare it's performance to FBSD 3.X on a 386.




How are you going to do that, Ted? From the 6.0R release notes: "Support for
80386 processors (the I386_CPU kernel configuration option) has been
removed. Users running this class of CPU should use FreeBSD 5.*X* or
earlier."


Ted



Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: What FreeBSD users really want

[Freminlins]

On 22/07/06, sammy sumer <[EMAIL PROTECTED]> wrote:


To Whom It May Concern:





1.Reinvent the installer and interface.


Fundamental thing like system installer is still phenomenally arcane.
There
is no excuse for FreeBSD developers not to upgrade the system installer
and
why not using disk imaging technology like Norton ghost or Acronis
TrueImageinstead of the traditional installation.



One thing I would say is that FreeBSD installs a complete operating system
far faster than any other OS out there. This does matter to some people,
though not everyone.

A few years ago I was new to FreeBSD (and UNIX/Linux in general) and I went
through the installation. The only thing that caught me out was adding a
user (me) but not putting myself in the wheel group. After the installation
completed I removed the monitor and plugged it back into my usual desktop
machine. I could SSH in but not su. It really was the only thing that caught
me out.

2.Integrate a PHP shell into the core of the system.


PHP is by far the most popular computing language in the world. Why not
have a shell called PHP shell. So lots of web developers out there can
easily create shell scripts in PHP syntax to automate and run programs on
FreeBSD.

Who wants to learn bash or sh scripting? They are by far the least popular
and ugly programming language in the world.

It is astounding that FreeBSD developers have not clued in to the fact
that
millions of backend webmasters could easily migrate and adopt FreeBSD as
their O.S of their choice because of PHP




Would that be PHP and all its associated modules from the base install?
That's big, and for many people unnecessary. Why would I need it for my (10)
mail servers. It wouldn't serve any purpose. Also PHP "syntax" is not
consistent. Take a look here: http://tnx.nl/php.

sh scripting is a low level "standard" across other Unices.

The only thing I wish I had learned so much sooner was "set autolist" in my
.cshrc. I didn't know it was there, and I have no idea why it is not in the
default dot.cshrc file. No doubt good reasons, but I "got by" for months
before I found this out. All that time I was going "bash can do it, why
can't csh?"


Sammy Sumer




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Sendmail = mind boggling

[Freminlins]

Hello,

I've always steered clear of using Sendmail (prefering Exim), but I thought
I would give it a try and am having real difficulties with something I think
should be simple.

I want to keep running Exim on port 25. I want Sendmail to run on a
different port. So I am trying to use the  "O DaemonPortOptions=Port=2525"
option. However, I have 6 cf files in /etc/mail - freebsd.cf,
mail.xxx.co.uk.submit.cf, freebsd.submit.cf, sendmail.cf, mail.xxx.co.uk.cfand
submit.cf.  I've tried all sorts of combinations of adding the above option
to these files, but whenever I try to start Sendmail this error:
sm-mta[52968]: NOQUEUE: SYSERR(root): opendaemonsocket: daemon Daemon0:
cannot bind: Address already in use.

Which file(s) am I supposed to put this option into? I just don't
understand. Is it correct, also, that I have 6 cf files? I set this up as
per the handbook. At this rate I know why I've always used Exim :-)

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: named always binds to "*"

[Freminlins]

Oh, sorry. You already have that. It should work. It does for me.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

ntpd as a server on 5.4 and 6.0 just doesn't seem to work

[Freminlins]

Hello,

I have this problem whereby I just cannot make ntpd work as a server on
FreeBSD 5.4 and 6.0. It works flawlessly on 4.6.2, which I'm still running
somewhere.

The contents of my ntpd.conf file are:
server ntp0.bris.ac.uk
server ntp.linx.net
restrict A.B.0.0 mask 255.255.0.0 notrust nomodify notrap (munged).

statistics clockstats
statsdir /var/db/ntpd
filegen clockstats
driftfile /etc/ntp.drift

I've even tried removing the "restritct" line but to no avail. ntpd runs,
but it doesn't actually return anything.  The response on any machine
querying this machine is "ntpdate[46934]: no server suitable for
synchronization found". The same command works for the older box running
4.6.2. I don't believe it is at all a hardware issue as I've had exactly the
same problem on another machine. I've looked at the handbook here  "
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-ntp.html";
but it doesn't seem to suggest I have done anything wrong.

Does anyone have a clue what the problem is? I'm just getting nowhere.
Traffic gets to the box (seen via tcpdump). Alas ntpd is not compiled with
-ddebug so the -d and -D options in the man page is moot.


Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: pam_userdb.so: Where is it?

[Freminlins]

Kyrre,

On 5/10/06, Kyrre Nygard <[EMAIL PROTECTED]> wrote:



Hello!

Does anybody know where pam_userdb.so has gone?



FreeBSD doesn't appear to have ever had it, so it hasn't "gone" anywhere.
The thread you linked to below suggests exactly that.


Linux has it, but apparently FreeBSD does not.

I need it to set up virtual users with vsftpd.

I've been in contact with others with the same problem:


http://lists.freebsd.org/pipermail/freebsd-questions/2005-November/104571.html

As well as Freenode #vsftpd.

But nobody seems to know what the PAM module for Berkeley DB files is at.

Or perhaps somebody can suggest alternate methods?



You could download the source  (
http://cvs.sourceforge.net/viewcvs.py/pam/Linux-PAM/modules/pam_userdb/) and
try and build it.

Thanks,

Kyrre




Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: filling up UDP socket buffers like mad

[Freminlins]

Michael,

On 3/27/06, Michael W. Lucas <[EMAIL PROTECTED]> wrote:
>
> # netstat -na
> udp4   43414  0  *.514  *.*


That's a big queue.

I'm attaching the output of netstat -na and netstat -s for general
> informative purposes; if anyone has any further suggestions, I'm all
> ears.



Try running syslogd with -n.

Thanks,
> ==ml


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Process stuck in START state

[Freminlins]

Hello,

I have a process which I cannot kill. It is stuck in the START state. Top
shows this:

37028 frem 1 1000 0K 0K START5:20  6.54% acroread

and ps shows this:
frem37028  6.5  0.0 0 0  v0  RE2:40PM   5:20.33 [acroread]

There is no entry for this pid under /proc.


How do I actually kill this process? It's been running for over a day now.
I'm using FreeBSD 6.0 Release.



Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Dell SCSI problem (reasonly long with some debugging)

[Freminlins]

A quick follow up to my earlier post...

I tried using FreeBSD 6.0 stable Snap 10, but the problem still exists.

I tried installing Solaris 10 (01/06) and the card is not detected by
Solaris at all.

I tried with RHEL 4 and Win 2003 and the machine works fine.

Back on FreeBSD 6.0 release, I lowered the "tags" using camcontrol:
camcontrol tags da0 -N 64. The problem then seems to go away. At least
I can't provoke it into occurring. Previously if I built a kernel, or
untarred the ports tarball the machine would "freeze".

Naturally calling Dell support was a pointless exercise. They had
never even heard of "Free biscuit".

Does anyone have any feedback on what the problem may be, and ideally
a fix? Is it "safe" to run a machine in production with my lowered
tags? Also, is there a way to set the tags during the boot process? At
the moment I have a script in /usr/local/etc/rc.d which does this but
it doesn't seem ideal.

I'm guessing this has happened because it's a OEM version of the
Adaptec card and has presumably been modified in some way.

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Dell SCSI problem (reasonly long with some debugging)

[Freminlins]

Hello,

I have FreeBSD installed on well over a dozen Dells (plus 50 other makes)
and they run without any problem. Today I have a Dell arrive and I am
getting very serious SCSI problems with it duing and after installing
FreeBSD 6 release. Basically the machine "freezes" and them comes back to
life after about 30-40 seconds, then a few minutes later it "freezes" again.
During the "freeze" the machine is incapable of any sort of I/O. In dmesg I
then see a load of SCSI information (posted below).  I've also posted the
relevent info from pciconf.

Is this faulty SCSI equipment or is is a driver issue?

Thanks,
Frem.


dmesg:
ahd0: FIFO0 Free, LONGJMP == 0x825e, SCB 0x21
SEQIMODE[0x3f]:(ENCFG4TCMD|ENCFG4ICMD|ENCFG4TSTAT|ENCFG4ISTAT|ENCFG4DATA|ENSAVEPTRS)

SEQINTSRC[0x0] DFCNTRL[0x0] DFSTATUS[0x89]:(FIFOEMP|HDONE|PRELOAD_AVAIL)
SG_CACHE_SHADOW[0x2]:(LAST_SEG) SG_STATE[0x0] DFFSXFRCTL[0x0]
SOFFCNT[0x0] MDFFSTAT[0x5]:(FIFOFREE|DLZERO) SHADDR = 0x00, SHCNT = 0x0
HADDR = 0x00, HCNT = 0x0 CCSGCTL[0x10]:(SG_CACHE_AVAIL)

ahd0: FIFO1 Free, LONGJMP == 0x8063, SCB 0xf
SEQIMODE[0x3f]:(ENCFG4TCMD|ENCFG4ICMD|ENCFG4TSTAT|ENCFG4ISTAT|ENCFG4DATA|ENSAVEPTRS)

SEQINTSRC[0x0] DFCNTRL[0x0] DFSTATUS[0x89]:(FIFOEMP|HDONE|PRELOAD_AVAIL)
SG_CACHE_SHADOW[0x2]:(LAST_SEG) SG_STATE[0x0] DFFSXFRCTL[0x0]
SOFFCNT[0x0] MDFFSTAT[0x5]:(FIFOFREE|DLZERO) SHADDR = 0x00, SHCNT = 0x0
HADDR = 0x00, HCNT = 0x0 CCSGCTL[0x10]:(SG_CACHE_AVAIL)
LQIN: 0x8 0x0 0x0 0x21 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0
0x0 0x0 0x0
ahd0: LQISTATE = 0x1, LQOSTATE = 0x19, OPTIONMODE = 0x42
ahd0: OS_SPACE_CNT = 0x1f MAXCMDCNT = 0x4d
ahd0: SAVED_SCSIID = 0x0 SAVED_LUN = 0x0

SIMODE0[0xc]:(ENOVERRUN|ENIOERR)
CCSCBCTL[0x4]:(CCSCBDIR)
ahd0: REG0 == 0x21, SINDEX = 0x102, DINDEX = 0x102
ahd0: SCBPTR == 0x21, SCB_NEXT == 0x15, SCB_NEXT2 == 0xff2d
CDB 2a 0 0 61 7 73
STACK: 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0
< Dump Card State Ends >>
(da0:ahd0:0:0:0): SCB 65 - timed out
(da0:ahd0:0:0:0): Other SCB Timeout
(da0:ahd0:0:0:0): No other SCB worth waiting for...
ahd0: Issued Channel A Bus Reset. 90 SCBs aborted
Copied 18 bytes of sense data offset 12: 0x70 0x0 0x6 0x0 0x0 0x0 0x0 0xa
0x0 0x0 0x0 0x0 0x29 0x2 0x2 0x0 0x0 0x0
(da0:ahd0:0:0:0): WRITE(10). CDB: 2a 0 0 61 7 73 0 0 8 0
(da0:ahd0:0:0:0): CAM Status: SCSI Status Error
(da0:ahd0:0:0:0): SCSI Status: Check Condition
(da0:ahd0:0:0:0): UNIT ATTENTION asc:29,2
(da0:ahd0:0:0:0): Scsi bus reset occurred field replaceable unit: 2
(da0:ahd0:0:0:0): Retrying Command (per Sense Data)
ahd0: Recovery Initiated - Card was not paused
>> Dump Card State Begins <
ahd0: Dumping Card State at program address 0x4 Mode 0x22
INTSTAT[0x0] SELOID[0x0] SELID[0x0] HS_MAILBOX[0x0]
INTCTL[0x80]:(SWTMINTMASK) SEQINTSTAT[0x0] SAVED_MODE[0x11]
DFFSTAT[0x33]:(CURRFIFO_NONE|FIFO0FREE|FIFO1FREE)
SCSISIGI[0x27]:(P_DATAOUT_DT|ACKI|REQI|BSYI) SCSIPHASE[0x0]
SCSIBUS[0x18] LASTPHASE[0x1]:(P_DATAOUT|P_BUSFREE)
SCSISEQ0[0x40]:(ENSELO) SCSISEQ1[0x12]:(ENAUTOATNP|ENRSELI)
SEQCTL0[0x0] SEQINTCTL[0x0] SEQ_FLAGS[0x0] SEQ_FLAGS2[0x0]
QFREEZE_COUNT[0x6] KERNEL_QFREEZE_COUNT[0x6] MK_MESSAGE_SCB[0x4a]
MK_MESSAGE_SCSIID[0x7] SSTAT0[0x10]:(SELINGO) SSTAT1[0x8]:(BUSFREE)
SSTAT2[0x0] SSTAT3[0x0] PERRDIAG[0x0]
SIMODE1[0xa4]:(ENSCSIPERR|ENSCSIRST|ENSELTIMO)
LQISTAT0[0x0] LQISTAT1[0x0] LQISTAT2[0x80]:(PACKETIZED)
LQOSTAT0[0x0] LQOSTAT1[0x0] LQOSTAT2[0xc0]

SCB Count = 160 CMDS_PENDING = 128 LASTSCB 0x45 CURRSCB 0x63 NEXTSCB 0x47
qinstart = 20204 qinfifonext = 20204
QINFIFO:
WAITING_TID_QUEUES:
   0 ( 0x67 0x33 0x2a 0x3 0x57 0x3c 0x5d 0x7 0x29 0x2c 0x1a 0x4e 0xb
0x20 0x42 0x39 0x5e 0x70 0x4b 0x7a 0x49 0x22 0x34 0x18 0
x71 0x58 0x3f 0x62 0x31 0x43 0xf 0x7f 0x48 0x6c 0x4c 0x6d 0x56 0x3b 0x6f
0x1e 0x79 0x7e 0x19 0x1d 0x10 0x69 0x50 0x8 0x5a 0x25 0x
3e 0x24 0x28 0x52 0x16 0x4a 0x75 0x3a 0x1 0x4f 0x0 0x54 0x2b 0x6b 0x26 0x36
0x76 0x74 0x3d 0xe 0x5 0x44 0x30 0x7b 0x4 0x2 0x45 0x
63 0x47 0x66 0x72 0x13 0x73 0x64 0x55 0x14 0x27 0x65 0x5c 0x7c 0x5b 0x77 0xc
0x1b 0x2f 0x12 0x46 0x32 0x23 0x38 0x35 0x1c 0x51 0x
d 0x21 0x37 0x7d 0x68 0x59 0x15 0x78 0x9 0x6 0x41 0xa 0x17 0x61 0x40 0x11
0x2e 0x1f 0x53 0x6a 0x6e 0x60 0x2d 0x5f 0x4d )
Pending list:
 77 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 95 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 45 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 96 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
110 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
106 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 83 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 31 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 46 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 17 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 64 FIFO_USE[0x0] SCB_CONTROL[0x60]:(TAG_ENB|DISCENB) SCB_SCSIID[0x7]
 97 FIFO_USE[0x0] S

Re: overloaded webserver: nfs wait issue?

[Freminlins]

Can you provide the output of nfsstat -c 1 on the web servers (and nfsstat
-s 1 on the nfs server if they are running FreeBSD)? Run them for a minute
when you are getting bad performance.

Please also quantify "moderately high traffic". What is this in Mbit/s?
Also, it may be helpful to know a little bit about your network
infrastructure, i.e. are the switch ports set to full duplex and the
machines to match. NFS performance is very dependent on a fast reliable
network. Don't consider anything less than 100Mbit/s full duplex. Also what
platform is the NFS server?

We recently migrated from a cluster pair of NetApps to a single NFS server
running FreeBSD. This is mounted by four NFS clients running Zeus. We have a
lot of NFS experience and we felt confident this would work. It did and with
the most marginal degradation in performance.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Serial console prob on amd64

[Freminlins]

Hello Arden,

On 11/24/05, arden <[EMAIL PROTECTED]> wrote:
> has this ever worked? If its a new box could be a hardware prob could try a
> loop-back test if you have the wrap plugs

Yes, on Solaris 10 before I wiped it today. I just can't see what I am
doing wrong. During the boot up sequence I can see and choose from the
FreeBSD "menu". It just stalls later on. When I say "what am I doing
wrong" I don't really mean that as I know I've done it right!

> Arden

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Serial console prob on amd64

[Freminlins]

Hello,

I have a dual amd64 machine on which serial console is not working
properly. I've configured dozens of Intel machines without a problem.

I have set up boot.config and /etc/ttys. If I boot the machine some
data is printed to console. Rather than paste the whole lot here,
here's the last few lines:

Timecounters tick every 1.000 msec
ad0: 78167MB  at ata0-master UDMA133
ad1: 78167MB  at ata0-slave UDMA133
acd0: CDROM  at ata1-master UDMA33
SMP: AP CPU #1 Launched!
Trying to mount root from ufs:/dev/ad0s1a
bge0: link state changed to DOWN
bge1: link state changed to DOWN

And that's it. Serial console is frozen, but the machine boots up and runs fine.

Any suggestions?

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Status of 6.0 for production systems

[Freminlins]

Ted,

On 11/22/05, Ted Mittelstaedt <[EMAIL PROTECTED]> wrote:
[snipped a massive load of nonsense]

Why don't you do us all a favour and shut up. Your posts are off-topic
and a waste of storage bytes. AFAIK this mailing list is not your
personal soap box.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

dump/restore puzzle

[Freminlins]

I have a puzzling problem with dump and restore. I'm looking to implement a
dump and restore pipe to automatically make copy of a file system onto
another system completely. I've used / only as an example (because it's
small) and I'm not overwriting /.

I do the following:
1. Level 0 dump and restore pipe which works as expected

frem# dump 0auLf - / | ( restore -rf - )
DUMP: Date of this level 0 dump: Thu Oct 6 16:16:17 2005
DUMP: Date of last level 0 dump: the epoch
DUMP: Dumping snapshot of /dev/ad0s1a (/) to standard output
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 54582 tape blocks.
DUMP: dumping (Pass III) [directories]
DUMP: dumping (Pass IV) [regular files]
expected next file 8069, got 818
DUMP: DUMP: 55429 tape blocks
DUMP: finished in 19 seconds, throughput 2917 KBytes/sec
DUMP: level 0 dump on Thu Oct 6 16:16:17 2005
DUMP: DUMP IS DONE


2. I touch a file, just to make a difference, then do a level 1 dump and
restore which works as expected (the new file is extracted).

frem# touch /hello
frem# dump 1auLf - / | ( restore -rf - )
DUMP: Date of this level 1 dump: Thu Oct 6 16:16:50 2005
DUMP: Date of last level 0 dump: Thu Oct 6 16:16:17 2005
DUMP: Dumping snapshot of /dev/ad0s1a (/) to standard output
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 36 tape blocks.
DUMP: dumping (Pass III) [directories]
DUMP: dumping (Pass IV) [regular files]
DUMP: DUMP: 35 tape blocks
DUMP: finished in less than a second
DUMP: level 1 dump on Thu Oct 6 16:16:50 2005
DUMP: DUMP IS DONE
expected next file 24550, got 819

3. I remove the file and do another level 1 dump and restore. This doesn't
work.

frem# rm /hello
frem# dump 1auLf - / | ( restore -rf - )
DUMP: Date of this level 1 dump: Thu Oct 6 16:17:08 2005
DUMP: Date of last level 0 dump: Thu Oct 6 16:16:17 2005
DUMP: Dumping snapshot of /dev/ad0s1a (/) to standard output
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 35 tape blocks.
DUMP: dumping (Pass III) [directories]
DUMP: dumping (Pass IV) [regular files]
DUMP: DUMP: 34 tape blocks
DUMP: finished in less than a second
DUMP: level 1 dump on Thu Oct 6 16:17:08 2005
DUMP: DUMP IS DONE
Incremental tape too high

4. If I then touch another file, this still doesn't work.

frem# touch /hello2
frem# dump 1auLf - / | ( restore -rf - )
DUMP: Date of this level 1 dump: Thu Oct 6 16:24:51 2005
DUMP: Date of last level 0 dump: Thu Oct 6 16:23:50 2005
DUMP: Dumping snapshot of /dev/ad0s1a (/) to standard output
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 37 tape blocks.
DUMP: dumping (Pass III) [directories]
DUMP: dumping (Pass IV) [regular files]
DUMP: DUMP: 36 tape blocks
DUMP: finished in less than a second
DUMP: level 1 dump on Thu Oct 6 16:24:51 2005
DUMP: DUMP IS DONE
Incremental tape too high


How can I go about achieving this (ignoring rsync, etc),

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Turning off file flags during restore

[Freminlins]

Hello,

I have some data on a NetApp which I am in the process of migrating to a 
machine running FreeBSD. I dump the data off the NetApp to one big file on 
the FreeBSD machine. When I restore the data I get some odd file flags like 
this: 
-rw-r--r-- 1 www www schg 73 Jul 27 22:04 foo.txt

I understand that some of the attributes on the NetApp may be confusing 
restore. I don't really have a problem with that. Is there a way to not set 
the file flags during restore? I don't really want to chflag a few million 
files if I don't have to. Also, of course, changing the flags updates the 
file's ctime.

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

/sbin/restore on 6.0 beta 3

[Freminlins]

Hello,

I'm not sure where I should send this. /sbin/restore on FreeBSD 6.0
beta 3 does not recognise dump files off a netapp:

zoe# /sbin/restore -if ./vol1_LEVEL1_2005-08-23 
Tape is not a dump tape

restore on a FreeBSD 4.7 box does however and works fine.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: disk fragmentation, <0%?

[Freminlins]

On 8/15/05, Jerry McAllister <[EMAIL PROTECTED]> wrote:
> >

> As someone mentioned, there is a FAQ on this.   You should read it.
> 
> It is going negative because you have used more than the nominal
> capacity of the slice.   The nominal capacity is the total space
> minus the reserved proportion (usually 8%) that is held out.
> Root is able to write to that space and you have done something
> that got root to write beyond the nominal space.

I'm not sure you are right in this case. I think you need to re-read
the post. I've quoted the relevent part here:
 
> > Filesystem SizeUsed   Avail Capacity  Mounted on
> > /dev/ar0s1e248M   -278K228M-0%/tmp

Looking at how the columns line up I have to state that I too have
never seen this behaviour.  As an experiment I over-filled a file
system and here's the results:

Filesystem SizeUsed   Avail Capacity  Mounted on
/dev/ad0s1f965M895M   -7.4M   101%/tmp

Note capacity is not negative. So that makes three of us in this
thread who have not seen negative capacity on UFS.

I have seen negative capacity when running an old version of FreeBSD
with a very large NFS mount (not enough bits in statfs if I remember
correctly).

> jerry

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Crashes on 5.4R

[Freminlins]

Kris,

On 8/1/05, Kris Kennaway <[EMAIL PROTECTED]> wrote:

> See the chapter on kernel debugging in the developers' handbook.

It should be obvious from my post that I did look at the handbook. I
have a dump and I've extracted the line where the kernel has crashed.
I do have a backtrace (not posted) .My question is "what am I supposed
to do next?" I'm not a kernel programmer.

> Kris

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Crashes on 5.4R

[Freminlins]

Hello,

I have a machine which I've installed afresh with 5.4R last week. It
is a dual Xeon machine with hyperthreading. Since I've updated it, it
has crashed regularly. So regularly that I have had to remove it from
service.

I built a debugging kernel and from my last crash dump I get the following:

#0  doadump () at pcpu.h:159
159 __asm __volatile("movl %%fs:0,%0" : "=r" (td));

Now, I'm in the fortunate position of not having FreeBSD crash
usually, so I'm not sure what to do next.

Can someone offer some help?

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: defragmentation in FreeBSD 4.11

[Freminlins]

Gayn Winters wrote:
 
> What I get from reading this article is that if the use of the file
> system is to store lots of small files, then use a small block size.  Am
> I missing something?

No and yes! There is a minimum block and fragment size. In this case
there were not enough contiguous fragments to enable an 8K file to be
created. Without checking I believe Solaris uses 8K blocks.

> Also, in most situations, buying a big enough disk is far better than
> worrying about what happens when a not-big-enough disk starts to get
> full.

Indeed. But... in the case I linked to there was apparently plenty of
free space, just not enough free contiguous space. The author also
implies that a bigger disk would not solve the problem:

  "it creates and deletes tons of small files and thus the fragmentation 
  over a period of time."

As mentioned though I have never seen this myself despite running very
busy mail and web servers with what must be billions of files being
created/deleted in that time.

It certainly grabbed my attention so I thought it may be of interest to others.

> -gayn

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: defragmentation in FreeBSD 4.11

[Freminlins]

On 7/28/05, Bob Johnson <[EMAIL PROTECTED]> wrote:

> > Why is it unnecessary to defragment UFS?
> >
> 
> In normal use, files never become fragmented enough to affect performance.  In
> a (loose) sense, files are intentionally fragmented in a controlled way so
> that fragmentation doesn't cause problems.  If you run fsck on a partition,
> you will typically see fragmentation levels of less than one percent.
> 
> Also, keep in mind that in the default formatting, a FreeBSD partition has 8%
> of the disk space withheld from normal users to help keep the disk from
> becoming so full the system can't operate, and it has the side effect of
> helping to prevent fragmentation as well.  It is why df can show a disk being
> as much as 108% full.  It is possible to make this space available for normal
> use if, for example, you are using a partition only for data storage and you
> want to squeeze every last bit of space out of it, but of course there will
> be some performance penalty as it starts to get full.  You can also adjust
> other disk parameters to optimize for your particular needs.  See tunefs(8).
> 
> If the disk gets close enough to full, the OS has no choice but to start
> fragmenting things.  Try to keep your disks less than about 90% full (that's
> a number I remember from somewhere -- it's just a guideline and not a firm
> limit).  My /home partition is 95% full according to df (which means it is
> actually a little under 90% full including the reserved capacity), and fsck
> shows 0.1% fragmentation.  Of course, it's a fairly big partition, so it
> still has over a gigabyte of free space.  Even the ISO CD images I downloaded
> a few days ago probably didn't get much fragmentation.

Apparently UFS can get fragmented even when there is lots of
apparently free space:
http://blogs.sun.com/roller/page/sprakki?entry=ufs_file_system_defragmentation

Obviously this is Sun UFS, but there is a common heritage. I've never
needed to do any sort of defrag on either FreeBSD or Solaris though.

> - Bob

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Question about FreeBsd 5.4 + BIND 9.3.1 and threads

[Freminlins]

Hello,

I have a FreeBSD 5.4R box running BIND 9.3.1 from base on a dual Xeon
with hyperthreading enabled.

According to the man page for named:

   -n #cpus
  Create #cpus worker threads to take advantage of multiple  CPUs.
  If not specified, named will try to determine the number of CPUs
  present and create one thread per  CPU.   If  it  is  unable  to
  determine  the  number  of  CPUs, a single worker thread will be
  created.

Yet I seem to have six threads:

mx01$ ps -xauwH | grep bind
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named
bind   95906  0.0  5.4 58444 56876  ??  SLs   1:16PM 143:20.01
/usr/sbin/named -u bind -t /var/named


Does this mean there is a bug in BIND, or am I missing something?

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Puzzling nfs/rpc log message

[Freminlins]

Hello,

I have a number of FreeBSD machines mounting two NetApps via NFS. I
updated one of the machines to 5.4 release. I have no prolems mounting
the filers, but I am getting the following log entry every so often:

  Jul 27 16:36:33 fred rpcbind: connect from 192.168.1.5 to
getport/addr(nlockmgr): request from unauthorized host

I am only getting this for one of the filers. They are both accessed
about the same amount, and are identically configured AFAIK.

Can anyone shed any light on this? I've Googled a bit with no luck.

Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

dump with -L flag - machine truely unresponsive

[Freminlins]

I have a FreeBSD 5.4 machine with a SCSI hard drive. I dump all our
machines onto a nfs-mounted filer. There are some files which it's
important not to clobber during the dump, so I've been experimenting
with dump using the -L flag like so: dump auLf /var /mnt/dump/var

What I have noticed is that during the time that the machine is taking
its snapshot, it becomes completely unreachable on the net. It is
impossible to ssh or even ping it. Our monitoring reports that the
machine has dropped off the net, but once the snapshot is finished it
is available again. The machine itself does not drop, it just appears
to drop.

Has anyone seen this before? I certainly haven't. And if so, does
anyone have a fix (apart from going back to dump without -L)?

For info the partitions are suspended during the snapshot for these
amounts of time:
Jun 21 03:00:08 gia kernel: /: suspended 7.003 sec, redo 0 of 4
Jun 21 03:08:41 gia kernel: /var: suspended 57.036 sec, redo 25 of 2775
Jun 21 03:22:19 gia kernel: /home: suspended 1.037 sec, redo 0 of 4
Jun 21 03:23:56 gia kernel: /usr: suspended 40.246 sec, redo 0 of 14


Thanks,
Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: set-uid bit: where am I going wrong?

[Freminlins]

On Apr 11, 2005 2:51 PM, Tim Stephens <[EMAIL PROTECTED]> wrote:

> Clearly the file is owned by root, and I kept it as part of my group. I've 
> read the man pages, and believe that when I call the script, it will assume 
> root's permissions. It doesn't, so where am I going wrong?

FreeBSD does not support setuid scripts. They are inherently insecure.
You have some options though to your problem. You could run the script
directly as root, which is what you are trying to do. Or you could
write a wrapper round your script, which may seem like overkill.

Given that you trust your script enough to try to run it setuid, I
would go for the first option. Make sure the script cannot be altered
by anyone other than root, then run it as root.

> Thanks,
> Tim

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: possible bug report re: (malformed?) internet addresses

[Freminlins]

Interesting. Host can look it up:
bash-2.05b$ host mr-chips-.deviantart.com
mr-chips-.deviantart.com has address 69.28.181.43

But the host name is itself invalid. From RFC 1035:

  The labels must follow the rules for ARPANET host names.  They must
  start with a letter, end with a letter or digit, and have as interior
  characters only letters, digits, and hyphen.

My guess is Windows is actually munging it, when it would be better to
return an error.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Wed, 23 Mar 2005 03:46:40 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:

> But I don't really know for sure, because nothing is documented, and
> nobody here knows anything.

So you keep telling us. So why do you bother posting in the first
place if you don't expect an answer that pleases you?

> Anthony

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Tue, 22 Mar 2005 12:51:36 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:

> And _hope_?

Yes, hope someone looks into it. You get the support you paid for...

> --
> Anthony


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Tue, 22 Mar 2005 11:26:13 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:
> Freminlins writes:

> No, FreeBSD doesn't work very well with the hardware.  As a matter of
> fact, it doesn't work very well with the hardware on my production
> server, either.

So, you seem to have a problem. 
 
> It seems that as soon as you install FreeBSD on a machine, the
> "hardware" fails.

Not true. I look after 60 or so FreeBSD machines. They run fine. I did
have a problem with a SCSI RAID on FreeBSD. I called the supplier and
they took a look - the firmware on one of the drives was different to
the others. They updated it and haven't had a problem since.

> What would you suggest that I do?  It takes a very long time to wade
> through OS source code.

Review the supported hardware list on the FreeBSD site. If the
hardware is not on there, stop complaining. If it's supposed to be
supported, raise it a PR and hope someone looks at it.

> --
> Anthony


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Tue, 22 Mar 2005 10:25:14 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:

> Exactly.  With _identical_ hardware.  So if the hardware ran under the
> other OS, but not under this OS, where do you look first for the
> problem?

Both, actually.
 
> If your car runs perfectly for years with one brand of oil, and then you
> change brands and the engine seizes, where do you look for the source of
> the problem?

Both, actually.

> I'm just pointing out the unavoidable implication of what you said.  NT
> ran on this hardware for eight years without a hitch; FreeBSD cannot do
> the same.  It's not the hardware.

So stick with NT. Why would you change from something that runs
perfectly for 8 years?

> UNIX is twenty years older than NT.

That doesn't mean nothing has changed in 20 years, does it? 

> I wanted to try FreeBSD.

And your hardware seems to not work very well with FreeBSD. Move on.
 
> Is that what you tell people who have trouble getting FreeBSD to work?
> "Reinstall your old OS"?

That's not what has been said. Having read this thread (and others by
you previously) you have no intention of helping yourself. You just
like to whinge that FreeBSD doen't work on your hardware.

> --
> Anthony

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Tue, 22 Mar 2005 04:19:25 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:

> It ran for eight years without errors.

On a different OS.
 
> So your saying an anciety copy of NT is more reliable than a current
> copy of FreeBSD?

Don't try and put your words in my mouth. On your ancient hardware
with an ancient OS you didn't have problems. Why not stick with it if
it's been so reliable?
 
> Anthony

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Anthony's drive issues.Re: ssh password delay

[Freminlins]

On Mon, 21 Mar 2005 19:39:11 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:

> Show me that it's not a bug in FreeBSD first.  

Alternatively, show us it is not a firmware problem first.

> I never had the problem
> in Windows NT.

Yawn. I had loads of problems with NT, virtually none with Win2K. Your
argument is meaningless.

>  I'm not going to upgrade every bit of hardware and
> software in the box just to prove it _isn't_ FreeBSD, when FreeBSD
> represents the only change to a machine that has run without fail for
> eight years.  I also don't believe in throwing darts to solve problems.

It's not throwing darts, it's sensible advice. NT is ancient, like
your firmware no doubt.

> --
> Anthony

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Location of openssl certs in FreeBSD 5.3-RELEASE

[Freminlins]

On Fri, 11 Mar 2005 22:29:24 -0500, Madhusudan Singh
<[EMAIL PROTECTED]> wrote:
>
>  Anyways, where are the certs installed in FreeBSD ?

There are no default certificates in FreeBSD. They are easy to create
however. Look in the Handbook here:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/openssl.html.
If you want to create your own CA you can search for a "howto" on
Google. There are enough of these published so you won't have a
problem finding one.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: What's the easiest way to do a backup and verify?

[Freminlins]

On Mon, 7 Mar 2005 14:56:44 +0100, Anthony Atkielski
<[EMAIL PROTECTED]> wrote:
> Is there an easy way to combine a backup and verify when doing backups
> with dump?

With the FreeBSD version of dump, no. You can use the L flag to ensure
a consistent dump, though this is not the same thing as verifying a
dump.
 
> On Windows NT it's just a matter of checking a box.  I seem to recall
> the last time I looked into this on UNIX there was no easy way to
> accomplish a verify operation for a backup, but perhaps things have
> changed with FreeBSD 5.3 (?).

Well on Solaris it is possible using the v flag. 
 
> I've never had a problem with backup (I backup to DAT tape), but I'd
> feel better if every backup was followed by a verify to make sure the
> tape is readable.

I have always had mixed feelings about this. Although it is nice to
know the dump was correctly written, that is only valid at that time.
There is no guarantee that reading from the tape weeks later is
possible. Noone cares how good your backups are, only "I need my files
back NOW!!"

> Anthony

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: no free inodes

[Freminlins]

On Sat, 5 Mar 2005 14:22:50 +, db <[EMAIL PROTECTED]> wrote:

> /var: create/symlink failed, no inodes free
> foo.domain.topdomain - error opening scoreboard: No space left on device

This means what it says - you have run out of inodes. Do df -i /var
and you will see all your inodes have been used up.

The number of inodes is fixed at newfs time. This really leaves you
three options:
 1.You can either find some files on /var which are no longer wanted
and delete them, thus freeing up inodes.
 2. You can dump the file system, and recreate it using a
different-than-the-default inode density. man newfs and look at -i for
full information.
 3.You evidently have lots on space on /home so move a lot of files
(preferably a single directory containing lots of file) to /home from
/var and symlinking it.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: ftp monitoring and limiting transfer rates

[Freminlins]

On Fri, 04 Mar 2005 18:26:26 +0100, Frank Staals <[EMAIL PROTECTED]> wrote:
> Hey everyone, I have a question:
> 
> I am running FreeBSD 5.3 with the default ftp-server ( ftpd ) as ftp
> deamon, everything works fine but I'd like to see what files peolpe are
> trying to download and when people log in. I would like I can get a html
> file which I can access on my httpd server. I know I can see who is
> loggin in by : tail -f /var/log/xferlog but that doesn't show what files
> are being transfered.

Have a look at the man page for ftpd, specifcally -l. The answer is there.

> My second question is about bandwith managing. How can I limit the
> ftp-transfer speeds that are used for downloading when people connect
> over rl1, my NIC which is connected to my modem and the internet. I am
> running pf for my firewalling and allready tried this in /etc/pf.conf:

I haven't used pf, but using ipfw this is simple enough. Check out the
man page for ipfw, the dummynet section.
 
> Frank Staals


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: NFS Write performance

[Freminlins]

On Wed, 02 Mar 2005 08:52:16 -0800, Tim Traver <[EMAIL PROTECTED]> wrote:
> Hi all,
> 
> ok, I've searched far and wide, but I have to ask the FreeBSD gurus
> about it...
> 
> I'm using a Netapp NFS server to serve up content to FreeBSD clients,
> and I am seeing terrible write performances.

I don't call myself a guru, but we have a similar setup with 4 NetApps
with about 20 FreeBSD (dual Xeon) clients.

> I've turned on these in the rc.conf file :
> 
> nfs_client_enable="YES"
> nfs_client_flags="-n 4"
> nfs_bufpackets=8

You don't need these for the client:
> nfs_server_enable="YES"
> rpc_lockd_enable="YES"
> rpc_statd_enable="YES"

> and I've got these in the sysctl.cnf file :
> 
> kern.maxfiles=32768
> net.inet.tcp.keepidle=3600
> net.inet.tcp.sendspace=65536
> net.inet.tcp.recvspace=65536
> net.inet.tcp.slowstart_flightsize=2
> kern.ipc.somaxconn=16384
> kern.ipc.shmall=65536
> kern.ipc.shmmax=268435456
> kern.ipc.nmbclusters=32768
> 
> I'm using 5.3-RELEASE on a dual AMD Opteron machine.
 
> I guess my question is, how do I make NFS writes fly ???
> 
> The reads seem to be pretty good. I know that the settings on the netapp
> are per their settings...

What sort of performance are you getting? On one of our already busy
mail servers, using just normal mounting on fast ethernet, I get:
 
bash-2.05b# time dd if=/dev/zero of=/mail/A-D/100M bs=1024k count=100
100+0 records in
100+0 records out
104857600 bytes transferred in 9.284085 secs (11294339 bytes/sec)

Not bad! We also get extremely good I/O on email, i.e. small files.

> Thanks,
> 
> Tim.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: kernfs

[Freminlins]

kernfs was removed some time ago (about 4.8 I think). It certainly
exists on a 4.7 machine I have.

Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Is Yahoo! moving from FreeBSD?

[Freminlins]

On Wed, 23 Feb 2005 16:36:36 +0100, Jorn Argelo <[EMAIL PROTECTED]> wrote:

> I don't think that they would. That'll be a massive migration involving lots
> and lots of costs. They have to pay for RedHat Enterprise too. The only reason
> I can think off is that they want support.Perhaps I missed a part, but I don't
> see the word FreeBSD in that article.

Although it doesn't state FreeBSD, I understand that Yahoo! runs stuff
on FreeBSD.
 
> Besides, the point of the article is not regarding a migration of Yahoo, but
> Linux and IT in general. It has nothing to do with Yahoo or FreeBSD. I think
> that the author of the article is simply mistaking.

I'n not sure I agree with that. The author stated "But in December,
Yahoo started to port ... to Red Hat Enterprise Linux 4.0" That would
suggest that Yahoo! is moving to Linux.

I am very interested in this as I have for several years used the
argument "we use the same OS as Yahoo!". We're not going to migrate to
Linux if Yahoo! does.

> Jorn.


Frem.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Is Yahoo! moving from FreeBSD?

[Freminlins]

"But in December, Yahoo started to port its homegrown infrastructure
applications from its custom operating system to Red Hat Enterprise
Linux 4.0, which was in beta at the time and was released last week.
Plans call for a gradual migration of more applications to Linux, but
the timing and number will depend on how successfully the early work
goes, Ng said."

http://www.computerworld.com/softwaretopics/os/linux/story/0,10801,99901,00.html
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"